completion_female.csv

completion_male.csv

employment_over_15.csv

income_per_person.csv

life_expectancy.csv

PCA demonstration.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PCA demonstration"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Outline\n",
    "- Intro and motivation for PCA\n",
    "- Dataset used. Reading and preprocessing the data\n",
    "- Exploratory visualizations and rescaling\n",
    "- Applying PCA. Using biplots to explore the transformed features. \n",
    "- An interactive visualization to compbine original features and principal components"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## When can we use PCA? \n",
    "\n",
    "- We can discover latent features (that are linear combinations of the original features) by extracting the features with maximum variance.\n",
    "- PCA is often used as a preprocessing step (dimensionality reduction) before supervised learning. It can drastically reduce the size of the feature space. This can be very useful especially when we have a lot of features but our data is not enough to \"fill\" the feature space.\n",
    "- One potential negative of PCA is that the resultant components may be harder to interpret than the original features when models are fit on them. But on high-dimensionality datasets, this can actually provide meaning to data that may not have an obvious prior grouping.\n",
    "- A reduced feature space can make it easier to visualize data that has a high dimensionality. In the example below we use PCA as an exploratory technique.\n",
    "\n",
    "We will use data from Gapminder that contains socio-economic and educational info about many countries around the world. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from sklearn import decomposition, preprocessing\n",
    "\n",
    "# plotting settings\n",
    "%matplotlib inline\n",
    "sns.set_context('notebook', font_scale=1.3)\n",
    "\n",
    "# for high resolution displays\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "\n",
    "# for interactive plotting with hover\n",
    "import mpld3\n",
    "from mpld3 import plugins\n",
    "\n",
    "# helper functions for the demonstration\n",
    "import utilities as utils\n",
    "\n",
    "%load_ext autoreload\n",
    "%load_ext watermark\n",
    "\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Package versions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2016-09-19 \n",
      "\n",
      "CPython 3.5.1\n",
      "IPython 4.2.0\n",
      "\n",
      "numpy 1.11.0\n",
      "matplotlib 1.5.1\n",
      "pandas 0.18.1\n",
      "seaborn 0.7.1\n",
      "mpld3 0.2\n",
      "scikit-learn 0.17.1\n"
     ]
    }
   ],
   "source": [
    "%watermark -d -v -p numpy,matplotlib,pandas,seaborn,mpld3,scikit-learn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Read and preprocess data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We combine several features:\n",
    "- primary school completion rates for *male* and *female* students. It is calculated by taking the total number of female/male students in the last grade of primary school, minus the number of repeaters in that grade, divided by the total number of male/female children of official graduation age. \n",
    "- income per person - Gross Domestic Product per capita (fixed 2011 international dollars).\n",
    "- employment over the age of 15 (as a percentage of the total population for that year). \n",
    "- life expectancy at birth\n",
    "\n",
    "All of these are taken for the year 2006. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/nikolay_shenkov/anaconda/envs/data_analysis/lib/python3.5/site-packages/IPython/core/formatters.py:92: DeprecationWarning: DisplayFormatter._ipython_display_formatter_default is deprecated: use @default decorator instead.\n",
      "  def _ipython_display_formatter_default(self):\n",
      "/Users/nikolay_shenkov/anaconda/envs/data_analysis/lib/python3.5/site-packages/IPython/core/formatters.py:669: DeprecationWarning: PlainTextFormatter._singleton_printers_default is deprecated: use @default decorator instead.\n",
      "  def _singleton_printers_default(self):\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>completion_male</th>\n",
       "      <th>completion_female</th>\n",
       "      <th>income_per_person</th>\n",
       "      <th>employment</th>\n",
       "      <th>life_expectancy</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Afghanistan</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1193.282161</td>\n",
       "      <td>56.000000</td>\n",
       "      <td>53.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Albania</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8090.423280</td>\n",
       "      <td>51.400002</td>\n",
       "      <td>74.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Algeria</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12088.087622</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>74.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Angola</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5303.703307</td>\n",
       "      <td>75.500000</td>\n",
       "      <td>56.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Argentina</th>\n",
       "      <td>97.91143</td>\n",
       "      <td>102.10666</td>\n",
       "      <td>12912.711243</td>\n",
       "      <td>58.400002</td>\n",
       "      <td>75.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             completion_male  completion_female  income_per_person  \\\n",
       "country                                                              \n",
       "Afghanistan              NaN                NaN        1193.282161   \n",
       "Albania                  NaN                NaN        8090.423280   \n",
       "Algeria                  NaN                NaN       12088.087622   \n",
       "Angola                   NaN                NaN        5303.703307   \n",
       "Argentina           97.91143          102.10666       12912.711243   \n",
       "\n",
       "             employment  life_expectancy  \n",
       "country                                   \n",
       "Afghanistan   56.000000             53.2  \n",
       "Albania       51.400002             74.5  \n",
       "Algeria       50.000000             74.8  \n",
       "Angola        75.500000             56.9  \n",
       "Argentina     58.400002             75.5  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from os.path import join as pjoin\n",
    "\n",
    "# data obtained from www.gapminder.org/data/\n",
    "filenames = [\"completion_male.csv\", \"completion_female.csv\", \"income_per_person.csv\",\n",
    "             \"employment_over_15.csv\", \"life_expectancy.csv\"]\n",
    "\n",
    "gapminder_data = []\n",
    "for name in filenames: \n",
    "    table = pd.read_csv(pjoin('gapminder_data', name), index_col=\"country\")\n",
    "    gapminder_data.append(table)\n",
    "    \n",
    "# create a dataframe with multiple features per country\n",
    "df = pd.concat(gapminder_data, join=\"inner\", axis=1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(169, 5)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will keep only countries for which we have a complete set of features (no missing data in any of the columns). Note that this **introduces a bias** - the missing data is not 'missing at random'. Poorer countries tend to have more missing data. So we are likely removing countries that have problems with their completion rates."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df.dropna(axis=0, how=\"any\", inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Income per person transformation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before we apply PCA, it is a good idea to check the distributions of the original features. Features with extreme values, or with highly skewed distributions might inversely affect PCA performance.\n",
    "\n",
    "In our case income per person is highly skewed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x10be89240>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA/AAAALvCAYAAADCqC70AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAWJQAAFiUBSVIk8AAAIABJREFUeJzs3Xm8bfd8N/DPzSAhElJiSo3RfInUTAxtUaQS1FBDhZJK\n0TyieB7lMT20qVZLxBNj1RCVEk9RNYUYgpo7EGnwM7dBEENkkvk8f/zWzt05Ofucc+/Z996se97v\n1+u89rSG39pr7X32Z/1+6/fbsLCwEAAAAODKbYdtXQAAAABgZQI8AAAAjIAADwAAACMgwAMAAMAI\nCPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAA\nAAAwAjtt6wIALKWqbpzkO8PD77bWbrYtywNsH6rqWUn+anh4z9baJ6dee1ySNw0Pn9Fae9kWWP8d\nk9y4tfbONSzjBUleMDx8WGvtXTNe+43W2mc2u7BzUlW/luQ3W2tvXOK17ya5UZJzWmt7bOWiAYyO\nGnjgym5hWxcA2C4t990y9++dqtqtql6V5HNJbjWnxW7VbdhUVbVTVf1Zki8n+Y0Zk23zcgKMiRp4\nAICNtlSgvGOSw+e4/IU5LmtL2TvJ87NyOcewLQBXCgI8AECS1tqbk7x5W5djJa21P0vyZ9u6HPPQ\nWrvpti4DwJhoQg8AAAAjIMADAGx5G7Z1AbaB9bjNAFuUJvTAKC3qLXr/JF9N8vgkj07vIGqPJKcn\n+UiSl7fWvrLC8nZI8oAkhw7Lu2GS85N8Lcm7k7ymtXbWMvPfJP361vsmuVmSXZL8OL3Dqre11t69\nzLyXDndf3Fp7TlXdNclTk9wtyV5JfpTkE0n+erIdQ3mfmOSxSW6Z5CpJvp7kbUmObq1dtML2PjzJ\n7ye587COc4b535vkVctt60qmtueFrbU/r6oHJPmTJLdN3y/fS/LhJC9rrX1jFcurJP8jyX2S/GqS\nHZP8IMnH0/fLF2fMd48kJw0PH5A+qsExSe6S5OIk30zyouX2zdSypkdFOLS19vdV9QdJnpRkvyRX\nHV5/f5KjWms/XMUy75TkCUnumeQGSS5N8t/px+wrWmvfmjHf4mP/V5L8Tfr7e176MfvMSe/jVbV7\n+rHywGH63ZP8Ism30/fDa1tr31+hrHul74P7Jdk3yW5JfpLkP5K8I8lxrbVLZ8z73fRext/RWnvE\n0CP5U5McmH6N9PlJTk3y1iSvb61dvFxZVlJVuyV5SpKHJblF+vvakhyb5DUrzLtiL/RVdfckhyX5\nzaH8l6Z/1j8/bOM7F00/fRwmPdS+sKpeODy+Z2vtk4uOsSPSv3dekX7cb0jy3SSvbK29brle6Jco\n7zWT/O8kD07fD2cl+VKS45L8Q2vtCteeLyrL+1prv7vM8j+e5LeGh9durf1s0fyTbT60qg4dHh/a\nWvv7Yf7vZoVe6Ifvu8l31p2SXDvJuenH8IfS35clP3NT79VC+rF/cZInJ3lEkl9L/+yeluSE9O+k\n02ZtK8CVgRp4YOwW0kPhx5P8XZJ7pP+4u0qSGyf5oyQnT/1wvIKqukGST6f/YH5wkn2G+fdIckD6\nkFOnVtVtZsz/rPSA8KdJbpP+I/Eq6WHz4UneVVWfGELQctuRqnpekn8Z5tt7WM4Nk/xBks9V1d2r\n6hrD9r56KN8eSXYd1v3iJCcMP3iXKut1qurTSd6e5CHpwXHnJHsOy3pRkm9V1X2WKetqTLbnRUne\nk+TeSa41rOum6cH3lOFEwkxDyDklPZDdIsnV039w75O+b/+tqo6etb1TZdk3fR//dpKrpb9nt08P\nXpu6XRuq6i3p10rfNck10vdTJflfSb5aVb81awFDz9x/mx74/mjYlqumh+Jbpp/s+EpVPXMVZfmN\n9MB/QPpJo8l+/O9hXfsn+UqSl6QHzj3TT95fKz0IPTfJN1f4fDw2ybfSQ9Cdk1wzfT9eP/3EyLFJ\nvlRV+yxTzsnx8MgkJ6efDNgn/bi95rAdr04/xjd7KLHh5MApSf4yff9O3tfbp4fhjw+PV7Jkh2pV\n9fL0z+eh2Xii7qrp3zWPTPKPVfWZITQvXt7CEo+XWs9C+km1T6d/Rq8+lPlW6SfaViznlFuknxx5\nZnpY3WVY9oFJ/j7JZ6vq2svMv5qO5RYW3c6af9Y2L9uB3bBP/zX95OSD0o+7ndOPm9sneU76Mfz4\nVZT1V9NPOh2VfizvmX4M7pt+UulrVXXfVSwHYJtRAw9sD96Y/kP1O0nekOQbSa6bXkt26/Qa29dU\n1Umttf+annEIw59PD8sLwzKOTa+Nvm6SQ9J/6O2d5INVdevW2hlT878wyf8Z5r00yT+m12qek14z\n+/hh3t9M8qmqutMytdsPTw815w7b9Pn0H9tPGJa1W/pJiu+lB57PJ/mH9BB6h/Rau12T3Cu9xvW1\nS2zrZ9KDx0KSLyY5Pr1m75rptasPTg9376+q32mtfXxGWVfj97OxBvRtST6Y/sP7QUl+Nz30vrWq\nzm+tvXfxzFV1VJKnD2X9eXpg/rfh8W2T/OFQ1qemh5wnLFOWv0wPL8el19jdIL3mc3PGyH7WsF0X\nph9v/5J+QuBR6TWR10g/ifJbrbV/X2L+45M8dNiO09OPt1PS35sD0sPh1ZK8uKp2bq29aJmyvDz9\nZPyr0vftryW5WWvte1V1lfSTUpPa/bcn+Vj6e3nd9P39gPT35fVV9e+ttVOmFz4E+zdmY8j6YHor\njZ+mH6uPSw8/+yf5zHB8//eMst4ufd/vMLwHJya5IL2lyWFDOW6XfrLhScts85Kq6jpJPpV+Am8h\nPai9Ob0Fy/7DMn8jvTf4TVZVh6WfXFlI/8y8Kf17Ysf04+EJ6e/rAUlen94CIEn+Mz2I75/kL4b5\n357+HkxeX+xZ6e/H+5L8v/TP50OTzKxpn+HV6cfV19KP1f9OP9H0pPTj4s5JPlRVd26tXbKJy17O\nj9O/S66T5HXp23xSeguYpO+baUs2ta+qvdOP2cn383fTPy9fS/+c3T/9u+Rq6cfwrq21Vy9Trnen\nb//JSd6S/n7cJP39uFn69+c/VNU+rbWzV7uxAFuTAA9sDyrJPyf5/dbaBZc9WfXa9LB2r/Sw+IdJ\nXrho3pdk44/DdyZ5TGvtwqnXX1FVbxjmvU56jeXThuXfKRvD+zlJfre19onLFazqJelNjH8nyc3T\nm/A+esZ27JPeLPme003+q+qN6c29r50eFCrJ61prfzw17/+rqk8N78NCem3g5QJ8+o/5SXh/YWvt\nyEWvv36oeX93eq3i5Ifs+TPKu5JJyH1Ya+19U88fO9SWvT49zB1TVScu2ncHZmN4/0KSB7bWfjK1\njLdV1YuTfCA9MD2+qt7XWvvnGWXZJb1p+3St9kvXsF1nJvmd1tq/Tj3/uqkTOrsmeWV6Df1lquoJ\n2Rje35/kUa21c6cmOa6qXpp+Eujm6U2tT2itLQ48SQ89uyR5yozQclA27u/nttb+etHrr6mq/5X+\nGdiQ3srhiVNl3Tv9eF1Ib3b82Nba2xdtz0vTA9qh6cfnWzN7vO990pvLLz4x9Laq+uf0QJ8kj6mq\np27GcfdX6Se8FpL83aLPxz9W1SvSw+DmjsF+xHD7gyR3bK39fPrFqjomvab4JkkeUlV7t9a+31r7\naZL3VNUvpib/WmvtPcusa5f05viPnHrulZtR5p3Tv38eM31ZzfBenJjeCuO26d9pR23G8pfUWvtl\n+jbfeOrp/15hm5fyuiz//fz64RKdf0z/jn9ZVX18mUumKv178YjpSweGFjGfTT9Req0kv5d+ogDg\nSkcTemDsNqRf93vodABMkuFa2r+Yeuou069X1Z7pwXwhvSbmsYt+HE48OT2wJb2WdeL5U/efsji8\nD2U4N/1ay+8NZZ1cA7yUSbC+3I/PoSbo+GyspTojPWwtXtd702u+NqTX9l2mqm6WHuoXkrx7ifA+\nWcZH0murNyS5Xnow21wLSV6wKLxP1vPG9JrdDenXvz5q0STPHW7PS/LgReF9soyfp58Mmez3Z88o\nx4b0Gui/2tQNmGEhyeGLwvukTC9MD4kbkty5qu45ea2qNqRfi5z0EPiIReF9soz/Sj8uk/5/ermm\n9D9J8rczXqup+1fYB4OXp1/+8fn0WvVpk1rgpPcV8PZFr08+Y09Ir1HdkOSuVXXvJdazIf19e9lS\nrTqG4+5fhoe7pofKVauqa6W3BlhIb4FzxOJphmPoEUk2t6a5huV/bnF4H5b/0yQvS38/T0g/mbA5\nJp/z5VperNZ3kjxucZ8YrbVfpL8Xk8/OFd6vba2qbpd+EmohvaXD4vCeJBm+X56d/r7tnOW/B05L\n/66+XJP94Tt2+gTG5f5XAFyZCPDA2C0k+ejwg3QpJ0/dv9ai1w5Kb/6a9M6zlqzxG55/SvqP3EdV\n1Q5D8+T7DOv/QXpzzCUNPw5fMTzcIb0Z8WKTH+1XCEmDbw+3C0k+uEwndZPmy4uvwX14Nn7nv35W\nWQevm7r/wBWmXc452bjdSzl6qfVU1XXTLzlYSPKB1tqPZi2gtfbtJB9Nf//uNKOfgYUkX1kqdG2m\n7ywVZqcsuV3pzZVvOpTnrcvVMLfWPp1+7fqGJAcN4X+xSZicFUinT3o8o6p2XmI9l7TWbtlau1tr\nbXHwuf9we0GS/7tMWS9JMl27/5BZ06bXBs+y3Gd1Jb+Tjcf3G2d1hNda+1p6q5zN8ZP0/XHfqlqy\nGX5r7ZXD+/mA1tqXNnM9SXJWa+3La5h/4phlvtf+K71/ig1JblRVt5/D+ubpAVP3j55xcnXiNUl+\nlr4tD1zm8/LuWZ0tZm3HH8BWowk9sD1Yrof56WC/OMAcMHX/k8utoLX2D9OPhw7tdk3/UfjxpXpy\nXuTD2RhylqrdWUjyk6EWbylnTt3/5jLrOW+4XXyCdrop9z5VtdRJhGk/T+/g6U4rTLecT7fWzpv1\nYmvt1Kr6afqP5en3ZLqsO62irL+cun+n9Gb1i31tpcJughNXeP2T2dgp16zt2mMV2zU5Fq6e3rnd\nUsf5ctv1nvQm67um107fd2iq/qEkJy13je9wImRysuHflzlBNvHhqfvL1V5u7md1JdPrXKlfg09k\n48mJTfH29E4K90jvbO+k9GPtxNbaqZuxvFkmNc7z8LEVXp90mJn093CpSzW2lel9+tHlJmytXTBc\nQvS76Z2I3ipL9y2wpY4/gK1GgAe2B2fOeqG1dknVZS2JF4fa607d/69smul5lxzua5Hpaa47Y5qf\nrXLdM0NxZvfm/KtT94+ZMc1SfqWqNqziBMVS2iqm+e/0AD9dcz5d1odk+Rrdxa4z4/mZx8hmWHa7\nWmtnV9WZ6a0grjf10vR2PSmb1lHbdbJ0+Fju2P9JVf1hekduO6d3Wnb48HdxVX02Qydpizt3zCYe\n3621M6tqctJn1vF9wQq1qNO15pvaQnB6nbM60ZtY7gTYcl6QfhLmruk1vfce/lJVp6c3m/+nJB+a\n1QJgE8zreP32Cq9P7/frz2md8zK9T1fajuSK37FLBfjl3te1HH8AW40vKGB7sLk/ln9l6v4vZ061\ntOmhrq5wHfMSpkP3rGGslh27fY32yOWHe1rtX9JrgDfHasaSn7wvO1Yfv3tS1olNLevuM9ZzwYzn\nN8dqt2tDLr8tm7tdk/Grl7Lsdg1N/W+XPlrBOVPL2zH9MoW/TvLtqnrLMErBUmVdzfGdbNyXs47v\ntYba5UxfMrLSZ3mzehcfWpP8VnpP9Kfk8vvn+ukjTrw3/f2cOW76Ks3leF2uBcxget9edR7rnKPJ\nMXj+Kk8gruY7dksegwBbhRp4YD2b/sF3tU2cd3o85tWMKz0dglcbiOZpEigXkuy6zDX087TrKqaZ\nvHfnT3XoNr1fDm+tvS5XLqvdroX0Dgcnprfrfq21D2craK19Ncljh34b7pXkvuk1x7+ejX0vPDrJ\nDZPcc3i8qcd3svEY3xbH93TrlZU+y7us8PpMw/XTr0ryqqq6aXo/GvdJf98mJ0B+Nck7qurgoXO+\nbWYYVm253vynv5c2t9Z/NZ+HzTE5BnddZSugbf0dC7BVqIEH1rPpztFutNyEVXXDqrpbVU2aRP9w\n6uV9VrGu6Z7nv7/K8s3T9Lb+6syp5uuGq5hmcp316VPPTZd177mWaD6W3a6qumY21gj/YOqlbbEP\nLtNau7C19qHW2jNaa7dL345npw/1lyS/WVW/PdzfpOO7qq6dHmAXsm2O7+nj52YrTDuXY6q19p3W\n2qtbaw9NH0Lvvkk+Pby8Y/pwgtvaSsfZ9L6dvvRguqO3HbO8xR1mzsv0MbjSPk22/XcswFYhwAPr\n2b9N3b/rzKm6Jyb5VJIfVNVvJPlSNjZzveeMXo+nTQ+ttZbeqTfX56fu32u5Cavq6lX1zqo6pqr+\nxxrWebcV1nObbKy1nO5sa9VlHZbz3Kp6Y1X9n6pazUmDtZo1zvnEPabur2W7Xl5Vr6uq5yxq3r4q\nVfWAqnp2Vc0aMvD01trf5PLD691meO3H6ddHb0hyh1Ws/z5T97fF8f2pqfv3XGHau2/qwqvqRlX1\nhKo6qqruvPj11tqlrbWPJblfeu3vhgzv5TZ2wAqv//bU/c9N3Z9uwj/r8o1U1VWzunC9OaY/L/eZ\nOVUvxy7Z+H1zbvpQggDbJQEeWM9OyMYxoQ+rqiUvK6qqHdPHUE96T8VfGDrj+mD6D/UbJHnsrJVU\n1dWTTAfh96yx3Jvj3VP3/+dSQ4pNeXJ6x3FHZBVBcxk3XKGn9f85df+dkzvD0HCnpL+3dx9OmCyp\nqm6U5Pnp49U/J5dv+r0lbEhyt6pabpzyp03df9fU/U+nN6nfkORhVTUz+FTVXdKvtf6j4XY1190v\n9r/TxxJ/dlUt10HZdPPz6fdvcszskstv0+Ky7pDeO/vEP29iOefhhGy8ROGw4TN3BUMLmoduxvL3\nTfK36e/D42dNNFxzPmmyvvhYnK7V3lq/v54864Wq2j/9hMNCki+31qZ7vv9pNn437jfruzHJ72X5\nHtvXss3T31lPHS4BmeWP01sCTIaenDW0IsDoCfDAutVa+1760FAbktw8yd8OYeQyw+OXDq8vpI8X\nP2ly/NfDcxuSHFNV91y8jqq62rCOGw7Tvr21ttxQRltEa+0/0ocP25A+JNk/VNUVrl2tqvsm+bPh\n4UKSl6xhtRuSvLaq9ltiPYcn+YPh4Wdaa4vH5v7LqWUcX1W3XmIZ10z/kX+VoaxvmONY77MspP/v\nfFtV3WCJMv1Veg38QpLjh3HHk/ShrpK8bHi4S5L3LNViYHju+Kn1HbWZowBMxqrfkOTNQy3l4nVd\nMxtD3kIuP5ziy9PD6IYkz6mq319i/p2SvC7JHYanPttaO2EzyromrbVfJjkqvazXTz++L7e9QyeJ\nx2fzOmX8RPolEBuSPL6qfmepiarqj9NHVVhIctKil6f7QNhzM8qwOQ6oqqMWP1lVeyf5x2xsHv+i\n6deHAPzF9O3dM8kzlljG7dO/G5c7Njd7m1trJ2fjd1YlOW7GMXxwNn5fXDh1H2C7pBM7YL17anrP\n0nsn+cMkd66qN6dfD3rjJI9KMqlt/WY2htu01j5XVX+R5HnpzUw/UlXvTP/ReU56UD4s/TrUhSTf\nSR/Ca1s5LMm/pw9J9rD0bX1Tkq+m/7i+1/D8DtkYHL+whvVdOqzrX6vq79Kb6O42rGMSgM7OEsOp\ntdbeXlX3T+9c7QbDMt6a5OPpPUnfKr12+trDLN9Kv557a7g0vUb2lKp6bZKTh3I8OhsvxfhhLt/C\nYOIl6dt+jyT7JflKVR2b3lx4hyS3T99Pk07wPpfk6M0s59+lt6L4tfQmyN8Y9vc3h3XtO6zrOsO6\njm2tXTZEXmvtv6rqiGE5Oyd5a1U9Nr2G/WfpTacPTQ9XSa+1PWQzyzoPL0ry4CT7J3lg+v75uyTf\nTT8B96T0vi5+lNlD3S2ptXZRVT0v/b3YKckHqupd6Sc8fpQ+XOB9s3F8+XOzKBQn+d7U/cdV1anp\nHcd9trV22qaUZxOckeTpVXW3JMcl+UmSW6fXWO+Zvt+Pa629Y4l5X5fkjukB+i+HwH7CMM89kvx+\n+smzL2RGU/3W2k+r6pfpPdzfr6r+V/qlGacOnSuu5PHpY9NPvrPuOBzDLf079+D0fT7poPOZrbUv\nr2K5AKM1lwBfVbsn+dP0Jpf7pA/hcmqSY5O8aamag+GH2XtXWnZrTSsBWN9WurZ8TcsZfmD+Rnqz\n9l9PD1V/MzXJZJiok5M8uLV2zqL5X1BV5yb58/SQ8/Dhb/H8H07ymNbaLzalfPPUWvtBVd01vVn3\nbdJbBbxg0WQL6QH1pa21/73GVX46yWnpJ0H+ZPibXs/3kjxgmRYJj0sPhU9J/3/1uOFvehkL6Scl\nHtJa25xm5pvjHemB/V654kmDhfT/fwe31n60eMbW2qVDjeEb0y/LuFp6DfiTFy1jIcmJSR65uWOK\nt9bOH/7XfiD9f/Pe6ZcbLC7vQnrN9B8vsYw3VtWFSV6dflLhfsPf4vn/I8nDW2srjcG+xbTWLqyq\ne6T/trhb+jb/9dQkk8/xS5O8ZTOW/4bhko3npn9ef2/4m7aQHugfvTigttZ+VFUnpV+jv0eS1w4v\nHZ7ePH9L+L30Vip3zuVD9mS/vT6zTyq+If2E1KHD44cNfxMXJXl6+vfectfaHz8sY6dsbNHz11nF\nCbfW2unDd9Y/pZ94uHGmTqJObcs5SZ7UWnvbSssEGLs1h+Phn9mX0mugJs0kd0vvJOb1ST4xNCFd\n7HbD7XnpNRVL/Z2+xHzA+rF4jO+lXl/zcobQcbv0GvgPpn//XJhey/iJ9GBzwKxwMnQEdov0Jrwn\np9eqnZdeS3R8kgNba/drrf1kc8q3aLrVWG5bv9tau316Tek/pQfs84fyfi29hvGOcwjvSXJJa+0x\n6TXT/5Lef8B56WHvOUlu1Vo7ZdbMrbWF1trT0082vDLJfw7LuDB9H30gvRn+Aa21Wb1Ob07T85Wc\nl17bekT6yYNzhr/PDM/dYbg8Y0mttfNba4ekj8P+xiRfH+a/IH1/vCvJg1prBy1zUmJV29Va+2b6\niakj0k8inT6s55z0jr7elOS3W2uPnnWioLV2XHoYPjK9tvVn6cfMt9JPfD00fR98d5mirOb43pTp\nltRaOzO9dviJ6cfcWen769T0XuHvkv753KzvldbaC9Jrpf82vZ+Gs9OD7A/Ta+OfmeQWrbXFzecn\nHjLM+/304/jMXL5p+cKi2+WstA0LSf41/fPzmvSa7wvSj4F3pu/3Jw1D413B8Pk7LL1VwT8P23jB\nsJw3JblTa+2YVZTl8PTA/t1h/rNzxY7xlv3OSm+Z8gfpJ2e+Pyznh0k+m97Xwz4rhPdNOf6mbwGu\ndDYsLGz+d9TQ6/IX0q99+356c8YT008MPCS9+dU1kryltXboonnfMUxzZGvthZtdCACuVKrq0vQf\nwJ9orf32StOPQVXdOP0SiIUkb26tzezIDABgS1lrE/r7p4f3hSSPaK19dnj+kiTvGGrej01ySFU9\na1Fzwsk1pdtiuBkAAAAYlbU2ob9Penj/0lR4nzYZKmnHbAzsk2vmbzo8FOABAABgBWsK8K21p6X3\nrvyoGZNM1/BfOHX/NukdwJy1wjVzAAAAQObQC31r7fTM7mxuMjTQL9KvlZ+4rPl8VR2YPozNbYfy\nfD3J29Kvm9eJCMA4bfFe9QEA1pu5jwM/XPd+y/QhcR6X3sT+Ga21c6cmmwT4A9J7fJ4O6jdJHyP3\n0Kp68FYcFgiA+dhee3LeXrcLABiJNfVCv1hVHZA+pMfEBUkOba29fdF0/5Y+JMhC+tAmr0wfjuZ6\n6WOFPi/95MIJrbUHzK2AAAAAMFLzroG/SfrYsBenjwV/lSTHVNXurbXXT0138jDNW1prr5p6/rQk\nR1bVt5Icl+SgqjqwtXbinMsJAAAAozLvGvirttZ+Ody/eZIXJ3loek3741prx23Csk5Osn+SN7XW\n/mhuhQQAAIARWuswcpczCe/D/W+21h6W5N3pnRn9xSYu7pPDfPvNr4QAAAAwTnPvxG4JL0/y4CQ3\nrKrrD73Wr8YvhturzqEMOhwCAADYfq2LEXDWFOCrat8kN0/yrdZamzHZdGC/dlVdNclB6R3WHdla\nu3Dp2XLd4faHaynjxBlnnD2PxTAye+21exL7fz2y79c3+399s//XL/t+fbP/16/Jvl8P1tqE/vgk\n70vyp8tMM2kCf2l6J3U3TvKKJM9J8ltLzVBVG5LcO73m/DNrLCMAAACM3loD/AnD7SOr6oaLX6yq\nq6QPCZckJ7XWzkzy6SRnDs89c8Zyj0jv0f6iJMeusYwAAAAwemsN8Ecn+Wn6kHEfqar7VNUOSVJV\nt0/ykSR3SPLLJM9IkqHJ/JHp1yjcp6qOq6q9h3l2r6rnDstdSG9if9oaywgAAACjt6YA31r7SZL7\nJ/lx+rXwJyY5r6rOSvJvSe6eXtv+kNbal6fmOzrJq9ND+iFJTquqM5P8LBvD/Stbay9aS/kAAABg\ne7HmYeRaa19IH6/9xUlOTXLJsNz/TPI3SfZrrZ24xHxHJDk4yXvSTwDsOty+K8mBrbWnrrVsAAAA\nsL2YyzByQ038c4e/TZnvQ0k+NI8yAAAAwPZszTXwAAAAwJYnwAMAAMAICPAAAAAwAgI8AAAAjIAA\nDwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAA\nACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMg\nwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMA\nAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAI\nCPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjMBO27oAW9v555+fgx/0oFz9\nmtfb4utgZqPuAAAgAElEQVQ679yz8z+f/Ec5+KCDt/i6AAAA2L6tuwB/8cUXZZdr3DDXvt0hW3xd\nP//B1/KjH/14i68HAACA7Z8m9AAAADACAjwAAACMgAAPAAAAIyDAAwAAwAgI8AAAADACAjwAAACM\ngAAPAAAAIyDAAwAAwAgI8AAAADACAjwAAACMgAAPAAAAIyDAAwAAwAgI8AAAADACAjwAAACMgAAP\nAAAAIyDAAwAAwAgI8AAAADACAjwAAACMgAAPAAAAIyDAAwAAwAgI8AAAADACAjwAAACMwE7zWEhV\n7Z7kT5M8JMk+SX6Z5NQkxyZ5U2ttYcZ810vy/CQHJdk7yc+TfC7Jy1prn5xH2QAAAGB7sOYa+Kq6\nUZIvJXlekv2Gp3dLcvckr0/yiaq62hLz3WyY7/AkN05ybpJrJXlgkpOq6ulrLRsAAABsL9YU4Ktq\nQ5J3Jrlpku8nOTg9vO+W5JFJzkwP8q9eNN+OSd6fZK8kn09yy9bar6QH+GOSbEjykqq6+1rKBwAA\nANuLtdbA3z/JHZIsJHlEa+1DrbWF1tolrbV3JHlaehg/pKquOzXfY5JUkrOT3L+19vUkaa2d1Vp7\nepLjh7K9aI3lAwAAgO3CWgP8fdLD+5daa59d4vX3DLc7Jrnt1PN/PMz35tbaz5aY76+G298cmugD\nAADAuramAN9ae1qSX03yqBmTTHeSd2GSDNfD32l47qMzlntKkjOGh/dbSxkBAABge7DmXuhba6cn\nOX3Gy08abn+R5AvD/Uo/cbCQ5KvLLPobSa6d5FZrLSMAAACM3VyGkZs21LDfMsmTkzwuPag/o7V2\n7jDJDaYm/94yi/p++vXz1593GQEAAGBs5hrgq+qAJNPXwl+Q5NDW2tunnttjcqe19stlFjd5bY9l\npgEAAIB1Yc3jwC9ykyTnJzknveb9KkmOqao/mppmctLgohWWdcGi6QEAAGDdmnc4fk9r7WpJUlU3\nT/LiJA9N8rdVdX5r7bhsrFnfeYVl7TLcXjiPgu211+5Jkl13TXbYYcM8Frkqu119l8vWzbZjH6xf\n9v36Zv+vb/b/+mXfr2/2P9uzudbATzeJb619s7X2sCTvTr+W/S+Gl86eTFNVu2S2qw23Z82zjAAA\nADBGW6N5+suTPDjJDavq+klOm3pt7yTfnjHf3unN8H8wj0KccUY/b3DOOWfn0ksX5rHIVTn3nAsu\nWzdb3+QMrH2w/tj365v9v77Z/+uXfb++2f/r13pqdbGmAF9V+ya5eZJvtdbajMmmh5i7dvrQcRcN\n6943swP8zYfbr6yljAAAALA9WGsT+uOTvC/Jny4zzX7D7aVJTmutXZzk08Nz915qhqq6dZK9hoef\nWGMZAQAAYPTWGuBPGG4fWVU3XPxiVV0lyfOGhye11s4c7h+ffl38YVW11+L5pub5WGvtG2ssIwAA\nAIzeWgP80Ul+mmS3JB+pqvtU1Q5JUlW3T/KRJHdI73n+GVPzvSlJS3LNJB+uqtsM81yjql6R5GFJ\nLknyZ2ssHwAAAGwX1hTgW2s/SXL/JD9Ov2b9xCTnVdVZSf4tyd2TnJnkIa21L0/Nd1GShyc5I8mv\nJ/liVf08yU+SPDm987qntNY+tZbyAQAAwPZizcPItda+kGT/9DHfT02vOd8hyX8m+Zsk+7XWTlxi\nvv8c5ntFekd2u6YPGXdCkvu21l671rIBAADA9mIuw8gNNfHPHf42Zb4zkjxt+AMAAABmWHMNPAAA\nALDlCfAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAI\nCPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAA\nAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAw\nAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8\nAAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAA\njIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAA\nDwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAA\nACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACOw\n07wWVFV7JPmTJA9Ksm+SXZKcnuSkJEe11k5dYp77J3nvSsturTnRAAAAwLo2lwBfVfsmOTHJjZIs\nJDk/yUXD40OTHFJVh7XW/mHRrLcbbs9LctaMxS/Mo4wAAAAwZmsO8FW1Y5J3p4f1byd5YmvtY8Nr\nt0pyTJJ7JXlDVZ3aWvvS1Oy3TQ/oL22tvXCtZQEAAIDt1Tyapj8iyS2SXJzkIZPwniRDs/mDk3w1\nyc5Jnr1o3tsOt18KAAAAMNM8AvxB6bXoJ7XWTln8YmvtgiTHJdmQ5B6T56tq9yQ3HR4K8AAAALCM\neVwD/8Ukuyb5zDLTnD7c7jH13G3SQ/0vWmvfnUM5AAAAYLu15gDfWjs6ydErTHb34fZ7U89d1ny+\nqg5Mctjw3E5Jvp7kbUne0lrTiR0AAADr3hYfnq2qbprkkPRm9u+femkS4A9I8sEkD0ty8yQ3SXJg\nkmOTfHQYng4AAADWtS0a4Ktql/Sa9KumDy131NTLkwC/S5JXJ7lVelP8myR5YfowdPdI8tYtWUYA\nAAAYg7mMA7+Uqto5ybuS3Dm99v0prbXpJvQnp/dc/5bW2qumnj8tyZFV9a30zu8OqqoDW2snbqmy\nAgAAwJXdFgnwVbVb+tjw904P70e11t44PU1r7bDlltFae2tVPSvJ/ulD1a0pwO+11+5Jkl13TXbY\nYcNaFrVJdrv6Lpetm23HPli/7Pv1zf5f3+z/9cu+X9/sf7Znc29CX1XXT/Iv2RjeX9Zae+ZmLu6T\n6T3V7zen4gEAAMAozbUGvqr2S/KBJDdKD+/Pb6395RoW+Yvh9qprLdsZZ5ydJDnnnLNz6aVbr2P7\nc8+54LJ1s/VNzsDaB+uPfb++2f/rm/2/ftn365v9v36tp1YXcwvwVXWX9F7m90y/tv2JrbVjZ0x7\nsyQHJblekiNbaxfOWOx1h9sfzqucAAAAMEZzCfBVdev0mvdrJjkvySNaax9YZpYbJ3lFei39J5J8\nZIllbsjGZvifmUc5AQAAYKzWfA380GHdu9LD+7lJ7rdCeE+STyc5c7g/6/r4I9KHlLsofUx4AAAA\nWLfm0Ynd85LcLL2m/PDW2qdWmmFoMn9kegd196mq46pq7ySpqt2r6rlJjh6WeWRr7bQ5lBMAAABG\na01N6KvqKkkOn3rqJVX1khVmu2Nr7futtaOrap9h/kOSHFJVZyXZLcmO6eH9la21F62ljAAAALA9\nWOs18Psn2T09bCfJdVaYfiE9nCdJWmtHVNV700P8XdKb4f84yWeTvKa19tE1lg8AAAC2C2sK8K21\n/8hUIN/MZXwoyYfWsgwAAADY3s3jGngAAABgCxPgAQAAYAQEeAAAABgBAR4AAABGQIAHAACAERDg\nAQAAYAQEeAAAABgBAR4AAABGQIAHAACAERDgAQAAYAQEeAAAABgBAR4AAABGQIAHAACAERDgAQAA\nYAQEeAAAABgBAR4AAABGQIAHAACAERDgAQAAYAQEeAAAABgBAR4AAABGQIAHAACAERDgAQAAYAQE\neAAAABgBAR4AAABGQIAHAACAERDgAQAAYAQEeAAAABgBAR4AAABGQIAHAACAERDgAQAAYAQEeAAA\nABgBAR4AAABGQIAHAACAERDgAQAAYAQEeAAAABgBAR4AAABGQIAHAACAERDgAQAAYAQEeAAAABgB\nAR4AAABGQIAHAACAERDgAQAAYAQEeAAAABgBAR4AAABGYKdtXYDt3Tv+6Z/zvhM/vlXWdb299syr\n/u/RW2VdAAAAbF0C/BZ27vkX57p3fexWWdcPv/z3W2U9AAAAbH2a0AMAAMAICPAAAAAwAgI8AAAA\njIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAA\nDwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAA\nACMgwAMAAMAI7DSvBVXVHkn+JMmDkuybZJckpyc5KclRrbVTZ8x3vSTPT3JQkr2T/DzJ55K8rLX2\nyXmVDwAAAMZsLjXwVbVvki8n+fMkt08/MXBRkhslOTTJv1fVo5eY72ZJvpTk8CQ3TnJukmsleWCS\nk6rq6fMoHwAAAIzdmgN8Ve2Y5N3pYf3bSe7bWtuttbZ7klun18BfJckbquq2i+Z7f5K9knw+yS1b\na7+SHuCPSbIhyUuq6u5rLSMAAACM3Txq4B+R5BZJLk7ykNbaxyYvDM3mD07y1SQ7J3n21HyPSVJJ\nzk5y/9ba14d5zmqtPT3J8UP5XjSHMgIAAMCozSPAH5RkIclJrbVTFr/YWrsgyXHpNer3mHrpj4f5\n3txa+9kSy/2r4fY3q+pGcygnAAAAjNY8AvwXk7wzyQnLTHP6cLtHklTV1ZLcaXjuo0vNMJwMOGN4\neL+1FxMAAADGa8290LfWjk5y9AqTTa5j/95wW+knDxbSm9fP8o0k105yq7WUEQAAAMZui48DX1U3\nTXJIelh///D0DaYm+d4VZtro++lN76+/ZUoHAAAA47BFA3xV7ZLkbUmumuT8JEcNL+0xmaa19stl\nFjF5bY9lpgEAAIDt3hYL8FW1c5J3Jblzeu37U1prk9r2SdP9i1ZYzAWLpgcAAIB1aYsE+KraLckH\nsrGH+qNaa2+cmmRSs77zCovaZbi9cL4lBAAAgHGZe812VV0//Vr326aH95e11p65aLKzp6bfZRhq\nbilXG27PWmu59tpr9yTJrrsmO+ywYa2LW7Uddtzi3QxcZqedd7xsO7k878v6Zd+vb/b/+mb/r1/2\n/fpm/7M9m2u6rKr9knw2G8P781trf7rEpKdN3d97mUXuPSznB3MrJAAAAIzQ3Grgq+ou6TXveya5\nOMkTW2vHzpj86+nXv++UZN8k354x3c2H26+stXxnnNEr/c855+xceunCWhe3apdeculWW9fFF11y\n2XbSTc7Ael/WH/t+fbP/1zf7f/2y79c3+3/9Wk+tLuZSA19Vt06/5n3PJOclefAy4T2ttYuTfHp4\neO9llrnX8PAT8ygnAAAAjNWaA/zQYd27klwzyblJ7tda+8AqZj0+fYz3w6pqryVef95w+7HW2jfW\nWk4AAAAYs3nUwD8vyc3Sr1U/vLX2qVXO96YkLT34f7iqbpMkVXWNqnpFkocluSTJn82hjAAAADBq\na7oGvqqukuTwqadeUlUvWWG2O7bWvt9au6iqHp7ko0l+PckXq+oXSa6eZMdsHDt+tScEAAAAYLu1\n1hr4/ZPsnh62k+Q6K/ztlR7OkySttf8clvGK9I7sdk0fMu6EJPdtrb12jeUDAACA7cKaauBba/+R\nqUC+mcs4I8nThj8AAABgCXMdBx4AAADYMgR4AAAAGAEBHgAAAEZAgAcAAIAREOABAABgBAR4AAAA\nGAEBHgAAAEZAgAcAAIAREOABAABgBAR4AAAAGAEBHgAAAEZAgAcAAIAREOABAABgBAR4AAAAGAEB\nHgAAAEZAgAcAAIAREOABAABgBAR4AAAAGAEBHgAAAEZAgAcAAIAREOABAABgBAR4AAAAGAEBHgAA\nAEZAgAcAAIAREOABAABgBAR4AAAAGAEBHgAAAEZAgAcAAIAREOABAABgBAR4AAAAGAEBHgAAAEZA\ngAcAAIAREOABAABgBAR4AAAAGAEBHgAAAEZAgAcAAIAREOABAABgBAR4AAAAGAEBHgAAAEZAgAcA\nAIAREOABAABgBAR4AAAAGAEBHgAAAEZAgAcAAIAREOABAABgBAR4AAAAGAEBHgAAAEZAgAcAAIAR\nEOABAABgBAR4AAAAGAEBHgAAAEZAgAcAAIAREOABAABgBAR4AAAAGAEBHgAAAEZAgAcAAIAREOAB\nAABgBAR4AAAAGAEBHgAAAEZAgAcAAIAREOABAABgBAR4AAAAGAEBHgAAAEZAgAcAAIAREOABAABg\nBAR4AAAAGAEBHgAAAEZAgAcAAIAR2GlLLbiqfi3JyUk+3lo7eMY090/y3pWW1VpzogEAAIB1bYsE\n+Kq6epK3JdllhUlvN9yel+SsGdMszKtcAAAAMFZzD/BVtWeS9yS5fVYO37cdpnlpa+2F8y4LAAAA\nbC/m2jS9qu6a5ItJ7p7V1Zzfdrj90jzLAQAAANubudTAV9XuSV6d5JDhqa8nOT3JPVaY56bDQwEe\nAAAAljGvGvibJXl0eq37a5PcIcl3V5jnNkk2JDmrtbbStAAAALCuzesa+EvTr3t/QWvt5CSpqpXm\nuaz5fFUdmOSw4bmd0mvw35bkLa01ndgBAACw7s0lwLfWTkny4E2cbRLgD0jywVz+mvmbJPmdJIdW\n1YNba7N6qAcAAIB1YVuOrz4J8LukXz9/qyS7pof3Fya5KP0a+rdug7IBAADAlcoWGQd+lU5OcnF6\nM/lXTT1/WpIjq+pbSY5LclBVHdhaO3FbFBIAAACuDLZZgG+tHbbC62+tqmcl2T/JI5II8AAAAKxb\n27IGfjU+meTXk+y31gXttdfuSZJdd0122GHDWhe3ajvsuPWuUthp5x0v204uz/uyftn365v9v77Z\n/+uXfb++2f9sz7blNfCr8Yvh9qrbtBQAAACwjW2TGviqulmSg5JcL8mRrbULZ0x63eH2h2td5xln\nnJ0kOeecs3PppVtvZLpLL7l0q63r4osuuWw76SZnYL0v6499v77Z/+ub/b9+2ffrm/2/fq2nVhfb\nqgn9jZO8In3ouE8k+cjiCapqQ5J7D9N8ZquWDgAAAK5ktlUT+k8nOXO4/8wZ0xyRPqTcRUmO3fJF\nAgAAgCuvbRLghybzRybZkOQ+VXVcVe2dJFW1e1U9N8nR6bXvR7bWTtsW5QQAAIAri23WiV1r7egk\nr04P6YckOa2qzkzys2wM969srb1oW5URAAAAriy2dIBfGP6W1Fo7IsnBSd6T5MdJdh1u35XkwNba\nU7dw+QAAAGAUtlgndq21P0zyh6uY7kNJPrSlygEAAADbgyv7OPAAAABABHgAAAAYBQEeAAAARkCA\nBwAAgBEQ4AEAAGAEBHgAAAAYAQEeAAAARkCABwAAgBEQ4AEAAGAEBHgAAAAYAQEeAAAARkCABwAA\ngBEQ4AEAAGAEBHgAAAAYAQEeAAAARkCABwAAgBEQ4AEAAGAEBHgAAAAYAQEeAAAARkCABwAAgBEQ\n4AEAAGAEBHgAAAAYAQEeAAAARkCABwAAgBEQ4AEAAGAEBHgAAAAYAQEeAAAARkCABwAAgBEQ4AEA\nAGAEBHgAAAAYAQEeAAAARkCABwAAgBEQ4AEAAGAEBHgAAAAYAQEeAAAARkCABwAAgBEQ4AEAAGAE\nBHgAAAAYAQEeAAAARkCABwAAgBEQ4AEAAGAEBHgAAAAYAQEeAAAARkCABwAAgBHYaVsXgPn5zre/\nmd875NCtsq7r7bVnXvV/j94q6wIAAECA364s7LBzrnHrx26Vdf3wy3+/VdYDAABApwk9AAAAjIAA\nDwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwAAACMgwAMAAMAICPAAAAAwAgI8AAAAjIAADwD/\nv737DpesqhI2/t4ONKlBwJZkBGFJRlCHIAqKCiKKmEURc0BFdBT9xBlmHCOjiAhiREmigoyOKIqS\nxIQKSnQBQ7JVtJUWaEBouu/3xz5FF5equrGr6lS9v+e5z6lbZ+/au86qvl3rnH32liRJqgETeEmS\nJEmSasAEXpIkSZKkGjCBlyRJkiSpBkzgJUmSJEmqARN4SZIkSZJqwARekiRJkqQaMIGXJEmSJKkG\nTOAlSZIkSaoBE3hJkiRJkmrABF6SJEmSpBowgZckSZIkqQZM4CVJkiRJqoE5K+uFI2Iz4HfA+Zn5\n7A7lNgA+AOwNbAwsBn4BfDIzL1xZ/ZMkSZIkqU5WyhX4iFgT+Bowb5xymwC/Bd4MPAq4E1gP2Bc4\nLyIOXRn9kyRJkiSpbmY8gY+IdYDvAzuMU242cBawAPglsEVmrktJ4D8NjABHRsSuM91HSZIkSZLq\nZkYT+IjYGbgU2BUYHaf4K4AA7gD2ycxrADLz9sw8FDit6t+HZrKPkiRJkiTV0Ywk8BExPyJOAi4C\nHgFcA1xIuYrezpsoSf5XM/PWFvs/Um13i4hHzkQ/JUmSJEmqq5m6Ar8JcAAlIT8e2BG4sV3hiFgd\neGL1649blcnMy4FF1a97zVA/JUmSJEmqpZlK4JcD3wF2zMyDM/POccpHU9tXdyh3bbXdapr9kyRJ\nkiSp1mZkGbnqavl+k6iyUdPjhR3K/ZEyDH/DqfRLkiRJkqRBsVKWkZuAtRoPMvPuDuUa+9bqUEaS\nJEmSpIHXqwS+ceV/6Tjl7hlTXpIkSZKkodSrxLhxZX3uOOXmVdt7p9vgggXzAVh1VZg1q9Pk+DNr\n1uxenSNZuebMnX3/Ma2DOvVVM8vYDzfjP9yM//Ay9sPN+GuQ9SqBv6PxICLmZeY9bcqtXm1vX/ld\n0mRcd+01PO3ZL+5KWxuvvy4nnXB8V9qSJEmSpH7VqwT+D02PNwaub1NuY8rSdH+aboOLFpVzBkuW\n3MHy5aPTfbkJW75sedfa6qblzGGNLQ7oSls3XXbi/fGbrMYZ2KnWV30Z++Fm/Ieb8R9exn64Gf/h\nNUyjLno1vvsaVtz/vnmHco+ttlet3O5IkiRJktTfepLAZ+Z9wE+rX5/eqkxEbAssqH69oBv9kiRJ\nkiSpX/VyhrXTKGu8vzYiFrTYf3i1PTczr+1etyRJkiRJ6j+9TOBPABJ4CHBORGwHEBFrR8QxwAuB\nZcB/9K6LkiRJkiT1h54l8Jm5FHgRsAjYBrg0IhYDfwMOpkxe97bMvKhXfZQkSZIkqV+s7AR+tPpp\nKTOvALYGjqHMRL8qZcm47wPPyEzXDpMkSZIkiZW4jFxmvhp49QTKLQLeUf1IkiRJkqQWenkPvCRJ\nkiRJmiATeEmSJEmSasAEXpIkSZKkGjCBlyRJkiSpBkzgJUmSJEmqARN4SZIkSZJqwARekiRJkqQa\nMIGXJEmSJKkGTOAlSZIkSaoBE3hJkiRJkmrABF6SJEmSpBowgZckSZIkqQZM4CVJkiRJqgETeEmS\nJEmSasAEXpIkSZKkGjCBlyRJkiSpBkzgJUmSJEmqARN4SZIkSZJqwARekiRJkqQaMIGXJEmSJKkG\nTOAlSZIkSaoBE3hJkiRJkmrABF6SJEmSpBowgZckSZIkqQZM4CVJkiRJqgETeEmSJEmSasAEXpIk\nSZKkGjCBlyRJkiSpBkzgJUmSJEmqARN4SZIkSZJqwARekiRJkqQaMIGXJEmSJKkGTOAlSZIkSaoB\nE3hJkiRJkmrABF6SJEmSpBowgZckSZIkqQZM4CVJkiRJqgETeEmSJEmSasAEXpIkSZKkGjCBlyRJ\nkiSpBkzgJUmSJEmqARN4SZIkSZJqwARekiRJkqQaMIGXJEmSJKkGTOAlSZIkSaoBE3hJkiRJkmrA\nBF6SJEmSpBowgZckSZIkqQZM4CVJkiRJqgETeEmSJEmSasAEXpIkSZKkGjCBlyRJkiSpBkzgJUmS\nJEmqARN4SZIkSZJqwARekiRJkqQaMIGXJEmSJKkGTOAlSZIkSaoBE3hJkiRJkmrABF6SJEmSpBow\ngZckSZIkqQZM4CVJkiRJqoE5ve6AJE3FwYccyi2LFrfdP2fubADuW7ps2m1tsGAdjj36qGm/jiRJ\nkjQdJvCSaumWRYtZe9sDu9PWZSd2pR1JkiSpE4fQS5IkSZJUAybwkiRJkiTVgAm8JEmSJEk1YAIv\nSZIkSVINmMBLkiRJklQDPZ+FPiJ+DewwTrEjMvM/u9EfSZIkSZL6UU8T+IiYDWwFjAKLgOVtii7p\nWqckSZIkSepDvb4CvwUwD1gKPDwz7+txfyRJkiRJ6ku9vgd++2p7tcm7JEmSJEnt9UsC/9ue9kKS\nJEmSpD7XDwn8KCbwkiRJkiR11Ot74Lertn+IiH8D9gI2Bm4FfgIclZk39KpzkiRJkiT1i54l8BHx\ncGA9yhX4kyiT2TU8nJLcvz4iDsrMr/egi5IkSZIk9Y1eDqHfvunxQuBFlIR+DeAZwK8pSf2JEbFz\n97snSZIkSVL/6OUQ+nuAsykJ+36Zubhp37kR8VTgF8DWwCeAXbrfRUmSJEmS+kPPEvjMPAc4p8P+\nuyPiCOAM4F8i4uGZuXCq7S1YMB+AVVeFWbNGpvoykzZrdq/nCay/OXNn3x+/qZpuffWfOXNnd7Ut\nP0P1ZNyGm/EfXsZ+uBl/DbJ+zy4vbHq8Zc96IUmSJElSj/V6Fvrx3Nb0eLXpvNCiRXcAsGTJHSxf\nPjqdl5qU5cuWd62tQXXf0mX3x2+yGmdgp1pf/eu+pcu62pafoXrx3/5wM/7Dy9gPN+M/vIZp1EUv\nZ6F/EbAJcGVmfrdNsfWbHt+y8nslSZIkSVJ/6uUV+LcCuwE/A9ol8M+qtncDl3ajU5IkSZIk9aNe\n3gPfSNp3iojdxu6MiLWA91OtE5+Z93azc5IkSZIk9ZNeJvCfBf5Y9eEbEbF/RMwFiIidgAuAx1CG\nzn+gZ72UJEmSJKkP9CyBz8wlwHOAPwEPA04HlkTE7ZRh9dtSEvxnZubfetVPSZIkSZL6QU+XkcvM\n34X1TT0AACAASURBVFES9Q8DlwNLq12XA/8JbJWZV/aoe5IkSZIk9Y2eLyOXmbdShsg7TF6SJEmS\npDZ6egVekiRJkiRNjAm8JEmSJEk1YAIvSZIkSVINmMBLkiRJklQDJvCSJEmSJNWACbwkSZIkSTVg\nAi9JkiRJUg2YwEuSJEmSVANzet0BaTw3XH8dL3j5QVOqO2fubADuW7ps3LIbLFiHY48+akrtSJIk\nSdLKZgKvvjc6ay5rb3vgSm/nlstOXOltSJIkSdJUOYRekiRJkqQaMIGXJEmSJKkGTOAlSZIkSaoB\nE3hJkiRJkmrABF6SJEmSpBowgZckSZIkqQZM4CVJkiRJqgETeEmSJEmSasAEXpIkSZKkGjCBlyRJ\nkiSpBkzgJUmSJEmqARN4SZIkSZJqwARekiRJkqQaMIGXJEmSJKkGTOAlSZIkSaoBE3hJkiRJkmpg\nTq87IPWLG66/jhe8/KCutLXBgnU49uijutLWwYccyi2LFnelrW6+r27q1mdjUI+fJEmSZoYJvFQZ\nnTWXtbc9sCtt3XLZiV1pB+CWRYsH8n11U7c+G4N6/CRJkjQzHEIvSZIkSVINmMBLkiRJklQDJvCS\nJEmSJNWACbwkSZIkSTVgAi9JkiRJUg2YwEuSJEmSVAMm8JIkSZIk1YAJvCRJkiRJNWACL0mSJElS\nDZjAS5IkSZJUAybwkiRJkiTVgAm8JEmSJEk1YAIvSZIkSVINmMBLkiRJklQDJvCSJEmSJNWACbwk\nSZIkSTVgAi9JkiRJUg2YwEuSJEmSVAMm8JIkSZIk1YAJvCRJkiRJNWACL0mSJElSDZjAS5IkSZJU\nAybwkiRJkiTVgAm8JEmSJEk1YAIvSZIkSVINmMBLkiRJklQDJvCSJEmSJNWACbwkSZIkSTVgAi9J\nkiRJUg2YwEuSJEmSVAMm8JIkSZIk1YAJvCRJkiRJNWACL0mSJElSDczpdQekYXTD9dfxgpcf1JW2\nbl64kG227UpTA/u+uqWbx+/Pf7yJDTd+VFfa2mDBOhx79FFdaaubDj7kUG5ZtLgrbXUrXt2M1co8\nfnPmzgbgvqXLgMH9DHZLNz/rg/IZHMvPYH34uVC/M4GXemB01lzW3vbArrR1340f7Eo7MLjvq1u6\nefyuv/GDXWvrlstO7Eo73XbLosUDF69uxqqbx29QP4PdMqixGtT3penxc6F+5xB6SZIkSZJqwARe\nkiRJkqQaMIGXJEmSJKkGTOAlSZIkSaqBvpjELiJeDBwMbE/p0w3AN4EjM/OuXvZNkiRJkqR+0PMr\n8BFxJHAa8GRgHnAfsAXw78AlEbGgh92TJEmSJKkv9DSBj4gDgHcBy4BDgPmZuTawB3ATsBlwcu96\nKEmSJElSf+hZAh8Rs4AjgFHgY5n5mcxcCpCZFwL7AMuBPSNi9171U5IkSZKkftDLK/B7AptSEvhP\njd2ZmVcB36l+PbCL/ZIkSZIkqe/0MoHfo9pelpl/a1PmR8AIsFd3uiRJkiRJUn/qZQK/JeXq+9Ud\nylxbbdePiHVWfpckSZIkSepPvUzgN6q2CzuU+WPT4w1XYl8kSZIkSeprvUzg16q2d3Yoc3eL8pIk\nSZIkDZ1eJvBzqu29Hcrc06K8JEmSJElDp5cJfOPq+iodysxretwp0ZckSZIkaaD18qr2HdV2tQ5l\nVm96fPt0GluwYD4Aa645h6W3L2TxZSdN5+Um5M4lt7HKvE7nJyRp8M2ZO/v+v8HdtjLbnTN39kp7\n7V7pZqy6efx6+RkcBHWM1UReo47vSxMznWPt50L9bmR0dLQnDUfEN4AXAKdkZst13iPiGcAPKLPV\nr5uZt3Wxi5IkSZIk9Y1eDqG/krLG++YdymxWbf9s8i5JkiRJGma9TODPq7aPj4i125TZs9qev/K7\nI0mSJElS/+plAv8Tyjrvc4D3jN0ZEdsA+1KGzx/f3a5JkiRJktRfepbAZ+Yo8H7KMPrDIuJ9EbEq\nQETsDny36t+PMvOiXvVTkiRJkqR+0LNJ7Boi4jjgjZREfinwT2A+5cr774FdM/MfveuhJEmSJEm9\n1/MEHiAi9gfeAuxAWVbuJuAM4KOZeUenupIkSZIkDYO+SOAlSZIkSVJnvZzETpIkSZIkTZAJvCRJ\nkiRJNWACL0mSJElSDZjAS5IkSZJUAybwkiRJkiTVgAm8JEmSJEk1MKfXHVhZIuLFwMHA9pT3eQPw\nTeDIzLyrl30bJhGxFvB24HnA5sA84M/AecAnMvPKNvU2AD4A7A1sDCwGfgF8MjMvHKfNKcW+F20O\no4j4EPA+4PzMfFqbMsZ/AETEFsC7gT2ADYE7gV8Dx2bmdzrUM/41FhFrAocC+wObAbOBm4HvU47J\nHzvUNfY1FBGbAb+j/F1/dodyQxHf6bRZR5OI/1Mpx3Nn4GHAXcAVwNeAL2Tm0g51jX+fmmj8W9Tb\ngBL/dYHdOx0b4/9AA7kOfEQcCbwLGAWWAvcAawIjwDXAbpm5qHc9HA4RsTnwQ+CRlFj8E1gOrE6J\nxb3AazPzlDH1NgF+RvnjPgrcBsxnxYiRf83Mo9q0OaXY96LNYRQRuwLnU47rBa0SeOM/GCLiIOBz\nrDhRfAflmEI5Np/KzHe2qGf8aywiNqb8G9+UckzuoxyX1SjH5B/APpn58xZ1jX0NVSdszgceD/yg\n3Rf4YYnvdNqso0nE/yPAYZRj0jguD6l2jwA/B/bKzDta1DX+fWqi8W9T9/vAsyjHaY92ya3xf7CB\nG0IfEQdQDvgy4BBgfmauTbkCdBPlasDJvevhcIiI2cD/UJL364FnZOYamTkf2JZyBX4V4EsRsf2Y\nemcBC4BfAltk5rrAesCnKf9wjqwSwbFtTin2vWhzGFV/5E+iw98d4z8YIuIpwJcoyftXgA0z8yHA\nBsCXq2KHRMS+Y+oZ//o7kZK8LwZeCqyemWsCTwaupnxhPz0i1miuZOzrKSLWoYys2GGcckMR3+m0\nWUeTiP8BrEjePwVskJnrAWsBbwOWADux4v+HsXWNfx+aaPzb1D2YkryPV874tzBQV+AjYhaQwCbA\nRzLz8DH7t6QM8ZgFPD0zz+96J4dERLwMOIVy9WXHzLx8zP55wCXA44DTM/Ml1fOvAk4Abgc2ycxb\nx9Q7lfKl8MLM3L3p+SnHvhdtDqOI+DJwEGXI3Gq0uAJv/AdDRFwObAmckJmva7H/POApjDlbb/zr\nLSIeCdxI+ZL+ysw8dcz+x1KuXIwCB2TmaU37jH3NRMTOlKHPjVF20OYK3LDEd6pt1tEk45/AY4Ez\nMvPFLfYfSDnZOwpsnZlXV88b/z41mfi3qPs44DdVvdVpcwXe+Lc3aFfg92TFsL1Pjd2ZmVcBjfsu\nD+xiv4bR3pQ4nDc2eQfIzHsoZ69GgKc27XpTVe+rYz/4lY9U292qL4sN04l9L9ocKhHxfEryfilw\nGiXurRj/mouIfwG2ovwH9q9tih0GvJNylb6Z8a+3DZse/3Lszsy8DvhT9esjxuw29jUREfMj4iTg\nIkocrwEupP3fdRie+E61zdqYbPyjDCnerPq13dDhk4G7q8fN3wmNf5+Z4r//5vpzKPFelfJdoBPj\n38agJfB7VNvLMvNvbcr8iPIh26s7XRpalwJnUIbWtPPnarsWQESsDjyxeu7HrSpUJwMa95w0x3BK\nse9Fm8MmItYHPk+ZA+GVlHuJWpUz/oPhedX2e5n5j1YFMvPizDw6M89oPGf8B8L/UYYcQpmk6gEi\n4hGU2ygArmt63tjXyybAAZQvqscDO1JGXrQ0LPGdZpt1Mqn4Uyax/DJlTqSrWhXIzOWsOC5rNe0y\n/v1nsvEf6wjKkPtTgNPHKWv82xi0BH5Lygfq6g5lrq2261f3bmglyMyjMvPFmfmgs1dNGveALKy2\nwYrP5ERiuFXTc1ONfS/aHDYnUGYYPbw6c9mO8R8M21GOy2+h3E4TEd+LiOsi4oqIOD4iNm1Rz/jX\nXPVl57PVr0dFxP7V1RYiYkfg25QvPZdWjxuMfb0sp1yB2jEzD87MO8cpPyzxnU6bdTKp+GfmtZn5\nuszcOzNva1UmIh7OilE5C5t2Gf/+M9l///eLiF0oV93/ALx1AlWMfxuDtozcRtV2YYcyzcvXbEiZ\naEddFhGPAV5O+UdyVvX0Rk1FxovhCA8crjnV2PeizaEREW+hnGW8IDM/OU5x4z8YHldtb48HzjDb\nsCVwYES8NB+4lJzxHwCZ+faIWEiZAOh04L6IuJdyn+NS4KvAO6srbg3Gvkaqq0j7TaLKsMR3Om3W\nxhTiPxGHU47JUuCcpueNf5+ZavyjTFx6EuW9vyYzb4+I1capZvzbGLQr8I1hN53OBt3d9HittqW0\n0kSZwO5rlInM/gl8otp1fzwy8+4WVRsa+5rjN9XY96LNoRARAXycsnzYQROoYvwHQ2NZoH8Dngkc\nQxlytypl1M2vq8enRlknvsH4D4BqGOF6wFzKiZvZlL/1o5TvHLMpS/E0M/aDbVjiO502h1ZE7A+8\nnvI34nP5wKW5jP/g+DTwGOC4zGw5xLwF49/GoCXwjREF93Yoc0+L8uqSiJgLfAt4EuWP9dsys3Gm\nqhGPlvdIN2nEsDl+U419L9oceFGW0jiZ8sX9HZl50wSqGf/B0EjONgCOzMxDMvOmzFyamb8AnkZZ\nWnI14INN9Yx/zUXEqpT7XN9NOaYHAutQPhPPoUx29Erggohovvpg7AfbsMR3Om0OpYjYkxXLcf0e\neO+YIsZ/AETEfsCrKbO7v2cSVY1/G4OWwDfOaqzSocy8psedgqMZVg2f+R4rZqj/RGY2r/nZiN/c\ncV6qEcPm+E019r1ocxgcQZnY5H8z84QJ1jH+g2OEcvb6P8fuyMwlwH9XZfapRuSA8R8EbwJ2oYys\nenpmnpKZt2fm3Zn5fcpa8DcCjwY+2lTP2A+2YYnvdNocOhHxPOB/KSOybgH2zcy7xhQz/jVXTWT8\nBcqy0q/MzH9Oorrxb2PQEvg7qm2neypWb3p8+0rsi5pUV1t+Ajydkrx/MjPHnoW7o6n8PNprxLA5\nflONfS/aHGgRsRPlLPoi4A2TqGr8B8MdlH/jl7T4MtZwUbVdhRXLCxn/+nsFJfYnZ+YVY3dm5mLg\nvygnb15aDbcHYz/ohiW+02lzqETEwZQ5Mlah3C+8R2Ze36Ko8a+/L1MmMv5oZv56zL7xlp4z/m0M\nWgL/h2q7cYcyzfv+3LaUZkxEbAn8HNie8uXuA5n57hZF/9D0eLwYjrJiPeHmupONfS/aHHRvoNzn\nOh/4XUT8ufmHFetm7tr0/E4Y/0Hxl2q7pEOZ5gm9Gv+ZGf/6a6wu8PMOZX5SbedQ7ocEYz/ohiW+\n02lzaETExylzo8yizMj95My8pk1x419j1Vrne1MS9Te1+D74u6biZ1bPn9H0nPFvY9AS+CspH5LN\nO5RpXO35c7vlLDRzqsTsJ8AjKcNnXpuZH25T/BpW3DvSKYaPrbbNS5JNNfa9aHPQjVD+OM0DHtbi\nZ9Wq3Jzq9wWUs/DGfzBcTjkuj+hQZt2mx7dUW+Nff42hg52+WzQPG2z8LTD2g21Y4judNgdeRIxE\nxEnAv1K+I1wM7JqZN3eoZvzrrfF9cJQyuenY74MPbSr7kOq5hzQ9Z/zbGLQE/rxq+/iIWLtNmT2r\n7fkrvzvDLSK2pdzzvg5wF7BfZn6lXfnMvA/4afXr0zu85oLq1wuadk0p9r1oc9Bl5qszc3a7H+Dz\nVdELqufmZOaFxn9gnF9tt4qIdmehd6u2f6M6a238B0LjKtouHcrsWG2XA/8Hxn7QDUt8p9nmMPgC\ncAAlmTsbeFpm/n2cOsa/xqoJbDt9H2xeem336vnmY2f82xi0BP4nlPX15tBilsOI2AbYl/LH4/ju\ndm24VBPWfYtyJu1OYK/M/N4Eqp5GOfP12ohY0GL/4dX23My8tun56cS+F22qNeNff9+gTOYyQrnf\n+QGqdV/fQTkuX8/M5jXijX+9nc6K+9sfdAUiIuawYpbpczPzH027jf1gG5b4TrXNgRYRbwZeQzlm\nZwLPHWeprQbjPzxa3Q9v/NsYqAS++iL4fsrBOywi3lcta0NE7A58l/Kef5SZF7V9Ic2EwylrP48C\nb57E8T6BsszEQ4BzImI7gIhYOyKOAV4ILAP+o7nSNGPfizbVmvGvucy8lbIG/Ajwqog4vvEfWURs\nShmV81jgVh64jBwY/7r7NHADZfKfcyPiuRExCyAitgC+DzyBsozOYWPqGvvBNizxnVKbgywi1gU+\nRvk+eCXwisxcNpG6xn+4Gf/2RkZHR8cvVTMRcRzwRsrBX0pZ0mY+5Y/H7yn33Pyj/StoOiJiFeCv\nwFqUY/7XCVR7Qmb+saq/NfBjyr0xI8BtlHWEZ1evd3BmtryqMdXY96LNYRURn6Ucr/Mz82kt9hv/\nARARn6BcaW+cVb8NWJtyXP4BPD8zL2xRz/jXWEQ8DjiLslRc45jcxYrY30X5Av/tFnWNfU1FxAnA\nq4CzM/PZbcoMRXyn02ZddYp/RLwXaMx9dAflb0Anp2XmoWNew/j3sYn8+29Tb33KJHCjlJUIHvSd\noCpn/McYqCvwDZn5FuBFwLmU4duNCbI+Cuw0LP+h9tDWrPiQQ+uJzJp/FlA+2ABkWX5oa8ospddT\nJjq6nXL15hmdPvhTjX0v2hxyjUlNHsT4D4bMfBflPrAzKTPTz6Pc8/wZ4PHt/qM2/vWWmb8HtqOM\nwrqE8qVnHmW26WOBbVol71VdY19vbf+uw/DEdzpt1ly7+O/StG9Nxv9OuNbYFzD+tdDx3/8E6rZl\n/B9sIK/AS5IkSZI0aAbyCrwkSZIkSYPGBF6SJEmSpBowgZckSZIkqQZM4CVJkiRJqgETeEmSJEmS\nasAEXpIkSZKkGjCBlyRJkiSpBkzgJUmSJEmqARN4SZIkSZJqwARekiRJkqQaMIGXJEmSJKkGTOAl\nSZIkSaoBE3hJkiRJkmrABF6SJEmSpBqY0+sOSJL6X0Qsb/H0pzLzndX+84GnAKOZOXuCr/kq4ITq\n14My88Q25WYDLwL2A54ArA+MAn8FrgD+FzgtM++cZP8blgP/BBYBVwJfB76Zmf+cyPuQJiMitgMu\nbbHrvZn58W73R5JUL16BlyRN1OiYn7H7pvO6LUXEtsAlwKmUJP4xwOrAGtXjfYEvAL+PiH0n0E6r\nnxFgNeCRwN7AV4ErImLnqb8lqaNWn0NJksblFXhJ0mScDnyoerxoZTYUEY8AzgY2oFwh/xJwLvBH\nSsLzCGAv4EBgY+D0iNg3M3/Y4WV/A7xuzHOzgfmUEwL7APsDmwDnRMRTMvOSGXtTEiTw+OrxRsD3\nMIGXJE2QCbwkaTL+npmXdamtj1CS9zuA3Vq0+yvgWxHxZeAcypX5L0TEJpm5rM1rLunQ/wuBr0bE\n3sC3KFflvxMR22Tm4um+GQmgujXjMoCIuK3H3ZEk1YxD6CVJfSci1qAMmR8Fjul00iAzfwF8mDIU\n/uHAc6fTdmZ+H3hH9XobAodO5/UkSZJmigm8JKkfbQrMrR5fN4HyJzU93nq6jWfm5yhXSUeAt0bE\n6tN9TUmSpOlyCL0kqd/tAnylU4HMXFjNan8rZSb5mfBF4NPA2sCTgU731j9A06z8RwAfBA4GXgts\nBtxHmYX8C5l56jivswbwduB5wObAqsCfgPOBT7cbmdA06/4Lgb8An6Kc2LgNuBh4QWYuHaftxioB\nN2bmJhHxL8D7gZ2AtYAbgf8BjszMv4/zWs8BXl3VXa/qx6XAKcDJmfmge8Aj4iuU+Q2+C7wC+Ex1\nHKCc1HlXZp5fld2Bcox3p8yH8E9gIfBj4NjMvKZD31YDXg+8gHKM1qCscPBT4POZeV6beudTYvyZ\nzHx7NYniW4AdquNzM2WFhI9l5kqdL0KSNDy8Ai9J6kfXAndRroC/JiLeGhEdTzpn5kmZeVZm3jhD\nfWhO3J46ybqNmcVnAd+knAjYlnJf/fzq9U6OiNOqZfIeJCJ2pByHDwFPpJxImEeZbO81wCUR8YFx\n+rADZX6AHYBVgAXAGuMl7y368gLgJ5RJ/tajjI7YDHgP8LuIeFybeqtFxJnAdyjLAG5Q1X0o8AzK\njP8XRcSCNv0fpVxsOIuSxK9Z/WwHXF+18UbKSYlXA4+uXn8+sAXwNuDKiHhNm/5tC1xOOcHxZOAh\nVf2NgZcAP46Ir0TE3BbV7589PiI+BXwbeGb13lYBHgu8s2p/81btS5I0WSbwkqS+k5l3A1+ufh2h\nJMA3R8QxEbFPRKzZhW5cy4rZwac6LP+NlFntFwJvAnYGXkZZGm+Ucp//MWMrRcSmlKvH61Nm+z+c\ncnV5Z8rV4sspx+WIiOh0j/57gaWUq/i7AocAR03yPaxHSbRHqr7uDjydsnzfcspM6udGxFot6n6T\nctV8OXAyJYl/EmWegpOr53cGvtvmBM0IJSnehXK1fnfg+cCHM/PmKjE+pip3CXAQsBvwNMpogcWU\nVQaOi4hHNb9wtcrBDygrDiwHTqScoNiJMlriCkqMDqSMxmjnRZTj+3vKCgc7Ve/zgqr+esBnO9SX\nJGnCHEIvSepXh1GSvSdWv29AGSZ9MLAsIn5Dubr87cz89Uw3npn3RsRiYB1KEjZZI5QE/FrgyU3D\nqC+urkqfDewBvD4ijsvMK5rqfp4yDPu6qu5fm/ZdHBEnUa747gV8OCJOzcy/tOnDGzPztOr3X0zh\nfcynJLgvycwzmp4/PyJ+RUnk16ckzIc1dkbEK4FnV3Vfmpmnj3ndsyLiR5STA0+gXC1vdXJhBPhh\nZr6y6bnvVNuXUb7L/BV4Smbe1VTmgoi4iLK6wFzg5ZSVDRqOrPo9ChyUmSc37ftVRJxStfMs4BUR\n8bXMPLtF3x5GGQGwR3XiqfH+vwf8jPL53T0iNsrMP7V4f5IkTZhX4CVJfalKhp4CHA3cS9OQZcpV\n1SdRksaLI+LyiNhjJXSjMYx/3SnWHwVeP/Ye6GoI+6uBZZT/i1/d2BcRW1MS+1Hg0DHJe3P9N1Zl\nVqFcMW7lduDrU+x783s4cUzy3ujHlygjBUaa30Pl7VXdM1sk7436JzXVf3OHPnyhzfMbVNtFY5L3\nxutfREnUj6DcAgBARGxImR9gFPjmmOS9UXcp8ErKZwDKygTtvK85ea/qL+OBkytu1aG+JEkTYgIv\nSepbmXlvZr6Tsjzc24AfAXezIplvJPRbUe5X/vcZ7sK8qo3l4xVs49rMvLDVjsy8mXJ1eIRypbrh\nWU2PW06gVtX/A3B19evubYpd2mqCuCn4fId9J1bb9aqJ7oiI9YAdq+fbvofKD6rtphGxUZsy7UZY\nZLXdMiI+ExEbPKhA5mGZ+cEqmW/YgxXfgU5o17HM/Btlor4RYLc2w/yX0nRyYIwbmh5347YPSdKA\ncwi9JGkmTDdJ7Fi/SqSOo9zLvArlfu5nUO5ZbtyfPgL8W0Rck5lfm2Z/Ghr3dd82hbqjwM/HKfNb\nSjK5adNz2zU9XhIRE2nrMW2en4kh20tpn0BDeQ8NmwO/5IHv4TMR8ZkJtvUYWve53fs4GXgfZeK4\ntwBvjohfU25POBv4RZsTGFs0Pf7VOH26mDL8flXK/fJjZ7T/S2be16bunU2P/c4lSZo2r8BLkmbC\nvY0HETEywTqrNj2+Z6INVVflz8vM/5eZ21GG0jcnyv820dfqJCIas74D/HmKLzNevb9V29kR8dDq\ncfP99qMT/Fm7zevfMYU+j3VrhwQVVrwHKPeUw9TeA5RZ4Me6t1371YmdZwJXNb3GE4EPUJaBu6Wa\n+PARY6o2+rc8M2/t8N6gTCLYsE6L/Xe2eK6h+eTBRP9dSJLUlmeDJUkzofkK9ZpMLHGc36Z+Ywj2\nRsCSzLyBDjLzNxGxJ+XK7zbA5hGxcWb+cUI9b+/xTY8vneJrdEp8odzL37B0zHOLKKMMJpL4LWvz\n/EwMn5/Oe4Byb/tEJ89rFevxRmf8DtgmIp5Bua99H2DDavdDKZMeHhQRz87MxlD3ySTTze9lJo6n\nJElTZgIvSZoJC5seb0xZUms8zcO+F47Zt5By9fs0yvDljjLznxHxRcqEd1CS/+km8E9pevzTKb7G\neJPfNdY/vyczGycxFlfb+Zl5+RTbnUmtrjo3a17D/ZZqu7jpudsy87KZ7dKDZeY5lFUJiIitgL0p\ns9RvD6wOnBARm1VD6htX3WdFxDqZubjVa1aa3994V+slSVqpHEIvSZoJzQna49uWeqDGJGf3sGIy\nsobGGuy7t5k4rJXmoczTuve7ug3goOrXv1DW9J6sER54L3grO1Tb5kT9qmq7akRs06lyRLwnIt4Q\nEbtOoX8TtXq1Ln07OzQ9bnwOrmp67kmdXjwinhkR74qI/SJi9cl0LCJWjYjtI2KT5ucz88rM/O/M\n3BE4nRKLxwCNCQWubCr+RDpr9P9e4KbJ9E+SpJlmAi9JmgnnsGJ48cHjFY6IzSmJ0yhwbot7nL/L\ninXU3zLBPuxdbW+YgeHz7wAeXfXv2GnM5L5TRDyy1Y6I2AzYpWrjf5p2/ajp8RvavXBE7AJ8FDie\nsqTcyvSyDvsOrLbXZebVcP8M+ddSYviycRLzoylLvZ0yhX4tBC6hLBPXzg+aHjfmXTifFbcdvKZd\nxYh4GLAvJUY/q5aWkySpZ0zgJUnTlpl/Ar5NSdh2jogPtysbEQ8BvsKK+5A/1aLY0ZT74keA/46I\nTmuEExFvB15ASbQ+Odn+j3mt/YAPVb/eNM3XmwV8sZo5v7mN1SjLl40ASyjHA4DM/DklKR0B3hQR\nz2nRx7VZsbTbKHDsNPo4nhHgPa1GA1THfbeqD8eM2d2YeX59yjF40HeOiPgvylXxxlrzD1rLfRxn\nV9sXRsQObcq8sNreSTXSo/q8nkl5by+KiAfdplHF7ETK8HtYcXuGJEk94z3wkqSZcghlebeHAu+t\nJpY7gXI//BLgYcDOwOuqx6PA8Zn5o7EvlJl/iYjnA98HVgGOjYi3AF+jTCi3iLLE2zaUq8NPdO+z\n5wAAA8RJREFUql7v25l5XIc+rhkRY4e1z6XM4r4l8Fzg6dXztwP7Z+bdkzkILewJ/DwijgSupyy1\ndhhl7fpR4N8zc+xs9a+jzKw/DzgzIk6gDAVfQnnP76YsaTYKfDkzfznNPnYyCqwBXBQRH6Os674m\n8ArgldX+SynL/DU7DngJZZTBS4HNIuJoyjJsG1JuUXheVfYWyszxk/VR4MWUK+vnRcSnKfMV/AN4\nFGVkwu5VH48ZE8t3VPseCpxUTYL3dcp97lsA76IsUTgKnJyZ35lC/yRJmlEm8JKkGZGZf4iIp1KS\noK0p97g/oUXRUcrM5h/PzMM7vN75EfF04HOU5HpL4L/avN5yylXowzp0caTqT6cZ5RtLml0JHDDN\nSeRGgYuqdncFTm3RzuGZ+aARCJn524jYG/gmZSK811U/Y+t/gzLL+8r2ccpJg7HHv7HW/XMzc3nz\njsxcFhH7AGdQ1rrfgXJFe2z9m4HnVEvCTUpmXhERr6OMRlgTeH+LYqOUyRA/MKbunyJid8rtGo8C\nXlX9NNcbpZyIOHSyfZMkaWUwgZckzZjM/H1EPJ4ybHl/StK2EeUq+hLgOsr9x1/MzGsm8Ho/i4jt\nKUuD7Uu5b/5hlKT2Nso90D8ETsnMKzq8VKd72JdXfVtISe7PpFzJX96hzkSMUIZt7wu8k3LF+dGU\nSfF+CnwyM9ueTKhOYDwWeCvwHGAzytJ7f6csmffFzDxrnD7M1LJnR1HuJX8fZbTDfZSJ6r4InJqZ\nLZexq2bW3zMi9qdcsX8i5Yr3PZSRGWcCx2Vmu2UHm9eIbykzT4yIi4G3AU+lJONzKMf5F8BXMvPs\nNnWviogtgTdRbsHYAlgN+APlxMRnM/PiDs2P278JlpEkaUJGRkf9P0WS1FlELKckIZ/LzIlOKje0\nIuI8SjJ5dmY+u9f9mYqIeBXlFohRYMPM/GuPuzRwIuJRwA2UY/y+zPx4j7skSepzTmInSZIkSVIN\nmMBLkiRJklQD3gMvSZqM9ZpmcV9ULcel1rxHTQ8SEatSls4D2LiXfZEk1Y8JvCRpokaAF1U/UNZv\nf2fvutP3RsYvoiEUdF4JQZKktkzgJUkT0epqsleYOxuE2ccH4T30I4+pJGlKnIVekiRJkqQacBI7\nSZIkSZJqwARekiRJkqQaMIGXJEmSJKkGTOAlSZIkSaoBE3hJkiRJkmrABF6SJEmSpBowgZckSZIk\nqQZM4CVJkiRJqgETeEmSJEmSasAEXpIkSZKkGjCBlyRJkiSpBkzgJUmSJEmqARN4SZIkSZJq4P8D\no9qh39Z80i0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x105017940>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 375,
       "width": 504
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.income_per_person.hist(bins=30)\n",
    "plt.title('Income per person distribution');\n",
    "plt.xlabel('[USD per person]')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is common to apply a log transform to this type of variable to reduce the skewness."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>completion_male</th>\n",
       "      <th>completion_female</th>\n",
       "      <th>employment</th>\n",
       "      <th>life_expectancy</th>\n",
       "      <th>log_income</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Argentina</th>\n",
       "      <td>97.91143</td>\n",
       "      <td>102.10666</td>\n",
       "      <td>58.400002</td>\n",
       "      <td>75.5</td>\n",
       "      <td>9.465967</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Armenia</th>\n",
       "      <td>91.53127</td>\n",
       "      <td>94.58897</td>\n",
       "      <td>39.400002</td>\n",
       "      <td>72.2</td>\n",
       "      <td>8.749288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Austria</th>\n",
       "      <td>100.63802</td>\n",
       "      <td>101.12819</td>\n",
       "      <td>57.099998</td>\n",
       "      <td>80.0</td>\n",
       "      <td>10.656714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Azerbaijan</th>\n",
       "      <td>91.68856</td>\n",
       "      <td>90.66259</td>\n",
       "      <td>59.299999</td>\n",
       "      <td>69.9</td>\n",
       "      <td>9.264091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bahamas</th>\n",
       "      <td>99.76959</td>\n",
       "      <td>96.84352</td>\n",
       "      <td>66.900002</td>\n",
       "      <td>72.1</td>\n",
       "      <td>10.133238</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            completion_male  completion_female  employment  life_expectancy  \\\n",
       "country                                                                       \n",
       "Argentina          97.91143          102.10666   58.400002             75.5   \n",
       "Armenia            91.53127           94.58897   39.400002             72.2   \n",
       "Austria           100.63802          101.12819   57.099998             80.0   \n",
       "Azerbaijan         91.68856           90.66259   59.299999             69.9   \n",
       "Bahamas            99.76959           96.84352   66.900002             72.1   \n",
       "\n",
       "            log_income  \n",
       "country                 \n",
       "Argentina     9.465967  \n",
       "Armenia       8.749288  \n",
       "Austria      10.656714  \n",
       "Azerbaijan    9.264091  \n",
       "Bahamas      10.133238  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# take the log of income_per_person \n",
    "# - this variable follows approximately a log-normal distribution\n",
    "df['log_income'] = np.log(df['income_per_person'])\n",
    "\n",
    "# for convenience, we drop the original feature\n",
    "df.drop('income_per_person', axis=1, inplace=True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA8QAAALJCAYAAAByReqdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1050179b0>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 356,
       "width": 482
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.log_income.hist(bins=30)\n",
    "plt.title('Log(income per person) distribution');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(105, 5)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Relationship between log_income and life_expectancy\n",
    "\n",
    "We can quickly use regplot to check the relationship between log_income and life_expectancy. We expect that these two variables are correlated."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10c6d12e8>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABAcAAALaCAYAAABNtDqJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10c6db0f0>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 365,
       "width": 515
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.regplot(data=df, x='log_income', y='life_expectancy')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While the relationship is roughly linear, note the presence of large number of outliers, especially for small log_income.\n",
    "\n",
    "If we have many variables that are correlated with each other, we can use PCA to reduce the dimensionality by extracting only the essential information from each variable. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Perform PCA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this section we perform PCA using the `sklearn` module. \n",
    "\n",
    "We first scale the data so that it has a mean = 0 and standard deviation = 1. This is done because our original features have very different ranges (e.g. life-expectancy is between 44 and 82, and log_income is between 6.4 and 11). Later we see what the effects are when we skip this rescaling step. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# we need to scale data with mean = 0 and stdev = 1\n",
    "scaled_data = preprocessing.scale(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-1.03975554  0.23058621]\n",
      " [-0.67166986 -1.75975392]\n",
      " [-1.80668446  0.46156378]\n",
      " [-0.21292093  0.13139593]\n",
      " [-0.85548708  1.05467569]\n",
      " [-0.56685139  1.01592655]\n",
      " [-0.51592526 -0.39585543]\n",
      " [-1.38476499 -0.376583  ]\n",
      " [-0.48263095 -0.06380433]\n",
      " [ 1.20094548 -0.04428261]]\n"
     ]
    }
   ],
   "source": [
    "# we will use PCA for exploratory visualizations - we keep 2 components only\n",
    "pca = decomposition.PCA(n_components=2)\n",
    "\n",
    "pca.fit(scaled_data)\n",
    "transformed = pca.transform(scaled_data)\n",
    "\n",
    "# let's take a look at the first ten rows of the transformed data\n",
    "print(transformed[:10, :])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Each row represents a country, and the values are the scores of that country along the particular component (PC1 or PC2 in our case)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Principal components"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can study the relationship between the original features and the components."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>completion_male</th>\n",
       "      <th>completion_female</th>\n",
       "      <th>employment</th>\n",
       "      <th>life_expectancy</th>\n",
       "      <th>log_income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PC1</th>\n",
       "      <td>-0.502837</td>\n",
       "      <td>-0.508865</td>\n",
       "      <td>0.226403</td>\n",
       "      <td>-0.466181</td>\n",
       "      <td>-0.468644</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PC2</th>\n",
       "      <td>-0.017325</td>\n",
       "      <td>-0.018845</td>\n",
       "      <td>0.936710</td>\n",
       "      <td>0.260947</td>\n",
       "      <td>0.232003</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     completion_male  completion_female  employment  life_expectancy  \\\n",
       "PC1        -0.502837          -0.508865    0.226403        -0.466181   \n",
       "PC2        -0.017325          -0.018845    0.936710         0.260947   \n",
       "\n",
       "     log_income  \n",
       "PC1   -0.468644  \n",
       "PC2    0.232003  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "utils.components_table(pca, df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "For example, PC1 places positive weight (0.23) on employment and negative weight on all the other features.\n",
    "PC2 is dominated by employment and contains essentially no information about completion rates."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we perform PCA on the unscaled data (the original df), the third feature (income_per_person) receives very little weight because its raw variance is much smaller than that of the rest of the features. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>completion_male</th>\n",
       "      <th>completion_female</th>\n",
       "      <th>employment</th>\n",
       "      <th>life_expectancy</th>\n",
       "      <th>log_income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PC1</th>\n",
       "      <td>-0.616577</td>\n",
       "      <td>-0.743552</td>\n",
       "      <td>0.129118</td>\n",
       "      <td>-0.222275</td>\n",
       "      <td>-0.029738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PC2</th>\n",
       "      <td>-0.080651</td>\n",
       "      <td>-0.073170</td>\n",
       "      <td>-0.988500</td>\n",
       "      <td>-0.104575</td>\n",
       "      <td>-0.008602</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     completion_male  completion_female  employment  life_expectancy  \\\n",
       "PC1        -0.616577          -0.743552    0.129118        -0.222275   \n",
       "PC2        -0.080651          -0.073170   -0.988500        -0.104575   \n",
       "\n",
       "     log_income  \n",
       "PC1   -0.029738  \n",
       "PC2   -0.008602  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca_unscaled = decomposition.PCA(n_components=2, whiten=False)\n",
    "pca_unscaled.fit(df)\n",
    "utils.components_table(pca_unscaled, df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploring PC properties"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Components are orthogonal to each other (dot product = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.dot(pca.components_[0], pca.components_[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Components are normaled to have length 1. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1.,  1.])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(pca.components_.T**2).sum(axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### The first two components together account for 84% of the variance in the data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.66304229,  0.18088559])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca.explained_variance_ratio_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Biplot visualizations\n",
    "\n",
    "A biplot is a scatterplot where each data point (country) is represented by its scores along the principal components. The axes are the principal components. In addition, the biplot shows the projection of the original features along the components. \n",
    "\n",
    "A biplot can help us interpret the reduced dimensions of the data, and discover relationships between the principal components and the original features. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10c69aef0>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABa4AAAMlCAYAAACbxGQqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10bed9b70>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 402,
       "width": 727
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "utils.biplot(df, transformed, pca)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Observations:\n",
    "- There are pairs of correlated variables: the completion rates, as well as life_expectancy and log_income. \n",
    "- Interestingly, the employment variable is largely uncorrelated with the rest.\n",
    "- PC2 largely represents the employment variable. \n",
    "- PC1 is a linear combination of the rest of the variables"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is now easy to interpret the relateve positoin of a given point. For instance, a point in the top right corner will have a large employment percentage, but low completion rates. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Difference in (male - female) completion rates\n",
    "Because the two completion rates are strongly correlated, we will construct a new feature (the difference male - female) and explore its relationship with the rest of the variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Let's take a look at the difference in completion rates: \n",
    "df_diff = df[[\"log_income\", \"employment\", \"life_expectancy\"]].copy()\n",
    "df_diff[\"diff_completion\"] = df.completion_male - df.completion_female"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-0.90597675  0.19996421]\n",
      " [-0.80896797 -1.78711098]\n",
      " [-1.61095973  0.46755818]\n",
      " [-0.10251782  0.13118676]\n",
      " [-0.34930845  1.08249442]]\n"
     ]
    }
   ],
   "source": [
    "pca_diff = decomposition.PCA(n_components=2)\n",
    "scaled_diff = preprocessing.scale(df_diff)\n",
    "\n",
    "pca_diff.fit(scaled_diff)\n",
    "transformed_diff = pca_diff.transform(scaled_diff)\n",
    "\n",
    "print(transformed_diff[:5, :])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>log_income</th>\n",
       "      <th>employment</th>\n",
       "      <th>life_expectancy</th>\n",
       "      <th>diff_completion</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PC1</th>\n",
       "      <td>-0.613868</td>\n",
       "      <td>0.287496</td>\n",
       "      <td>-0.593018</td>\n",
       "      <td>0.434560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PC2</th>\n",
       "      <td>0.229222</td>\n",
       "      <td>0.933890</td>\n",
       "      <td>0.265717</td>\n",
       "      <td>0.068568</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     log_income  employment  life_expectancy  diff_completion\n",
       "PC1   -0.613868    0.287496        -0.593018         0.434560\n",
       "PC2    0.229222    0.933890         0.265717         0.068568"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "utils.components_table(pca_diff, df_diff)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10c362048>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABa4AAAMlCAYAAACbxGQqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAIBWEVwDAAAAANAq/z+C+y11ddr9VwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10bf21d68>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 402,
       "width": 727
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "utils.biplot(df_diff, transformed_diff, pca_diff)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The new diff_completion variable behaves very differently - it is inversely correlated with life_expectancy and log_income - we can see that its projection points in the oppositve direction from the other two features.\n",
    "\n",
    "This means that poor countries tend to have a larger gender inequality (in favour of male completion rates). \n",
    "\n",
    "Remember that our dataset only captures 105 countries, and many poor countries are not included. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Interactive visualization\n",
    "\n",
    "Finally we will produce an interactive visualization that will allow us to explore the PCs together with the original features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# we create a dataframe out of the transformed_diff data\n",
    "df_transform = pd.DataFrame(index=df.index, data=transformed_diff, columns=[\"PC1\", \"PC2\"])\n",
    "\n",
    "# merge dataframes based on index - \"country\"\n",
    "# we now have a dataframe that contains both the components and the original features\n",
    "df_transform = df_transform.join(df_diff, how='inner')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will use the mpld3 library for the visualization. Hover over a data point (country) to learn more about it. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "\n",
       "<style>\n",
       "\n",
       "table\n",
       "{\n",
       "  border-collapse: collapse;\n",
       "}\n",
       "th\n",
       "{\n",
       "  color: #ffffff;\n",
       "  background-color: #000000;\n",
       "}\n",
       "td\n",
       "{\n",
       "  background-color: #cccccc;\n",
       "}\n",
       "table, th, td\n",
       "{\n",
       "  font-family:Arial, Helvetica, sans-serif;\n",
       "  border: 1px solid black;\n",
       "  text-align: right;\n",
       "}\n",
       "\n",
       "</style>\n",
       "\n",
       "<div id=\"fig_el82645032962085170635785\"></div>\n",
       "<script>\n",
       "function mpld3_load_lib(url, callback){\n",
       "  var s = document.createElement('script');\n",
       "  s.src = url;\n",
       "  s.async = true;\n",
       "  s.onreadystatechange = s.onload = callback;\n",
       "  s.onerror = function(){console.warn(\"failed to load library \" + url);};\n",
       "  document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
       "}\n",
       "\n",
       "if(typeof(mpld3) !== \"undefined\" && mpld3._mpld3IsLoaded){\n",
       "   // already loaded: just create the figure\n",
       "   !function(mpld3){\n",
       "       \n",
       "    mpld3.register_plugin(\"htmltooltip\", HtmlTooltipPlugin);\n",
       "    HtmlTooltipPlugin.prototype = Object.create(mpld3.Plugin.prototype);\n",
       "    HtmlTooltipPlugin.prototype.constructor = HtmlTooltipPlugin;\n",
       "    HtmlTooltipPlugin.prototype.requiredProps = [\"id\"];\n",
       "    HtmlTooltipPlugin.prototype.defaultProps = {labels:null, hoffset:0, voffset:10};\n",
       "    function HtmlTooltipPlugin(fig, props){\n",
       "        mpld3.Plugin.call(this, fig, props);\n",
       "    };\n",
       "\n",
       "    HtmlTooltipPlugin.prototype.draw = function(){\n",
       "       var obj = mpld3.get_element(this.props.id);\n",
       "       var labels = this.props.labels;\n",
       "       var tooltip = d3.select(\"body\").append(\"div\")\n",
       "                    .attr(\"class\", \"mpld3-tooltip\")\n",
       "                    .style(\"position\", \"absolute\")\n",
       "                    .style(\"z-index\", \"10\")\n",
       "                    .style(\"visibility\", \"hidden\");\n",
       "\n",
       "       obj.elements()\n",
       "           .on(\"mouseover\", function(d, i){\n",
       "                              tooltip.html(labels[i])\n",
       "                                     .style(\"visibility\", \"visible\");})\n",
       "           .on(\"mousemove\", function(d, i){\n",
       "                    tooltip\n",
       "                      .style(\"top\", d3.event.pageY + this.props.voffset + \"px\")\n",
       "                      .style(\"left\",d3.event.pageX + this.props.hoffset + \"px\");\n",
       "                 }.bind(this))\n",
       "           .on(\"mouseout\",  function(d, i){\n",
       "                           tooltip.style(\"visibility\", \"hidden\");});\n",
       "    };\n",
       "    \n",
       "       mpld3.draw_figure(\"fig_el82645032962085170635785\", {\"data\": {\"data01\": [[-0.9059767466709602, 0.1999642075296948], [-0.8089679667072682, -1.787110981705207], [-1.6109597326639027, 0.4675581840015952], [-0.10251781544973887, 0.1311867639929912], [-0.3493084496838704, 1.0824944223722484], [-0.555969690028334, 1.000619138771026], [-0.14798291653111562, -0.3693031336161806], [-1.8847937507295367, -0.40214247827618393], [0.6445143911682311, 0.06373349470193701], [0.43596108947192685, -0.08850265669440752], [0.6336702721810815, 1.0101904912985837], [-0.23575046953852663, -1.5487475660901404], [-1.7615791595047134, 1.1170318985889753], [-0.8019995765290763, -0.8239561791585369], [2.7350641255944415, 1.480492771233579], [1.5485378678894755, 1.4694312931656472], [1.9952125446418467, -0.441021882557023], [-1.4167830218634896, 0.9879846786060962], [-0.41037625893940155, -0.3590210904920519], [3.224026023437597, 0.4154867128857576], [-0.5847995908581678, 0.40784156256132703], [1.6028933221816086, 0.14162369640570466], [-1.0327687353836277, 0.24533682936197643], [2.8261279378091473, -0.3091272951140364], [-1.2041775980807685, -0.7407341756181904], [-0.9253688011631207, 0.12137755086948607], [-1.468202518030362, 0.5723541667342684], [-1.3531731857525442, 1.013337896102464], [-0.32790501885571566, 0.2339733964797726], [-0.37060263246639696, -1.4895779948937453], [-0.321007062576358, 0.016458027098669396], [2.6018509512595576, 0.08770052524740422], [-0.6344609909734393, 0.014002186819305887], [3.1693310984264427, 1.3200092661392027], [0.21638075114195932, -0.3372544929757867], [-1.4910155830462235, 0.38791026819665075], [1.7672172090352345, 0.6674684123094022], [-0.01436607408678462, -0.32062120993651394], [-1.6832380857104026, 0.045494072274930186], [1.4242160994269788, 0.18613381848993954], [-1.5479992174465664, -0.35186848526708253], [0.6185118031429417, 0.3881329647182456], [3.5576573946233587, 1.618504246533931], [0.48267806430719096, -0.29660002707900096], [-0.10948169817348458, -0.29252848369056705], [-1.6018641135261695, 0.6652766031718063], [-1.087385773580763, -0.7460884998993241], [-1.4290529114399957, 2.106432734108494], [1.0102580226459321, -0.5496481218208135], [0.1082802150919815, 0.05117396932611247], [-0.8778733783809136, -0.7860608767005898], [-1.5993336920258745, -0.17812566137913655], [-1.8809432732275373, -0.573438610080969], [-1.1836684947598555, -1.6317193911091208], [-0.05912613750871809, 0.38470467693386207], [-1.8599736067038959, 1.341981003640925], [-0.5108893313177494, 0.02730059441776359], [-1.30232850485144, -0.9298335886775063], [0.7774206520404754, -1.2793191562085755], [-0.7546508720799441, -0.3486673277038578], [-1.7011572309219847, 1.1760921694501814], [-1.0657736095951564, -2.021045683831008], [2.040249746197785, 1.8441135679401288], [2.9588104940861646, 0.13760742390873298], [-0.48820832440200607, 0.007299793112342251], [2.3447903497981937, -1.8007730240583963], [0.6902464549995853, -1.4294754388400974], [-0.763720655546284, -0.3434016965683857], [-0.8312660658526394, 0.35939269006814734], [0.06179183671813215, -1.4142915463527859], [0.42059418364209855, -0.8821985075910894], [0.026425664311242214, -0.9933442071942615], [3.5339610746191297, 0.9609343271363076], [0.06707709061265654, -2.1332257673145336], [-0.38483690554117433, -0.08590617959525451], [2.6694851258920433, -0.4799400659592541], [2.123344083929215, -0.9979830914802461], [-1.720393257474338, 1.2034158014897778], [1.535785779717043, -0.8857257240521148], [-0.7339030094818977, 0.2313970092009282], [0.17121271236485622, 1.34607721468068], [-0.1848454044463215, 1.0607314140824462], [-0.16625426409826466, -0.10334522734395346], [-1.3851432939673707, 2.3941243544829542], [-0.8112056964538618, -0.4736151850712644], [1.485881741345664, 0.23359001643653923], [-0.9529643874192013, -0.34374950732154425], [-1.6167340799767518, 0.04531591711333752], [-0.375174757102701, -0.3193580736309151], [-1.5746034832821052, -1.359470193968545], [1.3785585697672271, -1.565373273697286], [-1.5326823337877835, 0.7906825890899495], [-1.7303740795253182, 1.2390315784853807], [-0.6166806966265016, -1.1426037508269171], [0.8237042809253377, -0.9490890758887148], [2.334038512969227, 1.1550938084151021], [3.1891705539186463, 0.07634042732937146], [-0.761204766269025, -1.0851154903272002], [0.03568773175507887, -0.5141781335128335], [-1.389286997232564, 2.130024402856699], [-1.473911567378929, 0.9356001277058837], [-0.8661874467107501, 0.06405594133200992], [0.5649285476342275, -0.3939841324436752], [-1.027241655164723, 0.3275791204015193], [2.5268220343834447, -0.4909898761933704]]}, \"plugins\": [{\"type\": \"reset\"}, {\"type\": \"zoom\", \"button\": true, \"enabled\": false}, {\"type\": \"boxzoom\", \"button\": true, \"enabled\": false}, {\"type\": \"htmltooltip\", \"labels\": [\"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Argentina</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.905977</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.199964</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.465967</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.195230</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Armenia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.808968</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.787111</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.749288</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>39.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.057700</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Austria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.610960</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.467558</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.656714</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.490170</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Azerbaijan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.102518</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.131187</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.264091</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.025970</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bahamas</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.349308</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.082494</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.133238</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>66.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.926070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Barbados</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.555970</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.000619</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.666502</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>66.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.923830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Belarus</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.147983</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.369303</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.377423</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>53.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>3.137070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Belgium</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.884794</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.402142</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.615086</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>48.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.924920</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Belize</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.644514</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.063733</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.031668</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>13.471580</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bhutan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.435961</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.088503</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.462929</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>66.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.083280</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bolivia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.633670</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.010190</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.476296</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>70.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.028940</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Botswana</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.235750</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.548748</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.465450</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-6.489470</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Brunei</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.761579</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.117032</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.247214</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.287260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bulgaria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.802000</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.823956</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.498872</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.148880</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Burkina Faso</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.735064</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.480493</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.175434</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>81.199997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.247250</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cambodia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.548538</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.469431</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.654714</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>78.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>65.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.917590</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cameroon</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.995213</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.441022</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.844318</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>55.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>8.575600</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Canada</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.416783</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.987985</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.626688</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.470170</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cape Verde</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.410376</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.359021</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.523691</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-5.997490</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Chad</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.224026</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.415487</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.537504</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>69.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>53.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>19.968190</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Colombia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.584800</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.407842</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.173344</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>61.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.843070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Congo, Rep.</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.602893</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.141624</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.515644</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.300003</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.049310</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Costa Rica</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.032769</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.245337</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.327070</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.189040</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cote d'Ivoire</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.826128</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.309127</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.941214</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>54.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>21.766570</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Croatia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.204178</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.740734</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.948518</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>46.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.441510</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cuba</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.925369</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.121378</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.665211</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.338830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cyprus</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.468203</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.572354</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.361799</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.518550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Denmark</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.353173</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.013338</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.712460</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>63.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>78.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.588940</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ecuador</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.327905</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.233973</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.070115</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.297940</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Egypt</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.370603</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.489578</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.104431</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>41.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>3.662670</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>El Salvador</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.321007</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.016458</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.872884</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.456470</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Eritrea</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.601851</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.087701</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.140908</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>59.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>14.484550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Estonia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.634461</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.014002</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.074280</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>4.037860</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ethiopia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.169331</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.320009</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.678998</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>80.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>10.905990</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Fiji</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.216381</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.337254</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.914158</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>65.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.142810</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Finland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.491016</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.387910</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.601156</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.201330</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Gambia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.767217</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.667468</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.307449</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>71.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>61.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.999020</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Georgia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.014366</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.320621</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.536519</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.169760</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Germany</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.683238</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.045494</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.588125</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.540070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ghana</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.424216</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.186134</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.869484</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>61.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>3.343550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Greece</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.547999</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.351868</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.352177</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>49.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.031410</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Guatemala</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.618512</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.388133</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.802684</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.821130</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Guinea</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.557657</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.618504</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.087835</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>81.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>20.020490</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Guyana</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.482678</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.296600</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.466218</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.212650</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Honduras</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.109482</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.292528</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.330057</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-5.312550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Hong Kong, China</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.601864</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.665277</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.679305</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>82.124000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.604640</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Hungary</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.087386</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.746088</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.054576</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.506110</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Iceland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.429053</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>2.106433</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.663720</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>74.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>81.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.146030</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>India</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.010258</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.549648</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.176560</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>5.413260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Indonesia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.108280</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.051174</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.822841</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.202970</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Iran</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.877873</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.786061</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.517515</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.031680</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Israel</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.599334</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.178126</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.187703</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.121540</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Italy</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.880943</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.573439</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.551021</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>81.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.936880</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Jordan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.183668</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.631719</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.192079</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>39.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.360130</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Kazakhstan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.059126</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.384705</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.742618</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>65.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.182900</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Kuwait</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.859974</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.341981</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.530443</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.699997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>78.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.676570</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Latvia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.510889</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.027301</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.900620</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.947920</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Lebanon</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.302329</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.929834</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.386067</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>46.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.131060</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Lesotho</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.777421</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.279319</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.554775</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>44.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-26.551050</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Lithuania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.754651</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.348667</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.866371</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>71.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.384950</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Macao, China</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.701157</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.176092</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.178775</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.199997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>78.886000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.264630</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Macedonia, FYR</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.065774</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-2.021046</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.189139</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>34.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.805990</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Madagascar</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.040250</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.844114</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.245745</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>84.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>62.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.262260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Malawi</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.958810</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.137607</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.403049</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>71.699997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>49.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.115830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Maldives</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.488208</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.007300</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.089959</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.891300</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mali</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.344790</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.800773</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.349243</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>54.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>16.086970</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mauritania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.690246</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.429475</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.909027</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.131300</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mauritius</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.763721</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.343402</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.475324</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>53.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.102180</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mexico</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.831266</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.359393</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.677542</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.271540</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Moldova</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.061792</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.414292</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.152535</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.747480</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mongolia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.420594</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.882199</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.672546</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>62.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.243210</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Morocco</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.026426</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.993344</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.652641</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>46.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.208860</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mozambique</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.533961</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.960934</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.646684</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>77.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>54.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>14.250090</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Namibia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.067077</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-2.133226</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.958044</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>41.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>53.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-8.595690</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Nicaragua</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.384837</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.085906</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.249098</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-6.316270</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Niger</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.669485</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.479940</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.674577</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>58.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>12.607190</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Nigeria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.123344</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.997983</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.347697</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>20.773300</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Norway</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.720393</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.203416</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.067251</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.199997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.861550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Pakistan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.535786</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.885726</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.298761</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>50.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>17.711700</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Panama</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.733903</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.231397</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.361851</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.122110</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Paraguay</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.171213</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.346077</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.739353</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>72.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.173960</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Peru</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.184845</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.060731</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.996081</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>67.800003</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.550440</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Philippines</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.166254</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.103345</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.503630</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-7.342690</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Qatar</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.385143</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>2.394124</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.750505</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>75.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.853210</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Romania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.811206</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.473615</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.679253</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.049250</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Senegal</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.485882</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.233590</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.639393</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>4.013130</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Slovak Republic</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.952964</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.343750</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.977585</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.510610</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Spain</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.616734</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.045316</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.431980</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.412720</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Sri Lanka</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.375175</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.319358</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.724435</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>54.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.734160</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Suriname</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.574603</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.359470</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.470836</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>68.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-16.395260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Swaziland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.378559</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.565373</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.621692</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>44.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.912150</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Sweden</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.532682</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.790683</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.656070</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.047830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Switzerland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.730374</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.239032</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.876402</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.699997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>81.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.662370</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Syria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.616681</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.142604</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.680752</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>44.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.277510</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Tajikistan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.823704</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.949089</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.477146</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>4.506510</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Tanzania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.334039</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.155094</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.191813</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>78.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.362050</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Togo</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.189171</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.076340</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.097305</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>59.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>24.349910</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Turkey</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.761205</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.085115</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.686151</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>43.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>5.927800</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ukraine</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.035688</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.514178</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.972057</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>53.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>67.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.691150</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>United Arab Emirates</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.389287</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>2.130024</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.476259</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>75.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.332680</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>United States</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.473912</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.935600</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.832812</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.771030</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Uruguay</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.866187</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.064056</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.497198</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.448210</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Uzbekistan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.564929</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.393984</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.059925</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>68.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.730380</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Venezuela</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.027242</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.327579</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.689949</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-5.746690</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Zambia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.526822</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.490990</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.002437</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>48.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>10.728000</td>\\n    </tr>\\n  </tbody>\\n</table>\"], \"voffset\": 10, \"hoffset\": 10, \"id\": \"el8264508623984pts\"}], \"width\": 800.0, \"axes\": [{\"ydomain\": [-3.0, 3.0], \"axes\": [{\"nticks\": 7, \"position\": \"bottom\", \"scale\": \"linear\", \"fontsize\": 13.0, \"tickformat\": null, \"tickvalues\": null, \"grid\": {\"color\": \"#FFFFFF\", \"alpha\": 1.0, \"dasharray\": \"10,0\", \"gridOn\": true}}, {\"nticks\": 7, \"position\": \"left\", \"scale\": \"linear\", \"fontsize\": 13.0, \"tickformat\": null, \"tickvalues\": null, \"grid\": {\"color\": \"#FFFFFF\", \"alpha\": 1.0, \"dasharray\": \"10,0\", \"gridOn\": true}}], \"sharey\": [], \"zoomable\": true, \"xscale\": \"linear\", \"sharex\": [], \"xlim\": [-2.0, 4.0], \"markers\": [{\"data\": \"data01\", \"edgecolor\": \"#000000\", \"markerpath\": [[[0.0, 7.5], [1.9890232500000002, 7.5], [3.896849030886401, 6.709752686911813], [5.303300858899107, 5.303300858899107], [6.709752686911813, 3.896849030886401], [7.5, 1.9890232500000002], [7.5, 0.0], [7.5, -1.9890232500000002], [6.709752686911813, -3.896849030886401], [5.303300858899107, -5.303300858899107], [3.896849030886401, -6.709752686911813], [1.9890232500000002, -7.5], [0.0, -7.5], [-1.9890232500000002, -7.5], [-3.896849030886401, -6.709752686911813], [-5.303300858899107, -5.303300858899107], [-6.709752686911813, -3.896849030886401], [-7.5, -1.9890232500000002], [-7.5, 0.0], [-7.5, 1.9890232500000002], [-6.709752686911813, 3.896849030886401], [-5.303300858899107, 5.303300858899107], [-3.896849030886401, 6.709752686911813], [-1.9890232500000002, 7.5], [0.0, 7.5]], [\"M\", \"C\", \"C\", \"C\", \"C\", \"C\", \"C\", \"C\", \"C\", \"Z\"]], \"coordinates\": \"data\", \"xindex\": 0, \"edgewidth\": 1, \"zorder\": 2, \"yindex\": 1, \"alpha\": 0.6, \"id\": \"el8264508623984pts\", \"facecolor\": \"#0000FF\"}], \"bbox\": [0.125, 0.125, 0.775, 0.775], \"id\": \"el8264503983496\", \"lines\": [], \"texts\": [{\"color\": \"#262626\", \"coordinates\": \"axes\", \"v_baseline\": \"hanging\", \"rotation\": -0.0, \"h_anchor\": \"middle\", \"position\": [0.5, -0.0734767025089606], \"text\": \"PC1\", \"fontsize\": 14.3, \"zorder\": 3, \"alpha\": 1, \"id\": \"el8264504070856\"}, {\"color\": \"#262626\", \"coordinates\": \"axes\", \"v_baseline\": \"auto\", \"rotation\": -90.0, \"h_anchor\": \"middle\", \"position\": [-0.04831989247311827, 0.5], \"text\": \"PC2\", \"fontsize\": 14.3, \"zorder\": 3, \"alpha\": 1, \"id\": \"el8264504026080\"}, {\"color\": \"#262626\", \"coordinates\": \"axes\", \"v_baseline\": \"auto\", \"rotation\": -0.0, \"h_anchor\": \"middle\", \"position\": [0.5, 1.0149342891278375], \"text\": \"Interactive projection on PC plane\", \"fontsize\": 20.0, \"zorder\": 3, \"alpha\": 1, \"id\": \"el8264504134096\"}], \"axesbgalpha\": null, \"xdomain\": [-2.0, 4.0], \"ylim\": [-3.0, 3.0], \"images\": [], \"collections\": [], \"yscale\": \"linear\", \"paths\": [], \"axesbg\": \"#EAEAF2\"}], \"id\": \"el8264503296208\", \"height\": 480.0});\n",
       "   }(mpld3);\n",
       "}else if(typeof define === \"function\" && define.amd){\n",
       "   // require.js is available: use it to load d3/mpld3\n",
       "   require.config({paths: {d3: \"https://mpld3.github.io/js/d3.v3.min\"}});\n",
       "   require([\"d3\"], function(d3){\n",
       "      window.d3 = d3;\n",
       "      mpld3_load_lib(\"https://mpld3.github.io/js/mpld3.v0.2.js\", function(){\n",
       "         \n",
       "    mpld3.register_plugin(\"htmltooltip\", HtmlTooltipPlugin);\n",
       "    HtmlTooltipPlugin.prototype = Object.create(mpld3.Plugin.prototype);\n",
       "    HtmlTooltipPlugin.prototype.constructor = HtmlTooltipPlugin;\n",
       "    HtmlTooltipPlugin.prototype.requiredProps = [\"id\"];\n",
       "    HtmlTooltipPlugin.prototype.defaultProps = {labels:null, hoffset:0, voffset:10};\n",
       "    function HtmlTooltipPlugin(fig, props){\n",
       "        mpld3.Plugin.call(this, fig, props);\n",
       "    };\n",
       "\n",
       "    HtmlTooltipPlugin.prototype.draw = function(){\n",
       "       var obj = mpld3.get_element(this.props.id);\n",
       "       var labels = this.props.labels;\n",
       "       var tooltip = d3.select(\"body\").append(\"div\")\n",
       "                    .attr(\"class\", \"mpld3-tooltip\")\n",
       "                    .style(\"position\", \"absolute\")\n",
       "                    .style(\"z-index\", \"10\")\n",
       "                    .style(\"visibility\", \"hidden\");\n",
       "\n",
       "       obj.elements()\n",
       "           .on(\"mouseover\", function(d, i){\n",
       "                              tooltip.html(labels[i])\n",
       "                                     .style(\"visibility\", \"visible\");})\n",
       "           .on(\"mousemove\", function(d, i){\n",
       "                    tooltip\n",
       "                      .style(\"top\", d3.event.pageY + this.props.voffset + \"px\")\n",
       "                      .style(\"left\",d3.event.pageX + this.props.hoffset + \"px\");\n",
       "                 }.bind(this))\n",
       "           .on(\"mouseout\",  function(d, i){\n",
       "                           tooltip.style(\"visibility\", \"hidden\");});\n",
       "    };\n",
       "    \n",
       "         mpld3.draw_figure(\"fig_el82645032962085170635785\", {\"data\": {\"data01\": [[-0.9059767466709602, 0.1999642075296948], [-0.8089679667072682, -1.787110981705207], [-1.6109597326639027, 0.4675581840015952], [-0.10251781544973887, 0.1311867639929912], [-0.3493084496838704, 1.0824944223722484], [-0.555969690028334, 1.000619138771026], [-0.14798291653111562, -0.3693031336161806], [-1.8847937507295367, -0.40214247827618393], [0.6445143911682311, 0.06373349470193701], [0.43596108947192685, -0.08850265669440752], [0.6336702721810815, 1.0101904912985837], [-0.23575046953852663, -1.5487475660901404], [-1.7615791595047134, 1.1170318985889753], [-0.8019995765290763, -0.8239561791585369], [2.7350641255944415, 1.480492771233579], [1.5485378678894755, 1.4694312931656472], [1.9952125446418467, -0.441021882557023], [-1.4167830218634896, 0.9879846786060962], [-0.41037625893940155, -0.3590210904920519], [3.224026023437597, 0.4154867128857576], [-0.5847995908581678, 0.40784156256132703], [1.6028933221816086, 0.14162369640570466], [-1.0327687353836277, 0.24533682936197643], [2.8261279378091473, -0.3091272951140364], [-1.2041775980807685, -0.7407341756181904], [-0.9253688011631207, 0.12137755086948607], [-1.468202518030362, 0.5723541667342684], [-1.3531731857525442, 1.013337896102464], [-0.32790501885571566, 0.2339733964797726], [-0.37060263246639696, -1.4895779948937453], [-0.321007062576358, 0.016458027098669396], [2.6018509512595576, 0.08770052524740422], [-0.6344609909734393, 0.014002186819305887], [3.1693310984264427, 1.3200092661392027], [0.21638075114195932, -0.3372544929757867], [-1.4910155830462235, 0.38791026819665075], [1.7672172090352345, 0.6674684123094022], [-0.01436607408678462, -0.32062120993651394], [-1.6832380857104026, 0.045494072274930186], [1.4242160994269788, 0.18613381848993954], [-1.5479992174465664, -0.35186848526708253], [0.6185118031429417, 0.3881329647182456], [3.5576573946233587, 1.618504246533931], [0.48267806430719096, -0.29660002707900096], [-0.10948169817348458, -0.29252848369056705], [-1.6018641135261695, 0.6652766031718063], [-1.087385773580763, -0.7460884998993241], [-1.4290529114399957, 2.106432734108494], [1.0102580226459321, -0.5496481218208135], [0.1082802150919815, 0.05117396932611247], [-0.8778733783809136, -0.7860608767005898], [-1.5993336920258745, -0.17812566137913655], [-1.8809432732275373, -0.573438610080969], [-1.1836684947598555, -1.6317193911091208], [-0.05912613750871809, 0.38470467693386207], [-1.8599736067038959, 1.341981003640925], [-0.5108893313177494, 0.02730059441776359], [-1.30232850485144, -0.9298335886775063], [0.7774206520404754, -1.2793191562085755], [-0.7546508720799441, -0.3486673277038578], [-1.7011572309219847, 1.1760921694501814], [-1.0657736095951564, -2.021045683831008], [2.040249746197785, 1.8441135679401288], [2.9588104940861646, 0.13760742390873298], [-0.48820832440200607, 0.007299793112342251], [2.3447903497981937, -1.8007730240583963], [0.6902464549995853, -1.4294754388400974], [-0.763720655546284, -0.3434016965683857], [-0.8312660658526394, 0.35939269006814734], [0.06179183671813215, -1.4142915463527859], [0.42059418364209855, -0.8821985075910894], [0.026425664311242214, -0.9933442071942615], [3.5339610746191297, 0.9609343271363076], [0.06707709061265654, -2.1332257673145336], [-0.38483690554117433, -0.08590617959525451], [2.6694851258920433, -0.4799400659592541], [2.123344083929215, -0.9979830914802461], [-1.720393257474338, 1.2034158014897778], [1.535785779717043, -0.8857257240521148], [-0.7339030094818977, 0.2313970092009282], [0.17121271236485622, 1.34607721468068], [-0.1848454044463215, 1.0607314140824462], [-0.16625426409826466, -0.10334522734395346], [-1.3851432939673707, 2.3941243544829542], [-0.8112056964538618, -0.4736151850712644], [1.485881741345664, 0.23359001643653923], [-0.9529643874192013, -0.34374950732154425], [-1.6167340799767518, 0.04531591711333752], [-0.375174757102701, -0.3193580736309151], [-1.5746034832821052, -1.359470193968545], [1.3785585697672271, -1.565373273697286], [-1.5326823337877835, 0.7906825890899495], [-1.7303740795253182, 1.2390315784853807], [-0.6166806966265016, -1.1426037508269171], [0.8237042809253377, -0.9490890758887148], [2.334038512969227, 1.1550938084151021], [3.1891705539186463, 0.07634042732937146], [-0.761204766269025, -1.0851154903272002], [0.03568773175507887, -0.5141781335128335], [-1.389286997232564, 2.130024402856699], [-1.473911567378929, 0.9356001277058837], [-0.8661874467107501, 0.06405594133200992], [0.5649285476342275, -0.3939841324436752], [-1.027241655164723, 0.3275791204015193], [2.5268220343834447, -0.4909898761933704]]}, \"plugins\": [{\"type\": \"reset\"}, {\"type\": \"zoom\", \"button\": true, \"enabled\": false}, {\"type\": \"boxzoom\", \"button\": true, \"enabled\": false}, {\"type\": \"htmltooltip\", \"labels\": [\"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Argentina</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.905977</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.199964</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.465967</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.195230</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Armenia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.808968</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.787111</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.749288</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>39.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.057700</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Austria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.610960</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.467558</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.656714</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.490170</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Azerbaijan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.102518</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.131187</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.264091</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.025970</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bahamas</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.349308</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.082494</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.133238</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>66.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.926070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Barbados</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.555970</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.000619</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.666502</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>66.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.923830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Belarus</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.147983</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.369303</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.377423</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>53.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>3.137070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Belgium</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.884794</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.402142</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.615086</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>48.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.924920</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Belize</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.644514</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.063733</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.031668</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>13.471580</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bhutan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.435961</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.088503</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.462929</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>66.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.083280</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bolivia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.633670</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.010190</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.476296</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>70.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.028940</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Botswana</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.235750</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.548748</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.465450</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-6.489470</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Brunei</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.761579</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.117032</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.247214</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.287260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bulgaria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.802000</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.823956</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.498872</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.148880</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Burkina Faso</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.735064</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.480493</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.175434</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>81.199997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.247250</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cambodia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.548538</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.469431</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.654714</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>78.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>65.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.917590</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cameroon</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.995213</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.441022</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.844318</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>55.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>8.575600</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Canada</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.416783</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.987985</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.626688</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.470170</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cape Verde</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.410376</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.359021</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.523691</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-5.997490</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Chad</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.224026</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.415487</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.537504</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>69.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>53.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>19.968190</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Colombia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.584800</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.407842</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.173344</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>61.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.843070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Congo, Rep.</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.602893</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.141624</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.515644</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.300003</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.049310</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Costa Rica</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.032769</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.245337</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.327070</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.189040</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cote d'Ivoire</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.826128</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.309127</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.941214</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>54.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>21.766570</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Croatia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.204178</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.740734</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.948518</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>46.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.441510</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cuba</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.925369</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.121378</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.665211</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.338830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cyprus</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.468203</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.572354</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.361799</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.518550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Denmark</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.353173</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.013338</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.712460</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>63.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>78.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.588940</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ecuador</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.327905</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.233973</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.070115</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.297940</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Egypt</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.370603</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.489578</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.104431</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>41.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>3.662670</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>El Salvador</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.321007</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.016458</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.872884</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.456470</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Eritrea</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.601851</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.087701</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.140908</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>59.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>14.484550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Estonia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.634461</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.014002</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.074280</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>4.037860</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ethiopia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.169331</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.320009</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.678998</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>80.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>10.905990</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Fiji</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.216381</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.337254</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.914158</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>65.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.142810</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Finland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.491016</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.387910</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.601156</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.201330</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Gambia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.767217</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.667468</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.307449</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>71.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>61.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.999020</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Georgia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.014366</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.320621</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.536519</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.169760</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Germany</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.683238</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.045494</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.588125</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.540070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ghana</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.424216</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.186134</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.869484</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>61.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>3.343550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Greece</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.547999</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.351868</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.352177</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>49.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.031410</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Guatemala</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.618512</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.388133</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.802684</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.821130</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Guinea</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.557657</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.618504</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.087835</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>81.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>20.020490</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Guyana</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.482678</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.296600</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.466218</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.212650</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Honduras</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.109482</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.292528</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.330057</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-5.312550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Hong Kong, China</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.601864</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.665277</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.679305</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>82.124000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.604640</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Hungary</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.087386</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.746088</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.054576</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.506110</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Iceland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.429053</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>2.106433</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.663720</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>74.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>81.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.146030</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>India</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.010258</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.549648</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.176560</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>5.413260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Indonesia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.108280</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.051174</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.822841</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.202970</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Iran</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.877873</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.786061</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.517515</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.031680</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Israel</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.599334</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.178126</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.187703</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.121540</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Italy</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.880943</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.573439</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.551021</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>81.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.936880</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Jordan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.183668</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.631719</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.192079</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>39.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.360130</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Kazakhstan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.059126</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.384705</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.742618</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>65.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.182900</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Kuwait</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.859974</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.341981</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.530443</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.699997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>78.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.676570</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Latvia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.510889</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.027301</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.900620</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.947920</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Lebanon</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.302329</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.929834</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.386067</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>46.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.131060</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Lesotho</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.777421</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.279319</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.554775</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>44.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-26.551050</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Lithuania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.754651</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.348667</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.866371</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>71.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.384950</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Macao, China</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.701157</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.176092</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.178775</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.199997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>78.886000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.264630</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Macedonia, FYR</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.065774</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-2.021046</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.189139</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>34.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.805990</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Madagascar</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.040250</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.844114</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.245745</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>84.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>62.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.262260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Malawi</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.958810</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.137607</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.403049</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>71.699997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>49.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.115830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Maldives</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.488208</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.007300</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.089959</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.891300</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mali</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.344790</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.800773</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.349243</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>54.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>16.086970</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mauritania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.690246</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.429475</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.909027</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.131300</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mauritius</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.763721</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.343402</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.475324</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>53.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.102180</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mexico</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.831266</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.359393</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.677542</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.271540</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Moldova</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.061792</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.414292</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.152535</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.747480</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mongolia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.420594</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.882199</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.672546</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>62.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.243210</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Morocco</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.026426</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.993344</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.652641</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>46.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.208860</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mozambique</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.533961</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.960934</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.646684</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>77.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>54.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>14.250090</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Namibia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.067077</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-2.133226</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.958044</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>41.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>53.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-8.595690</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Nicaragua</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.384837</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.085906</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.249098</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-6.316270</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Niger</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.669485</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.479940</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.674577</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>58.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>12.607190</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Nigeria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.123344</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.997983</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.347697</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>20.773300</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Norway</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.720393</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.203416</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.067251</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.199997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.861550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Pakistan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.535786</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.885726</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.298761</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>50.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>17.711700</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Panama</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.733903</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.231397</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.361851</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.122110</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Paraguay</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.171213</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.346077</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.739353</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>72.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.173960</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Peru</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.184845</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.060731</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.996081</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>67.800003</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.550440</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Philippines</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.166254</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.103345</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.503630</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-7.342690</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Qatar</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.385143</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>2.394124</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.750505</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>75.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.853210</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Romania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.811206</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.473615</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.679253</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.049250</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Senegal</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.485882</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.233590</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.639393</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>4.013130</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Slovak Republic</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.952964</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.343750</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.977585</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.510610</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Spain</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.616734</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.045316</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.431980</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.412720</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Sri Lanka</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.375175</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.319358</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.724435</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>54.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.734160</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Suriname</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.574603</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.359470</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.470836</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>68.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-16.395260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Swaziland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.378559</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.565373</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.621692</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>44.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.912150</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Sweden</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.532682</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.790683</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.656070</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.047830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Switzerland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.730374</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.239032</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.876402</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.699997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>81.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.662370</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Syria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.616681</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.142604</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.680752</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>44.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.277510</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Tajikistan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.823704</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.949089</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.477146</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>4.506510</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Tanzania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.334039</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.155094</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.191813</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>78.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.362050</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Togo</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.189171</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.076340</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.097305</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>59.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>24.349910</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Turkey</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.761205</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.085115</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.686151</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>43.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>5.927800</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ukraine</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.035688</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.514178</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.972057</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>53.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>67.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.691150</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>United Arab Emirates</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.389287</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>2.130024</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.476259</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>75.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.332680</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>United States</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.473912</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.935600</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.832812</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.771030</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Uruguay</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.866187</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.064056</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.497198</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.448210</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Uzbekistan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.564929</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.393984</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.059925</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>68.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.730380</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Venezuela</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.027242</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.327579</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.689949</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-5.746690</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Zambia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.526822</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.490990</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.002437</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>48.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>10.728000</td>\\n    </tr>\\n  </tbody>\\n</table>\"], \"voffset\": 10, \"hoffset\": 10, \"id\": \"el8264508623984pts\"}], \"width\": 800.0, \"axes\": [{\"ydomain\": [-3.0, 3.0], \"axes\": [{\"nticks\": 7, \"position\": \"bottom\", \"scale\": \"linear\", \"fontsize\": 13.0, \"tickformat\": null, \"tickvalues\": null, \"grid\": {\"color\": \"#FFFFFF\", \"alpha\": 1.0, \"dasharray\": \"10,0\", \"gridOn\": true}}, {\"nticks\": 7, \"position\": \"left\", \"scale\": \"linear\", \"fontsize\": 13.0, \"tickformat\": null, \"tickvalues\": null, \"grid\": {\"color\": \"#FFFFFF\", \"alpha\": 1.0, \"dasharray\": \"10,0\", \"gridOn\": true}}], \"sharey\": [], \"zoomable\": true, \"xscale\": \"linear\", \"sharex\": [], \"xlim\": [-2.0, 4.0], \"markers\": [{\"data\": \"data01\", \"edgecolor\": \"#000000\", \"markerpath\": [[[0.0, 7.5], [1.9890232500000002, 7.5], [3.896849030886401, 6.709752686911813], [5.303300858899107, 5.303300858899107], [6.709752686911813, 3.896849030886401], [7.5, 1.9890232500000002], [7.5, 0.0], [7.5, -1.9890232500000002], [6.709752686911813, -3.896849030886401], [5.303300858899107, -5.303300858899107], [3.896849030886401, -6.709752686911813], [1.9890232500000002, -7.5], [0.0, -7.5], [-1.9890232500000002, -7.5], [-3.896849030886401, -6.709752686911813], [-5.303300858899107, -5.303300858899107], [-6.709752686911813, -3.896849030886401], [-7.5, -1.9890232500000002], [-7.5, 0.0], [-7.5, 1.9890232500000002], [-6.709752686911813, 3.896849030886401], [-5.303300858899107, 5.303300858899107], [-3.896849030886401, 6.709752686911813], [-1.9890232500000002, 7.5], [0.0, 7.5]], [\"M\", \"C\", \"C\", \"C\", \"C\", \"C\", \"C\", \"C\", \"C\", \"Z\"]], \"coordinates\": \"data\", \"xindex\": 0, \"edgewidth\": 1, \"zorder\": 2, \"yindex\": 1, \"alpha\": 0.6, \"id\": \"el8264508623984pts\", \"facecolor\": \"#0000FF\"}], \"bbox\": [0.125, 0.125, 0.775, 0.775], \"id\": \"el8264503983496\", \"lines\": [], \"texts\": [{\"color\": \"#262626\", \"coordinates\": \"axes\", \"v_baseline\": \"hanging\", \"rotation\": -0.0, \"h_anchor\": \"middle\", \"position\": [0.5, -0.0734767025089606], \"text\": \"PC1\", \"fontsize\": 14.3, \"zorder\": 3, \"alpha\": 1, \"id\": \"el8264504070856\"}, {\"color\": \"#262626\", \"coordinates\": \"axes\", \"v_baseline\": \"auto\", \"rotation\": -90.0, \"h_anchor\": \"middle\", \"position\": [-0.04831989247311827, 0.5], \"text\": \"PC2\", \"fontsize\": 14.3, \"zorder\": 3, \"alpha\": 1, \"id\": \"el8264504026080\"}, {\"color\": \"#262626\", \"coordinates\": \"axes\", \"v_baseline\": \"auto\", \"rotation\": -0.0, \"h_anchor\": \"middle\", \"position\": [0.5, 1.0149342891278375], \"text\": \"Interactive projection on PC plane\", \"fontsize\": 20.0, \"zorder\": 3, \"alpha\": 1, \"id\": \"el8264504134096\"}], \"axesbgalpha\": null, \"xdomain\": [-2.0, 4.0], \"ylim\": [-3.0, 3.0], \"images\": [], \"collections\": [], \"yscale\": \"linear\", \"paths\": [], \"axesbg\": \"#EAEAF2\"}], \"id\": \"el8264503296208\", \"height\": 480.0});\n",
       "      });\n",
       "    });\n",
       "}else{\n",
       "    // require.js not available: dynamically load d3 & mpld3\n",
       "    mpld3_load_lib(\"https://mpld3.github.io/js/d3.v3.min.js\", function(){\n",
       "         mpld3_load_lib(\"https://mpld3.github.io/js/mpld3.v0.2.js\", function(){\n",
       "                 \n",
       "    mpld3.register_plugin(\"htmltooltip\", HtmlTooltipPlugin);\n",
       "    HtmlTooltipPlugin.prototype = Object.create(mpld3.Plugin.prototype);\n",
       "    HtmlTooltipPlugin.prototype.constructor = HtmlTooltipPlugin;\n",
       "    HtmlTooltipPlugin.prototype.requiredProps = [\"id\"];\n",
       "    HtmlTooltipPlugin.prototype.defaultProps = {labels:null, hoffset:0, voffset:10};\n",
       "    function HtmlTooltipPlugin(fig, props){\n",
       "        mpld3.Plugin.call(this, fig, props);\n",
       "    };\n",
       "\n",
       "    HtmlTooltipPlugin.prototype.draw = function(){\n",
       "       var obj = mpld3.get_element(this.props.id);\n",
       "       var labels = this.props.labels;\n",
       "       var tooltip = d3.select(\"body\").append(\"div\")\n",
       "                    .attr(\"class\", \"mpld3-tooltip\")\n",
       "                    .style(\"position\", \"absolute\")\n",
       "                    .style(\"z-index\", \"10\")\n",
       "                    .style(\"visibility\", \"hidden\");\n",
       "\n",
       "       obj.elements()\n",
       "           .on(\"mouseover\", function(d, i){\n",
       "                              tooltip.html(labels[i])\n",
       "                                     .style(\"visibility\", \"visible\");})\n",
       "           .on(\"mousemove\", function(d, i){\n",
       "                    tooltip\n",
       "                      .style(\"top\", d3.event.pageY + this.props.voffset + \"px\")\n",
       "                      .style(\"left\",d3.event.pageX + this.props.hoffset + \"px\");\n",
       "                 }.bind(this))\n",
       "           .on(\"mouseout\",  function(d, i){\n",
       "                           tooltip.style(\"visibility\", \"hidden\");});\n",
       "    };\n",
       "    \n",
       "                 mpld3.draw_figure(\"fig_el82645032962085170635785\", {\"data\": {\"data01\": [[-0.9059767466709602, 0.1999642075296948], [-0.8089679667072682, -1.787110981705207], [-1.6109597326639027, 0.4675581840015952], [-0.10251781544973887, 0.1311867639929912], [-0.3493084496838704, 1.0824944223722484], [-0.555969690028334, 1.000619138771026], [-0.14798291653111562, -0.3693031336161806], [-1.8847937507295367, -0.40214247827618393], [0.6445143911682311, 0.06373349470193701], [0.43596108947192685, -0.08850265669440752], [0.6336702721810815, 1.0101904912985837], [-0.23575046953852663, -1.5487475660901404], [-1.7615791595047134, 1.1170318985889753], [-0.8019995765290763, -0.8239561791585369], [2.7350641255944415, 1.480492771233579], [1.5485378678894755, 1.4694312931656472], [1.9952125446418467, -0.441021882557023], [-1.4167830218634896, 0.9879846786060962], [-0.41037625893940155, -0.3590210904920519], [3.224026023437597, 0.4154867128857576], [-0.5847995908581678, 0.40784156256132703], [1.6028933221816086, 0.14162369640570466], [-1.0327687353836277, 0.24533682936197643], [2.8261279378091473, -0.3091272951140364], [-1.2041775980807685, -0.7407341756181904], [-0.9253688011631207, 0.12137755086948607], [-1.468202518030362, 0.5723541667342684], [-1.3531731857525442, 1.013337896102464], [-0.32790501885571566, 0.2339733964797726], [-0.37060263246639696, -1.4895779948937453], [-0.321007062576358, 0.016458027098669396], [2.6018509512595576, 0.08770052524740422], [-0.6344609909734393, 0.014002186819305887], [3.1693310984264427, 1.3200092661392027], [0.21638075114195932, -0.3372544929757867], [-1.4910155830462235, 0.38791026819665075], [1.7672172090352345, 0.6674684123094022], [-0.01436607408678462, -0.32062120993651394], [-1.6832380857104026, 0.045494072274930186], [1.4242160994269788, 0.18613381848993954], [-1.5479992174465664, -0.35186848526708253], [0.6185118031429417, 0.3881329647182456], [3.5576573946233587, 1.618504246533931], [0.48267806430719096, -0.29660002707900096], [-0.10948169817348458, -0.29252848369056705], [-1.6018641135261695, 0.6652766031718063], [-1.087385773580763, -0.7460884998993241], [-1.4290529114399957, 2.106432734108494], [1.0102580226459321, -0.5496481218208135], [0.1082802150919815, 0.05117396932611247], [-0.8778733783809136, -0.7860608767005898], [-1.5993336920258745, -0.17812566137913655], [-1.8809432732275373, -0.573438610080969], [-1.1836684947598555, -1.6317193911091208], [-0.05912613750871809, 0.38470467693386207], [-1.8599736067038959, 1.341981003640925], [-0.5108893313177494, 0.02730059441776359], [-1.30232850485144, -0.9298335886775063], [0.7774206520404754, -1.2793191562085755], [-0.7546508720799441, -0.3486673277038578], [-1.7011572309219847, 1.1760921694501814], [-1.0657736095951564, -2.021045683831008], [2.040249746197785, 1.8441135679401288], [2.9588104940861646, 0.13760742390873298], [-0.48820832440200607, 0.007299793112342251], [2.3447903497981937, -1.8007730240583963], [0.6902464549995853, -1.4294754388400974], [-0.763720655546284, -0.3434016965683857], [-0.8312660658526394, 0.35939269006814734], [0.06179183671813215, -1.4142915463527859], [0.42059418364209855, -0.8821985075910894], [0.026425664311242214, -0.9933442071942615], [3.5339610746191297, 0.9609343271363076], [0.06707709061265654, -2.1332257673145336], [-0.38483690554117433, -0.08590617959525451], [2.6694851258920433, -0.4799400659592541], [2.123344083929215, -0.9979830914802461], [-1.720393257474338, 1.2034158014897778], [1.535785779717043, -0.8857257240521148], [-0.7339030094818977, 0.2313970092009282], [0.17121271236485622, 1.34607721468068], [-0.1848454044463215, 1.0607314140824462], [-0.16625426409826466, -0.10334522734395346], [-1.3851432939673707, 2.3941243544829542], [-0.8112056964538618, -0.4736151850712644], [1.485881741345664, 0.23359001643653923], [-0.9529643874192013, -0.34374950732154425], [-1.6167340799767518, 0.04531591711333752], [-0.375174757102701, -0.3193580736309151], [-1.5746034832821052, -1.359470193968545], [1.3785585697672271, -1.565373273697286], [-1.5326823337877835, 0.7906825890899495], [-1.7303740795253182, 1.2390315784853807], [-0.6166806966265016, -1.1426037508269171], [0.8237042809253377, -0.9490890758887148], [2.334038512969227, 1.1550938084151021], [3.1891705539186463, 0.07634042732937146], [-0.761204766269025, -1.0851154903272002], [0.03568773175507887, -0.5141781335128335], [-1.389286997232564, 2.130024402856699], [-1.473911567378929, 0.9356001277058837], [-0.8661874467107501, 0.06405594133200992], [0.5649285476342275, -0.3939841324436752], [-1.027241655164723, 0.3275791204015193], [2.5268220343834447, -0.4909898761933704]]}, \"plugins\": [{\"type\": \"reset\"}, {\"type\": \"zoom\", \"button\": true, \"enabled\": false}, {\"type\": \"boxzoom\", \"button\": true, \"enabled\": false}, {\"type\": \"htmltooltip\", \"labels\": [\"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Argentina</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.905977</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.199964</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.465967</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.195230</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Armenia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.808968</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.787111</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.749288</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>39.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.057700</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Austria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.610960</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.467558</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.656714</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.490170</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Azerbaijan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.102518</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.131187</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.264091</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.025970</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bahamas</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.349308</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.082494</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.133238</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>66.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.926070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Barbados</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.555970</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.000619</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.666502</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>66.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.923830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Belarus</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.147983</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.369303</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.377423</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>53.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>3.137070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Belgium</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.884794</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.402142</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.615086</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>48.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.924920</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Belize</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.644514</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.063733</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.031668</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>13.471580</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bhutan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.435961</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.088503</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.462929</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>66.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.083280</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bolivia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.633670</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.010190</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.476296</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>70.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.028940</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Botswana</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.235750</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.548748</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.465450</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-6.489470</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Brunei</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.761579</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.117032</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.247214</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.287260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Bulgaria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.802000</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.823956</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.498872</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.148880</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Burkina Faso</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.735064</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.480493</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.175434</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>81.199997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.247250</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cambodia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.548538</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.469431</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.654714</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>78.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>65.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.917590</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cameroon</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.995213</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.441022</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.844318</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>55.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>8.575600</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Canada</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.416783</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.987985</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.626688</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.470170</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cape Verde</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.410376</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.359021</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.523691</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-5.997490</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Chad</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.224026</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.415487</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.537504</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>69.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>53.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>19.968190</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Colombia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.584800</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.407842</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.173344</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>61.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.843070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Congo, Rep.</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.602893</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.141624</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.515644</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.300003</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.049310</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Costa Rica</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.032769</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.245337</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.327070</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.189040</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cote d'Ivoire</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.826128</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.309127</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.941214</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>54.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>21.766570</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Croatia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.204178</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.740734</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.948518</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>46.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.441510</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cuba</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.925369</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.121378</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.665211</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.338830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Cyprus</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.468203</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.572354</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.361799</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.518550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Denmark</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.353173</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.013338</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.712460</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>63.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>78.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.588940</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ecuador</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.327905</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.233973</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.070115</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.297940</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Egypt</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.370603</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.489578</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.104431</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>41.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>3.662670</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>El Salvador</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.321007</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.016458</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.872884</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.456470</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Eritrea</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.601851</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.087701</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.140908</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>59.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>14.484550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Estonia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.634461</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.014002</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.074280</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>4.037860</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ethiopia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.169331</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.320009</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.678998</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>80.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>10.905990</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Fiji</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.216381</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.337254</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.914158</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>65.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.142810</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Finland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.491016</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.387910</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.601156</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.201330</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Gambia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.767217</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.667468</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.307449</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>71.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>61.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.999020</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Georgia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.014366</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.320621</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.536519</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.169760</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Germany</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.683238</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.045494</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.588125</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.540070</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ghana</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.424216</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.186134</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.869484</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>61.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>3.343550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Greece</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.547999</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.351868</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.352177</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>49.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.031410</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Guatemala</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.618512</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.388133</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.802684</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.821130</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Guinea</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.557657</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.618504</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.087835</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>81.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>20.020490</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Guyana</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.482678</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.296600</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.466218</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.212650</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Honduras</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.109482</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.292528</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.330057</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>57.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-5.312550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Hong Kong, China</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.601864</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.665277</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.679305</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>82.124000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.604640</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Hungary</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.087386</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.746088</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.054576</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.506110</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Iceland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.429053</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>2.106433</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.663720</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>74.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>81.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.146030</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>India</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.010258</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.549648</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.176560</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>5.413260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Indonesia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.108280</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.051174</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.822841</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.202970</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Iran</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.877873</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.786061</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.517515</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.031680</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Israel</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.599334</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.178126</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.187703</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.121540</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Italy</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.880943</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.573439</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.551021</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>81.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.936880</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Jordan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.183668</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.631719</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.192079</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>39.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.360130</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Kazakhstan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.059126</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.384705</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.742618</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>65.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.182900</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Kuwait</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.859974</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.341981</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.530443</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.699997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>78.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.676570</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Latvia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.510889</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.027301</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.900620</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>70.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.947920</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Lebanon</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.302329</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.929834</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.386067</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>46.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-4.131060</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Lesotho</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.777421</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.279319</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.554775</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>44.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-26.551050</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Lithuania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.754651</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.348667</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.866371</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>71.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.384950</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Macao, China</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.701157</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.176092</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.178775</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.199997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>78.886000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.264630</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Macedonia, FYR</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.065774</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-2.021046</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.189139</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>34.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.805990</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Madagascar</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.040250</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.844114</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.245745</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>84.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>62.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.262260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Malawi</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.958810</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.137607</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.403049</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>71.699997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>49.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.115830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Maldives</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.488208</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.007300</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.089959</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>55.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.891300</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mali</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.344790</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.800773</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.349243</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>54.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>16.086970</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mauritania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.690246</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.429475</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.909027</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>47.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.131300</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mauritius</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.763721</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.343402</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.475324</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>53.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.102180</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mexico</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.831266</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.359393</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.677542</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.271540</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Moldova</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.061792</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.414292</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.152535</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.747480</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mongolia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.420594</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.882199</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.672546</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>62.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.243210</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Morocco</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.026426</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.993344</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.652641</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>46.900002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>72.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>7.208860</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Mozambique</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.533961</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.960934</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.646684</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>77.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>54.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>14.250090</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Namibia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.067077</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-2.133226</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.958044</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>41.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>53.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-8.595690</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Nicaragua</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.384837</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.085906</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.249098</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-6.316270</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Niger</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.669485</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.479940</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>6.674577</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>58.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>12.607190</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Nigeria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.123344</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.997983</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.347697</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>20.773300</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Norway</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.720393</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.203416</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.067251</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.199997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.861550</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Pakistan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.535786</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.885726</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.298761</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>50.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>17.711700</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Panama</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.733903</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.231397</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.361851</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>58.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.200000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.122110</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Paraguay</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.171213</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.346077</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.739353</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>72.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.173960</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Peru</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.184845</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.060731</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.996081</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>67.800003</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.550440</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Philippines</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.166254</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.103345</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.503630</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-7.342690</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Qatar</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.385143</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>2.394124</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.750505</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>75.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>79.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.853210</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Romania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.811206</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.473615</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.679253</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>73.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.049250</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Senegal</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.485882</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.233590</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.639393</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>65.400002</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>63.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>4.013130</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Slovak Republic</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.952964</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.343750</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.977585</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.510610</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Spain</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.616734</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.045316</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.431980</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>52.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.412720</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Sri Lanka</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.375175</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.319358</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.724435</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>54.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-0.734160</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Suriname</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.574603</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.359470</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.470836</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>45.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>68.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-16.395260</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Swaziland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>1.378559</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.565373</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.621692</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>44.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.912150</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Sweden</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.532682</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.790683</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.656070</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.299999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>80.700000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.047830</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Switzerland</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.730374</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.239032</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.876402</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.699997</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>81.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.662370</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Syria</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.616681</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.142604</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.680752</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>44.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>76.100000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.277510</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Tajikistan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.823704</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.949089</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.477146</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>51.200001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>69.000000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>4.506510</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Tanzania</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.334039</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>1.155094</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.191813</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>78.099998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>57.900000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>2.362050</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Togo</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>3.189171</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.076340</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>7.097305</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>64.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>59.400000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>24.349910</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Turkey</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.761205</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-1.085115</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.686151</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>43.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>5.927800</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Ukraine</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.035688</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.514178</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.972057</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>53.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>67.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>0.691150</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>United Arab Emirates</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.389287</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>2.130024</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>11.476259</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>75.000000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.600000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-3.332680</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>United States</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.473912</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.935600</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>10.832812</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>62.599998</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>77.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-1.771030</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Uruguay</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-0.866187</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.064056</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.497198</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>75.500000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-2.448210</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Uzbekistan</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>0.564929</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.393984</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.059925</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>56.500000</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>68.300000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>1.730380</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Venezuela</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>-1.027242</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>0.327579</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>9.689949</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>59.700001</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>74.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>-5.746690</td>\\n    </tr>\\n  </tbody>\\n</table>\", \"<table border=\\\"1\\\" class=\\\"dataframe\\\">\\n  <thead>\\n    <tr style=\\\"text-align: right;\\\">\\n      <th></th>\\n      <th>Zambia</th>\\n    </tr>\\n  </thead>\\n  <tbody>\\n    <tr>\\n      <th>PC1</th>\\n      <td>2.526822</td>\\n    </tr>\\n    <tr>\\n      <th>PC2</th>\\n      <td>-0.490990</td>\\n    </tr>\\n    <tr>\\n      <th>log_income</th>\\n      <td>8.002437</td>\\n    </tr>\\n    <tr>\\n      <th>employment</th>\\n      <td>60.799999</td>\\n    </tr>\\n    <tr>\\n      <th>life_expectancy</th>\\n      <td>48.800000</td>\\n    </tr>\\n    <tr>\\n      <th>diff_completion</th>\\n      <td>10.728000</td>\\n    </tr>\\n  </tbody>\\n</table>\"], \"voffset\": 10, \"hoffset\": 10, \"id\": \"el8264508623984pts\"}], \"width\": 800.0, \"axes\": [{\"ydomain\": [-3.0, 3.0], \"axes\": [{\"nticks\": 7, \"position\": \"bottom\", \"scale\": \"linear\", \"fontsize\": 13.0, \"tickformat\": null, \"tickvalues\": null, \"grid\": {\"color\": \"#FFFFFF\", \"alpha\": 1.0, \"dasharray\": \"10,0\", \"gridOn\": true}}, {\"nticks\": 7, \"position\": \"left\", \"scale\": \"linear\", \"fontsize\": 13.0, \"tickformat\": null, \"tickvalues\": null, \"grid\": {\"color\": \"#FFFFFF\", \"alpha\": 1.0, \"dasharray\": \"10,0\", \"gridOn\": true}}], \"sharey\": [], \"zoomable\": true, \"xscale\": \"linear\", \"sharex\": [], \"xlim\": [-2.0, 4.0], \"markers\": [{\"data\": \"data01\", \"edgecolor\": \"#000000\", \"markerpath\": [[[0.0, 7.5], [1.9890232500000002, 7.5], [3.896849030886401, 6.709752686911813], [5.303300858899107, 5.303300858899107], [6.709752686911813, 3.896849030886401], [7.5, 1.9890232500000002], [7.5, 0.0], [7.5, -1.9890232500000002], [6.709752686911813, -3.896849030886401], [5.303300858899107, -5.303300858899107], [3.896849030886401, -6.709752686911813], [1.9890232500000002, -7.5], [0.0, -7.5], [-1.9890232500000002, -7.5], [-3.896849030886401, -6.709752686911813], [-5.303300858899107, -5.303300858899107], [-6.709752686911813, -3.896849030886401], [-7.5, -1.9890232500000002], [-7.5, 0.0], [-7.5, 1.9890232500000002], [-6.709752686911813, 3.896849030886401], [-5.303300858899107, 5.303300858899107], [-3.896849030886401, 6.709752686911813], [-1.9890232500000002, 7.5], [0.0, 7.5]], [\"M\", \"C\", \"C\", \"C\", \"C\", \"C\", \"C\", \"C\", \"C\", \"Z\"]], \"coordinates\": \"data\", \"xindex\": 0, \"edgewidth\": 1, \"zorder\": 2, \"yindex\": 1, \"alpha\": 0.6, \"id\": \"el8264508623984pts\", \"facecolor\": \"#0000FF\"}], \"bbox\": [0.125, 0.125, 0.775, 0.775], \"id\": \"el8264503983496\", \"lines\": [], \"texts\": [{\"color\": \"#262626\", \"coordinates\": \"axes\", \"v_baseline\": \"hanging\", \"rotation\": -0.0, \"h_anchor\": \"middle\", \"position\": [0.5, -0.0734767025089606], \"text\": \"PC1\", \"fontsize\": 14.3, \"zorder\": 3, \"alpha\": 1, \"id\": \"el8264504070856\"}, {\"color\": \"#262626\", \"coordinates\": \"axes\", \"v_baseline\": \"auto\", \"rotation\": -90.0, \"h_anchor\": \"middle\", \"position\": [-0.04831989247311827, 0.5], \"text\": \"PC2\", \"fontsize\": 14.3, \"zorder\": 3, \"alpha\": 1, \"id\": \"el8264504026080\"}, {\"color\": \"#262626\", \"coordinates\": \"axes\", \"v_baseline\": \"auto\", \"rotation\": -0.0, \"h_anchor\": \"middle\", \"position\": [0.5, 1.0149342891278375], \"text\": \"Interactive projection on PC plane\", \"fontsize\": 20.0, \"zorder\": 3, \"alpha\": 1, \"id\": \"el8264504134096\"}], \"axesbgalpha\": null, \"xdomain\": [-2.0, 4.0], \"ylim\": [-3.0, 3.0], \"images\": [], \"collections\": [], \"yscale\": \"linear\", \"paths\": [], \"axesbg\": \"#EAEAF2\"}], \"id\": \"el8264503296208\", \"height\": 480.0});\n",
       "            })\n",
       "         });\n",
       "}\n",
       "</script>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Example with modifications from:\n",
    "# http://mpld3.github.io/examples/html_tooltips.html\n",
    "\n",
    "# Define some CSS to control our custom labels\n",
    "\n",
    "css = \"\"\"\n",
    "table\n",
    "{\n",
    "  border-collapse: collapse;\n",
    "}\n",
    "th\n",
    "{\n",
    "  color: #ffffff;\n",
    "  background-color: #000000;\n",
    "}\n",
    "td\n",
    "{\n",
    "  background-color: #cccccc;\n",
    "}\n",
    "table, th, td\n",
    "{\n",
    "  font-family:Arial, Helvetica, sans-serif;\n",
    "  border: 1px solid black;\n",
    "  text-align: right;\n",
    "}\n",
    "\"\"\"\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10, 6))\n",
    "\n",
    "# labels for the tooltips\n",
    "labels = []\n",
    "for i in range(len(df_transform)):\n",
    "    label = df_transform.ix[[i], :].T\n",
    "    label.columns = [df_transform.index[i]]\n",
    "    # .to_html() is unicode; so make leading 'u' go away with str()\n",
    "    labels.append(str(label.to_html()))\n",
    "\n",
    "# scatter plot\n",
    "points = ax.plot(df_transform.PC1, df_transform.PC2, 'o', color='b',\n",
    "                 mec='k', ms=15, mew=1, alpha=.6)\n",
    "\n",
    "ax.set_xlabel('PC1')\n",
    "ax.set_ylabel('PC2')\n",
    "ax.set_title('Interactive projection on PC plane', size=20)\n",
    "\n",
    "# add tooltips\n",
    "tooltip = plugins.PointHTMLTooltip(points[0], labels,\n",
    "                                   voffset=10, hoffset=10, css=css)\n",
    "plugins.connect(fig, tooltip)\n",
    "\n",
    "mpld3.display()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}

utilities.py

'''A collection of helper procedures for the 
PCA demonstration'''

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

def components_table(pca, original):
    '''Organize the pca components_ attribute in a dataframe
    with labels for the orignal features and for the PCs'''

    table = pd.DataFrame(index=["PC{}".format(i + 1) for i in range(pca.n_components)],
                        columns=original.columns, data=pca.components_)
    return table


def biplot(original, reduced_data, pca):
    '''
    Produce a biplot that shows a scatterplot of the reduced
    data and the projections of the original features.
    
    original: original data, before transformation.
               Needs to be a pandas dataframe with valid column names
    reduced_data: the reduced data (the first two dimensions are plotted)
    pca: pca object that contains the components_ attribute

    return: a matplotlib AxesSubplot object (for any additional customization)
    
    This procedure is inspired by the script:
    https://github.com/teddyroland/python-biplot
    '''

    fig, ax = plt.subplots(figsize = (12,6))
    # scatterplot of the reduced data    
    ax.scatter(x=reduced_data[:, 0], y=reduced_data[:, 1], 
        facecolors='b', edgecolors='b', s=70, alpha=0.5)
    
    feature_vectors = pca.components_.T

    # we use scaling factors to make the arrows easier to see
    arrow_size, text_pos = 2.0, 2.5,

    # projections of the original features
    for i, v in enumerate(feature_vectors):
        ax.arrow(0, 0, arrow_size*v[0], arrow_size*v[1], 
                  head_width=0.1, head_length=0.2, linewidth=1, color='red')
        ax.text(v[0]*text_pos, v[1]*text_pos, original.columns[i], color='black', 
                 ha='center', va='center', fontsize=18)

    ax.set_xlabel("PC 1", fontsize=14)
    ax.set_ylabel("PC 2", fontsize=14)
    ax.set_title("PC plane with original feature projections.", fontsize=16);
    return ax