Articulation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Validation of Articulation tests   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Contents\n",
    "1. [Problem background](#bkgnd)\n",
    "2. [Data Prep](#dataprep)\n",
    "3. [Students who took one test](#one)\n",
    "4. [Students who took both tests](#both)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='bkgnd'></a> \n",
    "# Questions and Background"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Students can either take the Arizona or Goldman-Fristoe test  \n",
    "**Question:** Can these two tests be combined?  \n",
    "Similar studentss should score the same. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How should we define \"similar students\"?  \n",
    "For now I will consider the variable included in our population characteristics summary table"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Covariate | Redcap Variable\n",
    "----------|--------------------\n",
    "Age (Arizona) | `age_test_aaps`\n",
    "Age (Goldman Firstoe) | `age_test_gf2`\n",
    "Type of hearing loss | `type_hl_as`, `type_hl_ad`\n",
    "Degree of hearing loss | `degree_hl`\n",
    "Technology | `tech_as`, `tech_ad`\n",
    "Gender | `male`\n",
    "Race | `race`\n",
    "Primary language spoken | `prim_lang`\n",
    "Number of children in home | `sibs`\n",
    "Age of hearing loss diagnosis | `onset_1`\n",
    "Age first amplified | `age_amp`\n",
    "Age of intervention | `age_int`\n",
    "Age enrolled in OPTION | `age`\n",
    "Length of time in OPTION | `time`\n",
    "Known cause of hearing loss | `med_cause=0`\n",
    "Known syndrome | `synd_cause=0`\n",
    "Another diagnosed disability| `etiology`\n",
    "Another disability impacts ability to learn | `etiology_2`\n",
    "Other services outside of OPTION | `otherserv`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How will we asses this?\n",
    "* Linear model with score as outcome and test type as predictor, controlling for predictors noted above\n",
    "* If a student took both tests we could asses how well one test score predicts the other"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Statistical considerations:\n",
    "* Many covariates are listed above.  Asses sample size to determine degrees of freedom.\n",
    "* How to handle students with multiple observations? \n",
    "    - For now will take last observation\n",
    "* How to handle students with both tests?\n",
    "    - For now only looking at students who took one test\n",
    "    - Could try to predict Arizona score form Goldman-Fristoe and vice versa"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='dataprep'></a> \n",
    "# Data Prep"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "CHRIS = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# %pwd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "False    11601\n",
      "True      2567\n",
      "dtype: int64\n",
      "There are 716 null values for non_english\n",
      "_mother_ed:\n",
      "6    5181\n",
      "4    3022\n",
      "3    2037\n",
      "5    1628\n",
      "2    1421\n",
      "1     490\n",
      "0     213\n",
      "dtype: int64\n",
      "mother_ed:\n",
      "1    3458\n",
      "2    3022\n",
      "3    1628\n",
      "0     703\n",
      "dtype: int64\n",
      "\n",
      "There are 6073 null values for mother_ed\n",
      "There are 3494 null values for premature_weeks\n",
      "oad:\n",
      "0    4627\n",
      "1       4\n",
      "2       2\n",
      "dtype: int64\n",
      "There are 1715 null values for OAD\n",
      "\n",
      "hearing_aid:\n",
      "2    2162\n",
      "0    1638\n",
      "1     802\n",
      "dtype: int64\n",
      "There are 1768 null values for hearing_aid\n",
      "\n",
      "cochlear:\n",
      "0    3081\n",
      "2     918\n",
      "1     634\n",
      "dtype: int64\n",
      "There are 1715 null values for cochlear\n",
      "14884\n",
      "There are 0 null values for tech\n",
      "_race:\n",
      "0    7766\n",
      "2    2528\n",
      "1    1358\n",
      "3    1042\n",
      "6     721\n",
      "8     527\n",
      "7     239\n",
      "4      65\n",
      "5      33\n",
      "dtype: int64\n",
      "race:\n",
      "0    7766\n",
      "2    2528\n",
      "1    1358\n",
      "4    1346\n",
      "3    1042\n",
      "dtype: int64\n",
      "There are 844 null values for race\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/alicetoll/anaconda/envs/py34/lib/python3.3/site-packages/pandas/io/parsers.py:1170: DtypeWarning: Columns (46) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  data = self._reader.read(nrows)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEACAYAAAC+gnFaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10a4ba450>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if CHRIS:\n",
    "    print(\"change path to demographics file\")\n",
    "else:\n",
    "    %run '/Users/alicetoll/Documents/OPTION/LSL-DR/Demographics.ipynb'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "if CHRIS:\n",
    "    demographic = pd.read_csv('demographic.csv')\n",
    "else:\n",
    "    demographic = pd.read_pickle('demographic.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "covar_fields = ['study_id', 'redcap_event_name', 'male', 'race', 'prim_lang', 'sib', 'onset_1',\n",
    "                'age_amp', 'age_int', 'age', 'time', 'med_cause', 'synd_cause', 'etiology', 'etiology_2',\n",
    "                'otherserv', 'type_hl_ad', 'type_hl_as']\n",
    "\n",
    "covar = demographic[covar_fields]\n",
    "\n",
    "# Need to create vars and add: \n",
    "# - type of hearing loss\n",
    "# - degree of hearing loss\n",
    "# - technology\n",
    "# - CI Manufacturer\n",
    "\n",
    "# Do we want to collapse race?\n",
    "\n",
    "# Age is in articulation_fields\n",
    "# - age_test_aaps = Arizona\n",
    "# - age_test_gf2 = Goldman-Fristoe\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/alicetoll/anaconda/envs/py34/lib/python3.3/site-packages/IPython/kernel/__main__.py:7: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "/Users/alicetoll/anaconda/envs/py34/lib/python3.3/site-packages/pandas/core/indexing.py:415: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self.obj[item] = s\n",
      "/Users/alicetoll/anaconda/envs/py34/lib/python3.3/site-packages/IPython/kernel/__main__.py:10: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "/Users/alicetoll/anaconda/envs/py34/lib/python3.3/site-packages/IPython/kernel/__main__.py:14: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "/Users/alicetoll/anaconda/envs/py34/lib/python3.3/site-packages/IPython/kernel/__main__.py:17: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n"
     ]
    }
   ],
   "source": [
    "# Type of hearing loss\n",
    "# - Bilateral SNHL\n",
    "# - Unilateral SNHL\n",
    "# - Bilateral Auditory Neuropathy\n",
    "# - Unilateral Auditory Neuropathy\n",
    "\n",
    "covar['bilateral_snhl'] = 0\n",
    "covar.loc[(covar.type_hl_ad == 0) & (covar.type_hl_as == 0), 'bilateral_snhl'] = 1\n",
    "\n",
    "covar['unilateral_snhl'] = 0\n",
    "covar.loc[(covar.type_hl_ad == 0) & (covar.type_hl_as == 4), 'unilateral_snhl'] = 1\n",
    "covar.loc[(covar.type_hl_ad == 4) & (covar.type_hl_as == 0), 'unilateral_snhl'] = 1\n",
    "\n",
    "covar['bilateral_an'] = 0\n",
    "covar.loc[(covar.type_hl_ad == 3) & (covar.type_hl_as == 3), 'bilateral_an'] = 1\n",
    "\n",
    "covar['unilateral_an'] = 0\n",
    "covar.loc[(covar.type_hl_ad == 0) & (covar.type_hl_as == 3), 'unilateral_an'] = 1\n",
    "covar.loc[(covar.type_hl_ad == 3) & (covar.type_hl_as == 0), 'unilateral_an'] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# - Bilateral SNHL\n",
    "# - Unilateral SNHL\n",
    "# - Bilateral Auditory Neuropathy\n",
    "# - Unilateral Auditory Neuropathy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/alicetoll/anaconda/envs/py34/lib/python3.3/site-packages/pandas/core/indexing.py:249: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self.obj[key] = _infer_fill_value(value)\n"
     ]
    }
   ],
   "source": [
    "bi_snhl = covar.bilateral_snhl == 1\n",
    "uni_snhl = covar.unilateral_snhl == 1\n",
    "bi_an = covar.bilateral_an == 1\n",
    "uni_an = covar.unilateral_an == 1\n",
    "\n",
    "covar.loc[bi_snhl, 'type_hl'] = \"bi_snhl\"\n",
    "covar.loc[uni_snhl, 'type_hl'] = \"uni_snhl\"\n",
    "covar.loc[bi_an, 'type_hl'] = \"bi_an\"\n",
    "covar.loc[uni_an, 'type_hl'] = \"uni_an\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "articulation_fields = ['study_id','redcap_event_name', 'age_test_aaps','aaps_ss','age_test_gf2','gf2_ss']\n",
    "articulation = lsl_dr_project.export_records(fields=articulation_fields, format='df', \n",
    "                                           df_kwargs={'index_col':None,\n",
    "                                                      'na_values':[999, 9999]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# articulation.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    5760\n",
      "1      73\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# create indicator variable if a student took either test\n",
    "articulation.loc[articulation.aaps_ss.notnull() | articulation.gf2_ss.notnull(), 'test'] = 1\n",
    "# drop observations when neither test was taken\n",
    "temp = articulation.dropna(subset = ['test'])\n",
    "# Can drop test variable\n",
    "temp = temp.drop('test', 1)\n",
    "# Create new variable if student took both tests in one observation\n",
    "temp['both'] = 0\n",
    "temp.loc[temp.aaps_ss.notnull() & temp.gf2_ss.notnull(), 'both'] = 1\n",
    "print(temp.both.value_counts()) # 73 students took both tests, 5716 took only one\n",
    "# temp.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Merge covariates and articulation data frames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# merge\n",
    "temp2 = pd.merge(temp, covar, on=['study_id', 'redcap_event_name'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "AAPS = temp2.aaps_ss.notnull()\n",
    "GF2 = temp2.gf2_ss.notnull()\n",
    "\n",
    "temp2.loc[AAPS, \"test_name\"] = \"AAPS\"\n",
    "temp2.loc[GF2, \"test_name\"] = \"GF2\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create datasets which contain students who either took one or both tests in a single year.  If a student has multiple observations, take the last observation in each dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "We have 5760 observations where a student took one test in a single year, but only 2730 unique students\n",
      "We have 73 observations where a student took both test in a single year, but only 55 unique students\n"
     ]
    }
   ],
   "source": [
    "# One test\n",
    "single = temp2.loc[temp2.both==0,]\n",
    "a = single.shape[0]\n",
    "single = single.groupby('study_id').last()\n",
    "b = single.shape[0]\n",
    "print('We have', a, 'observations where a student took one test in a single year, but only', b, 'unique students')\n",
    "\n",
    "# Both tests\n",
    "both = temp2.loc[temp2.both==1,]\n",
    "a = both.shape[0]\n",
    "both = both.groupby('study_id').last()\n",
    "b = both.shape[0]\n",
    "print('We have', a, 'observations where a student took both test in a single year, but only', b, 'unique students')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "del temp\n",
    "del temp2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='one'></a> \n",
    "# Students who only took one test (within a single year)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Create variables to hold test score and age, and indicator for test name\n",
    "\n",
    "AAPS = single.aaps_ss.notnull()\n",
    "GF2 = single.gf2_ss.notnull()\n",
    "\n",
    "single.loc[AAPS, \"test_name\"] = \"AAPS\"\n",
    "single.loc[GF2, \"test_name\"] = \"GF2\"\n",
    "\n",
    "single.loc[AAPS, \"score\"] = single.aaps_ss\n",
    "single.loc[GF2, \"score\"] = single.gf2_ss\n",
    "\n",
    "single.loc[AAPS, \"age_test\"] = single.age_test_aaps\n",
    "single.loc[GF2, \"age_test\"] = single.age_test_gf2\n",
    "\n",
    "# temp.head()\n",
    "\n",
    "# single = single.drop(['age_test_aaps', 'aaps_ss', 'age_test_gf2', 'gf2_ss', 'both'], 1)\n",
    "# artShort1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2730"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Number of observations\n",
    "single.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "redcap_event_name       0\n",
       "age_test_aaps        2448\n",
       "aaps_ss              2448\n",
       "age_test_gf2          181\n",
       "gf2_ss                180\n",
       "both                    0\n",
       "male                   22\n",
       "race                   69\n",
       "prim_lang              42\n",
       "sib                   247\n",
       "onset_1               913\n",
       "age_amp               973\n",
       "age_int              1063\n",
       "age                    46\n",
       "time                  170\n",
       "med_cause              57\n",
       "synd_cause             60\n",
       "etiology               48\n",
       "etiology_2            527\n",
       "otherserv             974\n",
       "type_hl_ad            429\n",
       "type_hl_as            409\n",
       "bilateral_snhl          0\n",
       "unilateral_snhl         0\n",
       "bilateral_an            0\n",
       "unilateral_an           0\n",
       "type_hl               780\n",
       "test_name               0\n",
       "score                   0\n",
       "age_test                1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Missingness\n",
    "len(single.index)-single.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import seaborn as sns\n",
    "# sns.set(style=\"white\", color_codes=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10dcc5850>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfcAAAFXCAYAAAC/aQfJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x106c7dc10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "width = 3\n",
    "sns.kdeplot(np.array(single[single.test_name == 'GF2'].score), bw=width, label = \"Goldman-Fristoe\")\n",
    "sns.kdeplot(np.array(single[single.test_name == 'AAPS'].score), bw=width, label = \"Arizona\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bi_snhl     1811\n",
       "bi_an         73\n",
       "uni_snhl      57\n",
       "uni_an         9\n",
       "dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "single.type_hl.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We may want to consider looking at just bilateral SNHL vs other based on sample size."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjkAAAGGCAYAAACZn5BQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ntXZ8O6zXYxIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x112db5850>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Kernal Density Estimates broken out by type of hearing loss\n",
    "\n",
    "plt.figure()\n",
    "\n",
    "# Bilateral SNHL\n",
    "plt.subplot(2,2,1)\n",
    "sns.kdeplot(np.array(single.loc[(single.test_name == 'GF2') & (single.bilateral_snhl == 1), 'score']), \n",
    "                bw=width)\n",
    "sns.kdeplot(np.array(single.loc[(single.test_name == 'AAPS') & (single.bilateral_snhl == 1), 'score']), \n",
    "                bw=width)\n",
    "plt.title('Bilateral SNHL')\n",
    "plt.ylabel('Density')\n",
    "plt.xlabel('Standardized Score')\n",
    "\n",
    "# Unilateral SNHL\n",
    "plt.subplot(2,2,2)\n",
    "sns.kdeplot(np.array(single.loc[(single.test_name == 'GF2') & (single.unilateral_snhl == 1), 'score']), \n",
    "                bw=width)\n",
    "sns.kdeplot(np.array(single.loc[(single.test_name == 'AAPS') & (single.unilateral_snhl == 1), 'score']), \n",
    "                bw=width)\n",
    "plt.title('Unilateral SNHL')\n",
    "plt.ylabel('Density')\n",
    "plt.xlabel('Standardized Score')\n",
    "\n",
    "# Bilateral AN\n",
    "plt.subplot(2,2,3)\n",
    "sns.kdeplot(np.array(single.loc[(single.test_name == 'GF2') & (single.bilateral_an == 1), 'score']), \n",
    "                bw=width, label = \"Goldman-Fristoe\")\n",
    "sns.kdeplot(np.array(single.loc[(single.test_name == 'AAPS') & (single.bilateral_an == 1), 'score']), \n",
    "                bw=width, label = \"Arizona\")\n",
    "plt.title('Bilateral Auditory Neuropathy')\n",
    "plt.ylabel('Density')\n",
    "plt.xlabel('Standardized Score')\n",
    "\n",
    "# Unilateral AN\n",
    "plt.subplot(2,2,4)\n",
    "sns.kdeplot(np.array(single.loc[(single.test_name == 'GF2') & (single.unilateral_an == 1), 'score']), \n",
    "                bw=width)\n",
    "# no obs for Arizona with Unilateral AN\n",
    "# sns.kdeplot(np.array(single.loc[(single.test_name == 'AAPS') & (single.unilateral_an == 1), 'score']), \n",
    "#                bw=width, label = \"Arizona\")\n",
    "plt.title('Unilateral Auditory Neuropathy')\n",
    "plt.ylabel('Density')\n",
    "plt.xlabel('Standardized Score')\n",
    "\n",
    "\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Want to make this function"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "def mykdeplot(df, var, width):\n",
    "    sns.kdeplot(np.array(getattr(df[df.Group == 'a'], var)), bw=width, label = \"Group A\")\n",
    "    sns.kdeplot(np.array(getattr(df[df.Group == 'b'], var)), bw=width, label = \"Group B\")"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "def mykdeplot(df, var, width):\n",
    "    sns.kdeplot(np.array(getattr(df[(artShort1.test_name == 'GF2') & (df.var == 1)], 'score')),\n",
    "                         bw=width, label = \"Goldman-Fristoe\")\n",
    "    sns.kdeplot(np.array(getattr(artShort1[(df.test_name == 'AAPS') & (df.var == 1)], 'score')), \n",
    "                bw=width, label = \"Arizona\")"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "mykdeplot(single, 'bilateral_an', 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Linear Model"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "# https://github.com/justmarkham/DAT4/blob/master/notebooks/08_linear_regression.ipynb\n",
    "\n",
    "# this is the standard import if you're using \"formula notation\" (similar to R)\n",
    "import statsmodels.formula.api as smf\n",
    "\n",
    "# create a fitted model in one line\n",
    "lm = smf.ols(formula='score ~ test_name + age_test + male', data=artShort1).fit()\n",
    "\n",
    "# print the coefficients\n",
    "print(lm.params)\n",
    "\n",
    "# print the confidence intervals for the model coefficients\n",
    "print(lm.conf_int())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import statsmodels.formula.api as smf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Linear model predicting score from test name, age, gender, and type of hearing loss.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>          <td>score</td>      <th>  R-squared:         </th> <td>   0.019</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.016</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   7.470</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sun, 17 Apr 2016</td> <th>  Prob (F-statistic):</th> <td>6.29e-09</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:46:52</td>     <th>  Log-Likelihood:    </th> <td> -12016.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>  2707</td>      <th>  AIC:               </th> <td>2.405e+04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>  2699</td>      <th>  BIC:               </th> <td>2.409e+04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     7</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "          <td></td>            <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>        <td>   87.8178</td> <td>    1.979</td> <td>   44.369</td> <td> 0.000</td> <td>   83.937    91.699</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>test_name[T.GF2]</th> <td>    1.4962</td> <td>    1.589</td> <td>    0.942</td> <td> 0.346</td> <td>   -1.619     4.612</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test</th>         <td>   -0.0570</td> <td>    0.013</td> <td>   -4.545</td> <td> 0.000</td> <td>   -0.082    -0.032</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>male</th>             <td>    0.1904</td> <td>    0.792</td> <td>    0.241</td> <td> 0.810</td> <td>   -1.362     1.742</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bilateral_snhl</th>   <td>   -0.5525</td> <td>    0.880</td> <td>   -0.628</td> <td> 0.530</td> <td>   -2.279     1.174</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>unilateral_snhl</th>  <td>    9.0884</td> <td>    2.868</td> <td>    3.169</td> <td> 0.002</td> <td>    3.465    14.711</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bilateral_an</th>     <td>   -7.9264</td> <td>    2.547</td> <td>   -3.112</td> <td> 0.002</td> <td>  -12.921    -2.932</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>unilateral_an</th>    <td>  -17.2806</td> <td>    6.879</td> <td>   -2.512</td> <td> 0.012</td> <td>  -30.769    -3.792</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>180.329</td> <th>  Durbin-Watson:     </th> <td>   1.756</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td> 204.964</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.653</td>  <th>  Prob(JB):          </th> <td>3.11e-45</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 2.666</td>  <th>  Cond. No.          </th> <td>1.38e+03</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                  score   R-squared:                       0.019\n",
       "Model:                            OLS   Adj. R-squared:                  0.016\n",
       "Method:                 Least Squares   F-statistic:                     7.470\n",
       "Date:                Sun, 17 Apr 2016   Prob (F-statistic):           6.29e-09\n",
       "Time:                        18:46:52   Log-Likelihood:                -12016.\n",
       "No. Observations:                2707   AIC:                         2.405e+04\n",
       "Df Residuals:                    2699   BIC:                         2.409e+04\n",
       "Df Model:                           7                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "====================================================================================\n",
       "                       coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "------------------------------------------------------------------------------------\n",
       "Intercept           87.8178      1.979     44.369      0.000        83.937    91.699\n",
       "test_name[T.GF2]     1.4962      1.589      0.942      0.346        -1.619     4.612\n",
       "age_test            -0.0570      0.013     -4.545      0.000        -0.082    -0.032\n",
       "male                 0.1904      0.792      0.241      0.810        -1.362     1.742\n",
       "bilateral_snhl      -0.5525      0.880     -0.628      0.530        -2.279     1.174\n",
       "unilateral_snhl      9.0884      2.868      3.169      0.002         3.465    14.711\n",
       "bilateral_an        -7.9264      2.547     -3.112      0.002       -12.921    -2.932\n",
       "unilateral_an      -17.2806      6.879     -2.512      0.012       -30.769    -3.792\n",
       "==============================================================================\n",
       "Omnibus:                      180.329   Durbin-Watson:                   1.756\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              204.964\n",
       "Skew:                          -0.653   Prob(JB):                     3.11e-45\n",
       "Kurtosis:                       2.666   Cond. No.                     1.38e+03\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 1.38e+03. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lm = smf.ols(formula=\"\"\"score ~ test_name + age_test + male + \n",
    "                        bilateral_snhl + unilateral_snhl + bilateral_an + unilateral_an\"\"\", \n",
    "             data=single).fit()\n",
    "lm.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the output above, we see `test_name` is not significant (p = 0.334).  This tells us that after accounting for the student's age when taking the test, gender, and type of hearing loss, knowing which test the student took does not help us to predict there score.  This suggests that similar students score the same regardless of which test they took.\n",
    "\n",
    "We need to determine which variables should be included in our model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "<a id='both'></a>  \n",
    "# Students who only took both tests (within a single year)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We currently treat standardized Arizona and Goldman-Fristoe tests scores the same.  The goal of this analysis is to relate the two tests so they can be transformed for future modeling purposes.  \n",
    "\n",
    "We only have 55 observations (before accounting for missingness).  That only leaves us with 3-6 degress of freedom for modeling. A continuous variable will use up one degree of freedom, and the number of deggrees of freedom used by a categorical variable is one less than the number of categories.\n",
    "\n",
    "My initial intuition was to model Arizona score as a function of Goldman-Fristoe score, age, type of hearing loss, and whether or not the student receives other learning servies.  Unfortunatly on this small sample size the observations for type of hearing loss and other serivces are rather homongenous, so I do not believe they will be very helpful in predicted Arizona score.\n",
    "\n",
    "We may want to allow for nonlinearity by including a spline term for the Goldman-Fristoe score.  A spline with 3 knots would use 3 degrees of freedom.  \n",
    "\n",
    "The candidate models considered are sumarized in the subsection titled **Models**.  \n",
    "\n",
    "Variables that were eliminated from consideration due to homogenity\n",
    "* `type_hl` (47 out of 55 students have bilateral SNHL)\n",
    "* `etiology` (42 out of 55 have another diagnosed disability)\n",
    "* `otherserv` (35 out of 55 do not recieve other educational services, 12 out of 55 are missing)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Response:  Arizona scores (`aaps_ss`)**\n",
    "\n",
    "Considered Covariate | Redcap Variable | Degrees of Freedom\n",
    "----------|--------------------------------------\n",
    "Goldman Fristoe Score | gf2_ss | 1\n",
    "Goldman Fristoe Score, spline | gf2_ss | 3\n",
    "Age (Arizona) | `age_test_aaps` | 1\n",
    "Age (Goldman Firstoe) | `age_test_gf2` | 1\n",
    "Gender | `male` | 1\n",
    "Race (White/Non-White) | `white` | 1\n",
    "Age of hearing loss diagnosis | `onset_1` | 1\n",
    "Length of time in OPTION | `time` | 1\n",
    "Another disability impacts ability to learn | `etiology_impact` | 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Potential variables for modelings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(55, 28)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# With 55 observations (before missingness), we really only have 3-6 degrees of freedom to use in a model. \n",
    "both.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x10c25ae10>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgYAAAFmCAYAAAAWI1JnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQjej6vdIAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x109792590>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Arizona is more conservative towards \"normal\" (85)\n",
    "# May want to consider a spline for Goldman-Fristoe\n",
    "width = 4\n",
    "sns.kdeplot(np.array(both.gf2_ss), bw=width, label = \"Goldman-Fristoe\")\n",
    "sns.kdeplot(np.array(both.aaps_ss), bw=width, label = \"Arizona\")\n",
    "plt.ylabel('Density')\n",
    "plt.xlabel('Standardized Score')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x1110c81d0>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgYAAAFmCAYAAAAWI1JnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x109792650>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Large age range (in months).  May want to use a spline for age\n",
    "# Don't have enough degrees of freedom to spline both variables\n",
    "sns.kdeplot(both.age_test_gf2, bw=5, label = \"Goldman-Fristoe\")\n",
    "sns.kdeplot(both.age_test_aaps, bw=5, label = \"Arizona\")\n",
    "plt.ylabel('Density')\n",
    "plt.xlabel('Age (months)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NaN     2\n",
       " 0     26\n",
       " 1      6\n",
       " 2      8\n",
       " 3      4\n",
       " 4      9\n",
       "dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Race has a lot of categories\n",
    "# To preserve degrees of freedom we can reduce to white/non-white\n",
    "both.race.value_counts(dropna=False, sort=False)\n",
    "# 0 = Cacucasian\n",
    "# 1 = Black or African American\n",
    "# 2 = Hispanic or Latino\n",
    "# 3 = Asian\n",
    "# 4 = Native Hawaiians/Pacific Islanders\n",
    "# 5 = American Indian/Alaska Natives\n",
    "# 6 = Multiracial\n",
    "# 7 = Unknown\n",
    "# 8 = Other"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NaN     2\n",
       " 0     27\n",
       " 1     26\n",
       "dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "both['white'] = np.nan\n",
    "both.loc[(both.race == 0), 'white'] = 1\n",
    "both.loc[(both.race !=0) & (pd.notnull(both.race)), 'white'] = 0\n",
    "both.white.value_counts(dropna=False, sort=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NaN     1\n",
       " 0     11\n",
       " 1      5\n",
       " 2     12\n",
       " 4     26\n",
       "dtype: int64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Etiology 2: impact of the other condition(s) that contribute to the child's overall learning abilities\n",
    "# To preserve degrees of freedom we can reduce to impacts (1, 2, 3) and doesn't imact (0, 4)\n",
    "both.etiology_2.value_counts(dropna=False, sort=False)\n",
    "# 0 = None\n",
    "# 1 = Mild\n",
    "# 2 = Moderate\n",
    "# 3 = Severe\n",
    "# 4 = N/A (no additional disability suspected or identified)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 0     37\n",
       " 1     17\n",
       "NaN     1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "both['etiology_impact'] = np.nan\n",
    "both.loc[(both.etiology_2 == 0) | (both.etiology_2 == 4), 'etiology_impact'] = 0\n",
    "both.loc[(both.etiology_2 == 1) | (both.etiology_2 == 2) | (both.etiology_2 == 3), 'etiology_impact'] = 1\n",
    "both.etiology_impact.value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Amount of time enrolled in OPTION: 2 missing obs\n",
    "len(both.time.index)-both.time.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Age of onset: 10 missing obs\n",
    "# This is a lot of observations to sacrifice\n",
    "# Would be good if we could use multiple imputation\n",
    "len(both.onset_1.index)-both.onset_1.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Variables eliminated due to heterogeneity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bi_snhl     47\n",
       "NaN          4\n",
       "bi_an        3\n",
       "uni_snhl     1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Most of the students who took both tests have bilateral SNHL\n",
    "# 4 observations are missing\n",
    "# Because a large majority of the students only have bilateral SNHL, it will not be a useful predictor.\n",
    "both.type_hl.value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     9\n",
       "1    42\n",
       "2     1\n",
       "3     3\n",
       "dtype: int64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 42 out of 55 do not have another diagnosed disability\n",
    "# note: this doesn't agree with etiology_2 coding\n",
    "both.etiology.value_counts(dropna=False, sort=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NaN    12\n",
       " 0      3\n",
       " 1     35\n",
       " 2      5\n",
       "dtype: int64"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Other services: 12 missing obs\n",
    "# Most students do not recieve other services\n",
    "both.otherserv.value_counts(dropna=False, sort=False)\n",
    "# 0 = Yes\n",
    "# 1 = No\n",
    "# 2 = Unknown"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Model Name | Variables | n obs | df |R-squared\n",
    "-----------|-----------|-------|----|------------\n",
    "m1 | gf2_ss + age_test_gf2 + age_test_aaps +  male + white + etiology_impact | 52 | 6 | 0.782\n",
    "m2 | gf2_ss + age_test_gf2 + age_test_aaps | 55 | 3 | 0.771\n",
    "m3 | bs(gf2_ss, 3) + age_test_gf2 + age_test_aaps | 55 | 5 | 0.781\n",
    "m4 | cr(gf2_ss, df=3, constraints='center') + age_test_gf2 + age_test_aaps | 55 | 5 | 0.780\n",
    "m5 | gf2_ss + age_test_gf2 + age_test_aaps + time + onset_1 | 45 | 5 | 0.877\n",
    "m6 | gf2_ss + age_test_gf2 + age_test_aaps + time | 52 | 4 | 0.819"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>aaps_ss</td>     <th>  R-squared:         </th> <td>   0.782</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.753</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   26.85</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sun, 17 Apr 2016</td> <th>  Prob (F-statistic):</th> <td>2.44e-13</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:46:54</td>     <th>  Log-Likelihood:    </th> <td> -158.30</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    52</td>      <th>  AIC:               </th> <td>   330.6</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    45</td>      <th>  BIC:               </th> <td>   344.3</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     6</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "         <td></td>            <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>       <td>   53.1924</td> <td>    5.154</td> <td>   10.320</td> <td> 0.000</td> <td>   42.811    63.574</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>gf2_ss</th>          <td>    0.4439</td> <td>    0.038</td> <td>   11.787</td> <td> 0.000</td> <td>    0.368     0.520</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_gf2</th>    <td>   -0.3120</td> <td>    0.147</td> <td>   -2.125</td> <td> 0.039</td> <td>   -0.608    -0.016</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_aaps</th>   <td>    0.2512</td> <td>    0.134</td> <td>    1.881</td> <td> 0.066</td> <td>   -0.018     0.520</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>male</th>            <td>    0.5687</td> <td>    1.645</td> <td>    0.346</td> <td> 0.731</td> <td>   -2.745     3.882</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>white</th>           <td>    1.7043</td> <td>    1.638</td> <td>    1.041</td> <td> 0.304</td> <td>   -1.594     5.003</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>etiology_impact</th> <td>   -2.0195</td> <td>    1.702</td> <td>   -1.187</td> <td> 0.242</td> <td>   -5.447     1.408</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 2.316</td> <th>  Durbin-Watson:     </th> <td>   2.265</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.314</td> <th>  Jarque-Bera (JB):  </th> <td>   1.555</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.403</td> <th>  Prob(JB):          </th> <td>   0.460</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.263</td> <th>  Cond. No.          </th> <td>    880.</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                aaps_ss   R-squared:                       0.782\n",
       "Model:                            OLS   Adj. R-squared:                  0.753\n",
       "Method:                 Least Squares   F-statistic:                     26.85\n",
       "Date:                Sun, 17 Apr 2016   Prob (F-statistic):           2.44e-13\n",
       "Time:                        18:46:54   Log-Likelihood:                -158.30\n",
       "No. Observations:                  52   AIC:                             330.6\n",
       "Df Residuals:                      45   BIC:                             344.3\n",
       "Df Model:                           6                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================\n",
       "                      coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "-----------------------------------------------------------------------------------\n",
       "Intercept          53.1924      5.154     10.320      0.000        42.811    63.574\n",
       "gf2_ss              0.4439      0.038     11.787      0.000         0.368     0.520\n",
       "age_test_gf2       -0.3120      0.147     -2.125      0.039        -0.608    -0.016\n",
       "age_test_aaps       0.2512      0.134      1.881      0.066        -0.018     0.520\n",
       "male                0.5687      1.645      0.346      0.731        -2.745     3.882\n",
       "white               1.7043      1.638      1.041      0.304        -1.594     5.003\n",
       "etiology_impact    -2.0195      1.702     -1.187      0.242        -5.447     1.408\n",
       "==============================================================================\n",
       "Omnibus:                        2.316   Durbin-Watson:                   2.265\n",
       "Prob(Omnibus):                  0.314   Jarque-Bera (JB):                1.555\n",
       "Skew:                          -0.403   Prob(JB):                        0.460\n",
       "Kurtosis:                       3.263   Cond. No.                         880.\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m1 = smf.ols(formula=\"\"\"aaps_ss ~ gf2_ss + age_test_gf2 + age_test_aaps + \n",
    "                        male + white + etiology_impact\"\"\", \n",
    "             data=both).fit()\n",
    "m1.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>aaps_ss</td>     <th>  R-squared:         </th> <td>   0.771</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.757</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   57.12</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sun, 17 Apr 2016</td> <th>  Prob (F-statistic):</th> <td>2.52e-16</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:46:54</td>     <th>  Log-Likelihood:    </th> <td> -168.24</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    55</td>      <th>  AIC:               </th> <td>   344.5</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    51</td>      <th>  BIC:               </th> <td>   352.5</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     3</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td></td>           <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>     <td>   54.8203</td> <td>    4.437</td> <td>   12.354</td> <td> 0.000</td> <td>   45.912    63.729</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>gf2_ss</th>        <td>    0.4339</td> <td>    0.034</td> <td>   12.623</td> <td> 0.000</td> <td>    0.365     0.503</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_gf2</th>  <td>   -0.3497</td> <td>    0.135</td> <td>   -2.588</td> <td> 0.013</td> <td>   -0.621    -0.078</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_aaps</th> <td>    0.2834</td> <td>    0.124</td> <td>    2.280</td> <td> 0.027</td> <td>    0.034     0.533</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 2.865</td> <th>  Durbin-Watson:     </th> <td>   2.272</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.239</td> <th>  Jarque-Bera (JB):  </th> <td>   2.027</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.442</td> <th>  Prob(JB):          </th> <td>   0.363</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.320</td> <th>  Cond. No.          </th> <td>    798.</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                aaps_ss   R-squared:                       0.771\n",
       "Model:                            OLS   Adj. R-squared:                  0.757\n",
       "Method:                 Least Squares   F-statistic:                     57.12\n",
       "Date:                Sun, 17 Apr 2016   Prob (F-statistic):           2.52e-16\n",
       "Time:                        18:46:54   Log-Likelihood:                -168.24\n",
       "No. Observations:                  55   AIC:                             344.5\n",
       "Df Residuals:                      51   BIC:                             352.5\n",
       "Df Model:                           3                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "=================================================================================\n",
       "                    coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "---------------------------------------------------------------------------------\n",
       "Intercept        54.8203      4.437     12.354      0.000        45.912    63.729\n",
       "gf2_ss            0.4339      0.034     12.623      0.000         0.365     0.503\n",
       "age_test_gf2     -0.3497      0.135     -2.588      0.013        -0.621    -0.078\n",
       "age_test_aaps     0.2834      0.124      2.280      0.027         0.034     0.533\n",
       "==============================================================================\n",
       "Omnibus:                        2.865   Durbin-Watson:                   2.272\n",
       "Prob(Omnibus):                  0.239   Jarque-Bera (JB):                2.027\n",
       "Skew:                          -0.442   Prob(JB):                        0.363\n",
       "Kurtosis:                       3.320   Cond. No.                         798.\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m2 = smf.ols(formula=\"\"\"aaps_ss ~ gf2_ss + age_test_gf2 + age_test_aaps\"\"\", \n",
    "             data=both).fit()\n",
    "m2.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Use to print anova tables (for when we use splines)\n",
    "import statsmodels.api as sm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>aaps_ss</td>     <th>  R-squared:         </th> <td>   0.781</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.759</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   35.01</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sun, 17 Apr 2016</td> <th>  Prob (F-statistic):</th> <td>4.62e-15</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:46:54</td>     <th>  Log-Likelihood:    </th> <td> -166.93</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    55</td>      <th>  AIC:               </th> <td>   345.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    49</td>      <th>  BIC:               </th> <td>   357.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     5</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "          <td></td>            <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>        <td>   75.4249</td> <td>    4.181</td> <td>   18.040</td> <td> 0.000</td> <td>   67.023    83.827</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bs(gf2_ss, 3)[0]</th> <td>   -1.3268</td> <td>    7.914</td> <td>   -0.168</td> <td> 0.868</td> <td>  -17.230    14.576</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bs(gf2_ss, 3)[1]</th> <td>   22.9797</td> <td>    5.155</td> <td>    4.458</td> <td> 0.000</td> <td>   12.620    33.339</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bs(gf2_ss, 3)[2]</th> <td>   27.7088</td> <td>    4.170</td> <td>    6.645</td> <td> 0.000</td> <td>   19.330    36.088</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_gf2</th>     <td>   -0.3629</td> <td>    0.136</td> <td>   -2.671</td> <td> 0.010</td> <td>   -0.636    -0.090</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_aaps</th>    <td>    0.2951</td> <td>    0.126</td> <td>    2.342</td> <td> 0.023</td> <td>    0.042     0.548</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 3.445</td> <th>  Durbin-Watson:     </th> <td>   2.254</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.179</td> <th>  Jarque-Bera (JB):  </th> <td>   2.473</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.464</td> <th>  Prob(JB):          </th> <td>   0.290</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.465</td> <th>  Cond. No.          </th> <td>1.30e+03</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                aaps_ss   R-squared:                       0.781\n",
       "Model:                            OLS   Adj. R-squared:                  0.759\n",
       "Method:                 Least Squares   F-statistic:                     35.01\n",
       "Date:                Sun, 17 Apr 2016   Prob (F-statistic):           4.62e-15\n",
       "Time:                        18:46:54   Log-Likelihood:                -166.93\n",
       "No. Observations:                  55   AIC:                             345.9\n",
       "Df Residuals:                      49   BIC:                             357.9\n",
       "Df Model:                           5                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "====================================================================================\n",
       "                       coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "------------------------------------------------------------------------------------\n",
       "Intercept           75.4249      4.181     18.040      0.000        67.023    83.827\n",
       "bs(gf2_ss, 3)[0]    -1.3268      7.914     -0.168      0.868       -17.230    14.576\n",
       "bs(gf2_ss, 3)[1]    22.9797      5.155      4.458      0.000        12.620    33.339\n",
       "bs(gf2_ss, 3)[2]    27.7088      4.170      6.645      0.000        19.330    36.088\n",
       "age_test_gf2        -0.3629      0.136     -2.671      0.010        -0.636    -0.090\n",
       "age_test_aaps        0.2951      0.126      2.342      0.023         0.042     0.548\n",
       "==============================================================================\n",
       "Omnibus:                        3.445   Durbin-Watson:                   2.254\n",
       "Prob(Omnibus):                  0.179   Jarque-Bera (JB):                2.473\n",
       "Skew:                          -0.464   Prob(JB):                        0.290\n",
       "Kurtosis:                       3.465   Cond. No.                     1.30e+03\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 1.3e+03. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# bs() B-spline\n",
    "m3 = smf.ols(formula=\"\"\"aaps_ss ~ bs(gf2_ss, 3) + age_test_gf2 + age_test_aaps\"\"\", \n",
    "             data=both).fit()\n",
    "m3.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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>sum_sq</th>\n",
       "      <th>df</th>\n",
       "      <th>F</th>\n",
       "      <th>PR(&gt;F)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>bs(gf2_ss, 3)</th>\n",
       "      <td>4634.360674</td>\n",
       "      <td>3</td>\n",
       "      <td>54.312079</td>\n",
       "      <td>1.308633e-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>age_test_gf2</th>\n",
       "      <td>202.877916</td>\n",
       "      <td>1</td>\n",
       "      <td>7.132842</td>\n",
       "      <td>1.024204e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>age_test_aaps</th>\n",
       "      <td>156.014750</td>\n",
       "      <td>1</td>\n",
       "      <td>5.485213</td>\n",
       "      <td>2.328615e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Residual</th>\n",
       "      <td>1393.696551</td>\n",
       "      <td>49</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    sum_sq  df          F        PR(>F)\n",
       "bs(gf2_ss, 3)  4634.360674   3  54.312079  1.308633e-15\n",
       "age_test_gf2    202.877916   1   7.132842  1.024204e-02\n",
       "age_test_aaps   156.014750   1   5.485213  2.328615e-02\n",
       "Residual       1393.696551  49        NaN           NaN"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sm.stats.anova_lm(m3, typ=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>aaps_ss</td>     <th>  R-squared:         </th> <td>   0.780</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.757</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   34.68</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sun, 17 Apr 2016</td> <th>  Prob (F-statistic):</th> <td>5.52e-15</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:46:54</td>     <th>  Log-Likelihood:    </th> <td> -167.14</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    55</td>      <th>  AIC:               </th> <td>   346.3</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    49</td>      <th>  BIC:               </th> <td>   358.3</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     5</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                      <td></td>                         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>                                 <td>   90.0725</td> <td>    3.236</td> <td>   27.838</td> <td> 0.000</td> <td>   83.570    96.575</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>cr(gf2_ss, df=3, constraints='center')[0]</th> <td>   -0.6435</td> <td>    1.915</td> <td>   -0.336</td> <td> 0.738</td> <td>   -4.491     3.204</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>cr(gf2_ss, df=3, constraints='center')[1]</th> <td>   14.6513</td> <td>    1.804</td> <td>    8.121</td> <td> 0.000</td> <td>   11.026    18.277</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>cr(gf2_ss, df=3, constraints='center')[2]</th> <td>   16.6059</td> <td>    2.660</td> <td>    6.243</td> <td> 0.000</td> <td>   11.261    21.951</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_gf2</th>                              <td>   -0.3580</td> <td>    0.137</td> <td>   -2.621</td> <td> 0.012</td> <td>   -0.633    -0.083</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_aaps</th>                             <td>    0.2940</td> <td>    0.127</td> <td>    2.321</td> <td> 0.025</td> <td>    0.039     0.549</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 2.991</td> <th>  Durbin-Watson:     </th> <td>   2.239</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.224</td> <th>  Jarque-Bera (JB):  </th> <td>   2.083</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.432</td> <th>  Prob(JB):          </th> <td>   0.353</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.404</td> <th>  Cond. No.          </th> <td>    460.</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                aaps_ss   R-squared:                       0.780\n",
       "Model:                            OLS   Adj. R-squared:                  0.757\n",
       "Method:                 Least Squares   F-statistic:                     34.68\n",
       "Date:                Sun, 17 Apr 2016   Prob (F-statistic):           5.52e-15\n",
       "Time:                        18:46:54   Log-Likelihood:                -167.14\n",
       "No. Observations:                  55   AIC:                             346.3\n",
       "Df Residuals:                      49   BIC:                             358.3\n",
       "Df Model:                           5                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "=============================================================================================================\n",
       "                                                coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "-------------------------------------------------------------------------------------------------------------\n",
       "Intercept                                    90.0725      3.236     27.838      0.000        83.570    96.575\n",
       "cr(gf2_ss, df=3, constraints='center')[0]    -0.6435      1.915     -0.336      0.738        -4.491     3.204\n",
       "cr(gf2_ss, df=3, constraints='center')[1]    14.6513      1.804      8.121      0.000        11.026    18.277\n",
       "cr(gf2_ss, df=3, constraints='center')[2]    16.6059      2.660      6.243      0.000        11.261    21.951\n",
       "age_test_gf2                                 -0.3580      0.137     -2.621      0.012        -0.633    -0.083\n",
       "age_test_aaps                                 0.2940      0.127      2.321      0.025         0.039     0.549\n",
       "==============================================================================\n",
       "Omnibus:                        2.991   Durbin-Watson:                   2.239\n",
       "Prob(Omnibus):                  0.224   Jarque-Bera (JB):                2.083\n",
       "Skew:                          -0.432   Prob(JB):                        0.353\n",
       "Kurtosis:                       3.404   Cond. No.                         460.\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# cr() natural cubic spline\n",
    "m4 = smf.ols(formula=\"\"\"aaps_ss ~ cr(gf2_ss, df=3, constraints='center') + age_test_gf2 + age_test_aaps\"\"\", \n",
    "             data=both).fit()\n",
    "m4.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false,
    "scrolled": 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>sum_sq</th>\n",
       "      <th>df</th>\n",
       "      <th>F</th>\n",
       "      <th>PR(&gt;F)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>cr(gf2_ss, df=3, constraints='center')</th>\n",
       "      <td>4624.005992</td>\n",
       "      <td>3</td>\n",
       "      <td>53.791079</td>\n",
       "      <td>1.567175e-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>age_test_gf2</th>\n",
       "      <td>196.774953</td>\n",
       "      <td>1</td>\n",
       "      <td>6.867251</td>\n",
       "      <td>1.165756e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>age_test_aaps</th>\n",
       "      <td>154.307597</td>\n",
       "      <td>1</td>\n",
       "      <td>5.385183</td>\n",
       "      <td>2.451412e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Residual</th>\n",
       "      <td>1404.051233</td>\n",
       "      <td>49</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             sum_sq  df          F  \\\n",
       "cr(gf2_ss, df=3, constraints='center')  4624.005992   3  53.791079   \n",
       "age_test_gf2                             196.774953   1   6.867251   \n",
       "age_test_aaps                            154.307597   1   5.385183   \n",
       "Residual                                1404.051233  49        NaN   \n",
       "\n",
       "                                              PR(>F)  \n",
       "cr(gf2_ss, df=3, constraints='center')  1.567175e-15  \n",
       "age_test_gf2                            1.165756e-02  \n",
       "age_test_aaps                           2.451412e-02  \n",
       "Residual                                         NaN  "
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sm.stats.anova_lm(m4, typ=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>aaps_ss</td>     <th>  R-squared:         </th> <td>   0.877</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.862</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   55.82</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sun, 17 Apr 2016</td> <th>  Prob (F-statistic):</th> <td>9.93e-17</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:46:54</td>     <th>  Log-Likelihood:    </th> <td> -122.00</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    45</td>      <th>  AIC:               </th> <td>   256.0</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    39</td>      <th>  BIC:               </th> <td>   266.8</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     5</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td></td>           <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>     <td>   63.0834</td> <td>    3.795</td> <td>   16.622</td> <td> 0.000</td> <td>   55.407    70.760</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>gf2_ss</th>        <td>    0.3897</td> <td>    0.029</td> <td>   13.285</td> <td> 0.000</td> <td>    0.330     0.449</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_gf2</th>  <td>   -0.5494</td> <td>    0.113</td> <td>   -4.871</td> <td> 0.000</td> <td>   -0.778    -0.321</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_aaps</th> <td>    0.3424</td> <td>    0.103</td> <td>    3.327</td> <td> 0.002</td> <td>    0.134     0.551</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>time</th>          <td>    1.8769</td> <td>    0.523</td> <td>    3.590</td> <td> 0.001</td> <td>    0.819     2.934</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>onset_1</th>       <td>    0.0459</td> <td>    0.049</td> <td>    0.941</td> <td> 0.353</td> <td>   -0.053     0.144</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 0.832</td> <th>  Durbin-Watson:     </th> <td>   2.356</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.660</td> <th>  Jarque-Bera (JB):  </th> <td>   0.209</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.008</td> <th>  Prob(JB):          </th> <td>   0.901</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.333</td> <th>  Cond. No.          </th> <td>    847.</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                aaps_ss   R-squared:                       0.877\n",
       "Model:                            OLS   Adj. R-squared:                  0.862\n",
       "Method:                 Least Squares   F-statistic:                     55.82\n",
       "Date:                Sun, 17 Apr 2016   Prob (F-statistic):           9.93e-17\n",
       "Time:                        18:46:54   Log-Likelihood:                -122.00\n",
       "No. Observations:                  45   AIC:                             256.0\n",
       "Df Residuals:                      39   BIC:                             266.8\n",
       "Df Model:                           5                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "=================================================================================\n",
       "                    coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "---------------------------------------------------------------------------------\n",
       "Intercept        63.0834      3.795     16.622      0.000        55.407    70.760\n",
       "gf2_ss            0.3897      0.029     13.285      0.000         0.330     0.449\n",
       "age_test_gf2     -0.5494      0.113     -4.871      0.000        -0.778    -0.321\n",
       "age_test_aaps     0.3424      0.103      3.327      0.002         0.134     0.551\n",
       "time              1.8769      0.523      3.590      0.001         0.819     2.934\n",
       "onset_1           0.0459      0.049      0.941      0.353        -0.053     0.144\n",
       "==============================================================================\n",
       "Omnibus:                        0.832   Durbin-Watson:                   2.356\n",
       "Prob(Omnibus):                  0.660   Jarque-Bera (JB):                0.209\n",
       "Skew:                          -0.008   Prob(JB):                        0.901\n",
       "Kurtosis:                       3.333   Cond. No.                         847.\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m5 = smf.ols(formula=\"\"\"aaps_ss ~ gf2_ss + age_test_gf2 + age_test_aaps + \n",
    "                        time + onset_1\"\"\", \n",
    "             data=both).fit()\n",
    "m5.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>aaps_ss</td>     <th>  R-squared:         </th> <td>   0.819</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.804</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   54.32</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sun, 17 Apr 2016</td> <th>  Prob (F-statistic):</th> <td>3.13e-17</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:46:54</td>     <th>  Log-Likelihood:    </th> <td> -155.96</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    53</td>      <th>  AIC:               </th> <td>   321.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    48</td>      <th>  BIC:               </th> <td>   331.8</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     4</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td></td>           <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>     <td>   58.6568</td> <td>    4.169</td> <td>   14.069</td> <td> 0.000</td> <td>   50.274    67.040</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>gf2_ss</th>        <td>    0.3976</td> <td>    0.033</td> <td>   12.071</td> <td> 0.000</td> <td>    0.331     0.464</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_gf2</th>  <td>   -0.4447</td> <td>    0.127</td> <td>   -3.510</td> <td> 0.001</td> <td>   -0.699    -0.190</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_test_aaps</th> <td>    0.3094</td> <td>    0.114</td> <td>    2.719</td> <td> 0.009</td> <td>    0.081     0.538</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>time</th>          <td>    1.5959</td> <td>    0.539</td> <td>    2.963</td> <td> 0.005</td> <td>    0.513     2.679</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 5.319</td> <th>  Durbin-Watson:     </th> <td>   2.411</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.070</td> <th>  Jarque-Bera (JB):  </th> <td>   4.242</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.632</td> <th>  Prob(JB):          </th> <td>   0.120</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.567</td> <th>  Cond. No.          </th> <td>    819.</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                aaps_ss   R-squared:                       0.819\n",
       "Model:                            OLS   Adj. R-squared:                  0.804\n",
       "Method:                 Least Squares   F-statistic:                     54.32\n",
       "Date:                Sun, 17 Apr 2016   Prob (F-statistic):           3.13e-17\n",
       "Time:                        18:46:54   Log-Likelihood:                -155.96\n",
       "No. Observations:                  53   AIC:                             321.9\n",
       "Df Residuals:                      48   BIC:                             331.8\n",
       "Df Model:                           4                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "=================================================================================\n",
       "                    coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
       "---------------------------------------------------------------------------------\n",
       "Intercept        58.6568      4.169     14.069      0.000        50.274    67.040\n",
       "gf2_ss            0.3976      0.033     12.071      0.000         0.331     0.464\n",
       "age_test_gf2     -0.4447      0.127     -3.510      0.001        -0.699    -0.190\n",
       "age_test_aaps     0.3094      0.114      2.719      0.009         0.081     0.538\n",
       "time              1.5959      0.539      2.963      0.005         0.513     2.679\n",
       "==============================================================================\n",
       "Omnibus:                        5.319   Durbin-Watson:                   2.411\n",
       "Prob(Omnibus):                  0.070   Jarque-Bera (JB):                4.242\n",
       "Skew:                          -0.632   Prob(JB):                        0.120\n",
       "Kurtosis:                       3.567   Cond. No.                         819.\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m6 = smf.ols(formula=\"\"\"aaps_ss ~ gf2_ss + age_test_gf2 + age_test_aaps + time\"\"\", \n",
    "             data=both).fit()\n",
    "m6.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plots"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Seaborn Jointplot: https://stanford.edu/~mwaskom/software/seaborn/generated/seaborn.jointplot.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Peasron Corelation: 0.860\n",
      "Spearman Corelation: 0.848\n"
     ]
    }
   ],
   "source": [
    "import scipy\n",
    "# (correlation, p-value)\n",
    "pc = scipy.stats.pearsonr(both.gf2_ss, both.aaps_ss)\n",
    "print(\"Peasron Corelation: \" \"%.3f\" % pc[0])\n",
    "sc = scipy.stats.spearmanr(both.gf2_ss, both.aaps_ss)\n",
    "print(\"Spearman Corelation: \" \"%.3f\" % sc[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWEAAAFjCAYAAADsN7rcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nFhExkUJYRMRE/wApRuTplj7IigAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10d0bb8d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lmplot(\"gf2_ss\", \"aaps_ss\", data=both, fit_reg=False)\n",
    "sns.regplot(both.gf2_ss, both.aaps_ss, scatter=False, lowess=True, label='LOWESS')\n",
    "plt.plot([50, 110], [50, 110], 'red', label='Identity');\n",
    "legend = plt.legend(loc='upper left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x10f7d9f10>]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAawAAAGqCAYAAABXkXCxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nggUA2AWCBQDYBYIFANiFN1VukS+NF6LXAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10d40b650>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# pearson correlation\n",
    "g = sns.jointplot(x=\"gf2_ss\", y=\"aaps_ss\", data=both, marginal_kws=dict(bins=15, rug=True))\n",
    "    # for spearman correlation, add 'stat_func=spearmanr' to joinplot()\n",
    "x0, x1 = g.ax_joint.get_xlim()\n",
    "y0, y1 = g.ax_joint.get_ylim()\n",
    "lims = [max(x0, y0)-10, min(x1, y1)+10]\n",
    "g.ax_joint.plot(lims, lims, ':k') \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAawAAAGqCAYAAABXkXCxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCZKwhBBCxIT/D6GTiBpL9pk0AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10d09df10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# regression line with confidence bands\n",
    "sns.jointplot(x=\"gf2_ss\", y=\"aaps_ss\", data=both, kind=\"reg\", marginal_kws=dict(bins=15, rug=True));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1kAAAQoCAYAAAD42BTZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10d0afc10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(12,15))\n",
    "fig = sm.graphics.plot_regress_exog(m1, 'gf2_ss', fig=fig)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x10ec3e6d0>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfwAAAGKCAYAAADpKfPPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ebb3e50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8,6))\n",
    "ax.plot(np.array(both.aaps_ss), m3.fittedvalues, 'r.')\n",
    "plt.plot([65, 105], [65, 105], 'blue', label='Identity')\n",
    "legend = plt.legend(loc='upper left')\n",
    "plt.xlabel('Observed Arizona Score')\n",
    "plt.ylabel('Predicted Arizona Score')\n",
    "plt.title('Model 3')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60, 110)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfwAAAGKCAYAAADpKfPPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMcZYN8ANnzHGGOsGuOEzxhhj3cD/AJ47/GJ3cLx/AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ec74f50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8,6))\n",
    "ax.plot(np.array(both.gf2_ss), m3.fittedvalues, 'r.', label='Predicted')\n",
    "ax.plot(np.array(both.gf2_ss), np.array(both.aaps_ss), 'g.', label='Observed')\n",
    "plt.plot([35, 110], [35, 110], 'blue', label='Identity');\n",
    "legend = plt.legend(loc='upper left')\n",
    "plt.xlabel('Goldman-Fristoe Score')\n",
    "plt.ylabel('Arizona Score')\n",
    "plt.title('Model 3')\n",
    "ax.set_xlim([30,120])\n",
    "ax.set_ylim([60,110])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "aaps_m6 = both[both['time'].notnull()].aaps_ss\n",
    "gf2_m6 = both[both['time'].notnull()].gf2_ss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x10f10d190>"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfwAAAGKCAYAAADpKfPPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ec927d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8,6))\n",
    "ax.plot(aaps_m6, m6.fittedvalues, 'r.')\n",
    "plt.plot([65, 105], [65, 105], 'blue', label='Identity')\n",
    "legend = plt.legend(loc='upper left')\n",
    "plt.xlabel('Observed Arizona Score')\n",
    "plt.ylabel('Predicted Arizona Score')\n",
    "plt.title('Model 6')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60, 110)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfwAAAGKCAYAAADpKfPPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3X18k+XZ//FPkqZJa4FKW4qCKDIRpohDZIAiikWhoMgq\nYz8nE9RSuBWo4njQKqKgoLgqKA8pihPdjUyEyqwglVuYyoMMQV03halVqNLe1A5sm7Zp8vujNx0o\n0KckV3Ll+369fJmkyXUdZ1J65DqP88Hi8/l8iIiIiKlZjQ5AREREAk8JX0REJAIo4YuIiEQAJXwR\nEZEIoIQvIiISAZTwRUREIoASvojJHThwgG7dunHrrbf+5GczZ86kW7dulJWVNemYGRkZrF279rTP\n2bFjBzfccMNJf/bZZ58xZswYRo4cSVpaGn//+9+bdH4RaTolfJEI4HA4+OqrrygqKqp/rKKigr/9\n7W9YLJYmH89isTTrdQCVlZXccccdpKens3btWv7rv/6Le++9t1nHEpHGizI6ABEJPKvVSmpqKuvX\nrycjIwOAt99+m2uvvZYVK1bUP+/VV1/l5Zdfxmq1kpiYyIMPPsh5553HoUOHmDFjBiUlJbRv357v\nv/++/jX/+te/eOyxx/j+++/xer2MGTOGtLS0U8by/vvvc+6553LVVVcBMGjQIDp27BiglovIMbrC\nF4kQI0aM4I033qi/n5uby69+9av6+9u2beP555/npZdeIjc3l+HDh3PXXXcB8Mgjj/CLX/yCv/zl\nL8yaNYsvv/wSAI/Hw+TJk5k6dSqvv/46K1eu5IUXXmDv3r2njOPLL78kISGBBx54gLS0NG6//XZq\na2sD1GoROUYJXyRCXHTRRVitVv7+97/z7bffUl5ezgUXXACAz+fjr3/9K6mpqZx55pkAjBw5kkOH\nDnHgwAG2bdvGyJEjATjnnHO44oorAPjqq6/45ptvuP/++7npppsYM2YMVVVV/OMf/zhll7/H42Hr\n1q2MHj2aNWvWcOuttzJ+/HhqamqC8C6IRC516YtEkBtvvJE33niDtm3bMmLEiBN+5vP5+PHWGj6f\nD4/Hg8ViOeFnNpsNAK/XS+vWrVm3bl39z0pKSmjdujV79uw5aQzJycmcf/75XHLJJQBce+21ZGVl\n8c0333D++ef7pZ0i8lO6wheJIDfeeCNvvfUWeXl5J4ygt1gsDBgwgLfeeovS0lIA1qxZw5lnnsm5\n557LgAEDePXVVwH47rvv2LZtGwCdO3cmOjq6vlTw7bffMmLECAoKCk4Zw1VXXcXBgwfrR+Z/+OGH\nWK1W1fFFAkxX+CIR4Fj3enJyMj/72c9o1aoVrVu3PuFn/fv357bbbuO2227D5/PRtm1bli1bhsVi\n4aGHHuL+++8nNTWV9u3b061bNwDsdjuLFy9m7ty5LF++HI/Hw5QpU/jFL37Bjh07ThpLYmIizz33\nHLNnz6ayspLo6GgWLVpEdHR0EN4JkchlCdT2uDNnzmTLli0kJCSwfv16AN566y2effZZvvjiC157\n7TUuuuii+ucvW7aMNWvWYLVaycrK4sorrwxEWCIiIhEpYF36aWlpLF++/ITHunbtyrPPPkvv3r1P\neHz//v3k5eXx5ptvsnz5cmbPno3X6w1UaCIiIhEnYAm/d+/e9V2Gx3Tp0oXOnTv/5LnvvPMOw4YN\nw26307FjRzp16sTHH38cqNBEREQiTkgM2isuLqZ9+/b199u3b8+hQ4cMjEhERMRcQiLhn0xzl+0U\nERGRnwqJhJ+cnMx3331Xf/+7774jOTn5tK8J0FhDERExkfffh6goOOcc8HiMjsZYhk3LOz5hDxo0\niKlTpzJ27FgOHTpEYWFh/aIcp2KxWCgpORroMAMuKamV2hEizNAGMEc7zNAGUDuMVloKo0efgc9n\n4U9/svD99+HXhh9LSmrV7NcGLOHfe++97Ny5k7KyMgYOHMikSZOIj4/n0Ucf5fvvvycjI4Pu3buz\nfPlyfvaznzF06FCGDRuGzWZj1qxZ6tIXEZFm8/kgM9PJwYNWZsyoYsAAByUlRkdlrIDNww+GcPzG\n+WPh+s35x8zQDjO0AczRDjO0AdQOI7lcdrKynAwY4GH16kratw+/NpxMS67wQ6KGLyIi4i979liZ\nPdtBYqKXxYvd/N/WDxFPCV9EREzjyBFIT4/B44HFi90kJ4dtJ7bfKeGLiIgp+HwwdaqTwkIrkydX\nc/XVtUaHFFKU8EVExBRWrrSTm2unTx8P06dXGx1OyFHCFxGRsFdQYCUry8GZZ/pYtsxNlPaC/Qkl\nfD+66qo+jBt3C7/73WgefHAGVVXuZh9r7tyHeffddwCYP38OX3315Smf+9FHf+PTT5u+98DNN9/A\nkSP/bnaMIiKhoLwc0tOduN0Wnnmmkg4dVLc/GSV8P3I4nKxY8SdeeulV7HY769atOeHnniYs82Sx\nWOrXIpg+PYvzzvvppkPH7N69i08+aXrC11oHImIGM2c62bfPRkZGNUOGqG5/Kur0CJBLLrmUf/1r\nPx999DdycpbQunVrvv66kJdf/jNLlixiz56/UV1dw223jWHQoFR8Ph/Z2U+wa9dO2rVLxm631x/r\n7rvHc/fd99CtW3e2b/8Al2sxXq+X+Ph4Zsx4kDfeeB2r1cbbb+dxzz3TOOecc3nqqcc5dKhuueLJ\nk6fSo0dP/v3vMh5++AH+939LuPjiS7Q8sYiEvdWro1i1ys6ll9by4INVRocT0iI+4cenpgBQlpfv\nt2N6PB62b3+fvn2vAGDfvs9YuXI17dufRW7u68TFxZGT8xLV1dVMnjye7t0v5fPP/8k333zNK6+8\nxuHDh7n11lEMHz4C+M/V/vfff88TT8xl8eLltG9/FkePHqVVq1aMGJFGbGwsv/nNrQA8/PAD/PrX\nt3DJJZfy3Xffcd99k3j55T+zYkUOPXv+grFj72Tbtvf4y19y/dZmEZFg27/fwrRpTuLifCxbVkl0\ntNERhbaITvjxqSnYd+2sv93SpF9dXcW4cbcA0LNnL4YPH8HHH++he/eLaN/+LAA+/HA7//rX/vr6\nvNtdyYEDX7N370cMHjwEi8VCYmIil13W+4Rj+3w+/v73T7j00l71x2rVqtVxP//Pc3ft2klh4X9q\n/hUVFVRWVrJ370c89tgCAPr1u5JWrVq3qL0iIkZxu+HOO2OoqLCQk1NJ587qsWxIRCd8f4uOdrBi\nxZ9+8rjTGXPC/Xvvncbll/cF/rNk5bZt7zfYxd74mrsPl+uPJ5QF6n+ibnwRMYFZsxwUFNgYM6aa\nESMifBu8RoroQXtlefnU9O5DTe8+fu3SP50+ffrx+uuv1Q/g+/LLL3G73fTs2Yt33tmE1+vlf//3\nf9m9+28nvM5isXDRRT3Ys2c3335bBFA/wj42NpaKivL6515+eV/+/OdV9ff37fscqOt12LRpAwDb\ntr3P0aNHAtdQEZEAWb8+ihUrounevZY5c1S3b6yIv8L3Z6I/2RV4Xf39P/dvuOEmvv22iDvuuBWf\nz0e7dkk88sh8Bg68ht27P+TWW0eRnNyeHj1+uj1wfHw806Y9wAMP/B6v10fbtm35wx+e5YorriIr\nazrvvbeFe+6ZRmbmffzhD/O57bb/R21tLZde2ov77pvB7ben8/DDDzBmzK+5+OKe9aUBEZFwUVho\nITPTSWysj5wcNzExDb9G6mi3PIOF4y5UJ2OGdpihDWCOdpihDaB2+Ft1Ndx4Yyy7d9tYuLCS3/ym\n8V35odKGltJueSIiYnpz5zrYvdvGqFE1jB6tun1TKeGLiEjI27TJxpIl0XTp4mX+fDdaN6zplPBF\nRCSkFRVZmDTJicPhIyenkrg4oyMKTxE/aE9EREKXxwMTJjgpLbUyf76biy/2Gh1S2NIVvoiIhKwF\nC6LZvj2K4cNrGDu2xuhwwpoSvoiIhKStW21kZ0fTqZOX7GzV7VtKCd/PiosPMWPGvfzmN79i9Oib\neOaZp/B4POTlrSc7+wmjw/uJwYMHGB2CiMhPFBdbmDjRic0GLlclbdoYHVH4U8L3I5/PxwMP/J6B\nAwexatXr/Pd/v05lZQUu13MB2Yq2ttYf20DqK7OIhBavF+66y0lJiZWsrCp69VLd3h80aM+P/va3\nD3E4nAwdOhwAq9XK5Mn3MmrUjdx550SKiw8xaVIGJSUlXH/9UMaNS6eiooLf/34KJSUleL213Hbb\nnVx77WD++c9/8Oyz2VRWVtKmTTwPPDCLhIRE7r57PF27XsjHH+/liisG8Oabb/DnP7+BxWKhsrKS\n3/72Zv785zf47rtv+cMfnqCs7HucTifTpz9Ap07nUVR0kNmzs3C7K7niiqsMfsdERH5q0aJotmyJ\nYvBgDxMmqG7vLxGf8FPX1G2Pm5fW8iV2v/zyCy68sNsJj8XGnkFycntqaz0UFPydlStX43A4SE//\nHf36XUlFxfckJrbjySefAaC8/Ac8Hg9PP/0k8+f/gTZt4nnnnbdxuRYzc+ZDWCwWPB4Py5e/BMDn\nn/+Tjz76G7169eaDD/7KL3/ZH5vNxhNPzOX3v7+fjh3P4e9//5SnnprPM88s4ZlnFvCrX43i+utT\nef31P7e4zSIi/rRjh41586I56ywvCxe6saof2m8iOuGnrklh16Gd9bdbmvQb6rW//PJf0rp13Za0\nAwcO4uOP9zBs2HU8/vg8lixZRP/+A+jZ81K++GI/X375LzIz/wsAr9dLQkJS/XGuvfa6+tuDBg1m\n8+ZN9OrVm/z8t0lL+zUVFRV88snHPPjg9Prn1dTUrUr16acf12+Re/31Q1myZFGL2iwi4i+lpXVT\n8Hw+WLbMTUJC2K78HpIiOuH723nnnc+7724+4bHy8h84dOg7bLaoE+r4Pp8Pq9XCeeedxwsvvMK2\nbe+Rk7OY3r37cNVVV9O5cxeWLn3hpOc5frvdK664CpdrMUeOHOHzz//JZZddTkVFOa1atTrpVr0i\nIqHI54PMTCcHD1qZMaOKvn39MUZJjhfRnSV5afn0Tu5D7+Q+funS7927D263mw0b3gTqBtU9++zT\npKbeiNPp5MMPd3DkyBGqqtz89a9b6NHjUoqLi4mOjua664by//7fGD7//DM6dTqPsrLv+fTTTwDw\neDx8+eUXJz1nbGws3br9nGeeeZIrrhiAxWLhjDPiOPvss/mf/6lrk8/nY//+fQD06NGTd955G4C3\n397Q4jaLiPhDTo6dDRvsDBjgYcqUaqPDMaWIv8L3R6I/3mOPPclTT83jxRefx+fz0q/flYwf/1/k\n52+ke/eLyMqaRnFxMUOGpHLhhd347LO9PPbYPKxWC1FRUdx33/1ERUXx6KPzeeaZBfzwww/U1noY\nPfoWOnc+/6TnvPbawTz00EwWLVpW/9hDD81hwYJ5/PGPL+DxeEhJuY6f/ewCpky5j9mzs3jllT9y\n5ZUDAzJ7QESkKfbssTJ7toPERC+LF7ux2YyOyJy0Pa7BzLRlY7i3wwxtAHO0wwxtALWjMY4cgWuv\nPYOvv7bw6quVXH11YLryzfRZNFdEd+mLiIhxfD6YOtVJYaGVyZOrA5bspY4SvoiIGGLlSju5uXb6\n9PEwfbrq9oGmhC8iIkFXUGAlK8vBmWf6WLbMTVTEjygLPL3FIiISVOXlkJ7uxO224HJV0KFD2A4l\nCyu6whcRkaCaOdPJvn02MjKqGTJEdftgUcIXEZGgWb06ilWr7PTsWUtWVpXR4UQUJXw/OtVWs3Pn\nPsy7777TrGPu2/c527a9X3//vfe28vLLLwKwdeu7fPXVl806rohIsO3fb2HaNCdxcT5crkocDqMj\niixK+H518kVsLBZLsxe42bfvM7Zv/0/Cv/LKq7j11rEA/PWv7/LVVydfgU9EJJS43XDnnTFUVFjI\nznbTubPq9sGmQXsB4PP5yM5+gl27dtKuXTJ2u51j6xv9eNvbP/zhScDJ3XeP56KLerB79y5++OEo\nM2Y8xEUXXczy5Uuprq7m44/3cOut46iqcvPZZ/9g8OAhvP/+X9mz5yNeeukFHn10Pg8+OIMXXngZ\ngG+++ZpZs+6vvy8iYqRZsxwUFNgYM6aaESM8RocTkXSFHwBbt/4P33zzNa+88hpZWY/wyScf129r\n+/TTTzJ37hM8//xKhg27gezsbKCuF8Dr9ZKT80cmT57KihUuoqKiSE+fSErKdaxY8SeuvXZwfU/B\nxRdfwpVXXsXdd0/hhRdeoUOHjsTFxbFv3+cA5OWtZ9iwGw17D0REjlm/PooVK6Lp3r2WOXNUtzeK\naa/wH37Ywfr1/m3eDTd4ePjhhn9Z9+z5iMGDh2CxWEhMTOSyy3oD8PXXX/1k29uzzmpf/7qBA68B\n4MILu/Hdd98Cdb0Fp1v9+PifDR9+E3l565k06R42b95ETs5LTW+kiIgfFRZayMx0EhvrIyfHTUxM\nw6+RwDBtwjeSxcIpk/SPt709fn1nuz0aAKvVRm1t46aqHD824OqrB7FihYvLLutNt27dad26dXOb\nICLSYtXVkJERw9GjFhYurKRrV6/RIUU00yb8hx+uatTVeCD07NmL3NzXGTp0OKWlpeze/Teuu27o\nCdveXnxxDzweD/v376dNm+RTHuuMM86goqKi/v7xXyRiY2MpLy+vvx8dHc0vf9mPBQvmMXPmQ4Fp\nnIhII82d62D3bhujRtUwerTq9kZTDd+Pjl1tDxx4Deeccw633jqKuXNn0aPHJQD1294uXbqIsWNv\nYdy4W/joo49OdTQAfvGL3nz11ReMG3cL77yz6YQR/9deex1/+tNKbr/9VoqKDgKQkjIEq9VKnz59\nA9tYEZHT2LTJxpIl0XTp4mX+fDfaidt42h7XYP7esvFPf1pJZWUFd9yR4bdjNoYZtp40QxvAHO0w\nQxsgcttRVGRh0KBYysstvPVWBRdfbHxXvpk+i+YybZd+JJo58z6+/baIhQuXGB2KiEQojwcmTHBS\nWmpl/nx3SCR7qaOEbyKPP77A6BBEJMItWBDN9u1RDB9ew9ixNUaHI8dRDV9ERPxi61Yb2dnRdOrk\nJTtbdftQo4QvIiItVlxsYeJEJzYbuFyVtGljdETyY0r4IiICQOqaFFLXpDT5dV4v9B/1d0pKrGRl\nVdGrl+r2oUgJX0RESF2Twq5DO9l1aGeTk/5lGSs58o9fwgV/YX3S1YEJUFpMCV9ERJptxw4bB9eP\nh1YH4KaxWKxhO9Pb9DRKX0REyEvLr7+yz0vLb9RrSkvrpuBZsXBB+qO0Or9Lo18rwaeELyIiQOMT\nPYDPB5mZTg4etDJjRhX33qtpwaFOXfoiItJkOTl2NmywM2CAhylTqo0ORxpBCV9ERJpkzx4rs2c7\nSEz0snixG5vN6IikMZTwRUSk0Y4cgfT0GDweWLzYTXKyBumFCyV8ERFpFJ8Ppk51UlhoZfLkaq6+\nutbokKQJlPBFRKRRVq60k5trp08fD9Onq24fbgKW8GfOnEn//v254YYb6h8rKytj3LhxXH/99dx+\n++0cOXKk/mfLli3juuuuY8iQIbz33nuBCktERJqhoMBKVpaDM8/0sWyZmyjN8Qo7AUv4aWlpLF++\n/ITHXC4X/fv3Z+PGjfTt2xeXywXA/v37ycvL480332T58uXMnj0br1dLM4qIhILyckhPd+J2W3jm\nmUo6dFDdPhwFLOH37t2b1q1bn/DY5s2bGTlyJAAjR44kP79uzuc777zDsGHDsNvtdOzYkU6dOvHx\nxx8HKjQREWmCu++GfftsZGRUM2SIeev28akpxKc2fS+BcBHUGv7hw4dJTEwEIDExkcOHDwNQXFxM\n+/bt65/Xvn17Dh06FMzQRERCWnM3tmmpX057ihdfhJ49a8nKqjrl84yKz1/iU1MY2HMnA3vuNG3S\nN2zQnsViwXKazZJP9zMRkUjSko1tWmLQc7fz5Sv3QfQRatNG4XCEVnz+dFX/AradA9vOqbttRkEd\ndpGQkEBJSQlJSUkUFxfTtm1bAJKTk/nuu+/qn/fdd9+RnJzc4PGSkloFLNZgUjtChxnaAOZohxna\nAP5ph91uO+F2MN4btxu+eP4xqImDm39Nq7O/O+V5jYivOU4Xl+WSHnBgW/3tUG1DSwQ14Q8aNIi1\na9cyfvx41q1bR0pKSv3jU6dOZezYsRw6dIjCwkIuueSSBo9XUnI00CEHXFJSK7UjRJihDWCOdpih\nDeC/duTeuLH+yjn3xo1BeW+mT3dQceACEq9cywVDDpz2vEbE11QNfRZbltYysOf/3f5LLSU3hl4b\noGVfIAOW8O+991527txJWVkZAwcOZPLkyYwfP57MzEzWrFlDhw4dePrppwH42c9+xtChQxk2bBg2\nm41Zs2apS19E5DjB3IVu/fooVqyIpnv3Wja8kkKnTiMbTOJm2CXvg+fr/l/T29g4AsXi8/nCdn5F\nKH6LbCpdyYQOM7QBzNEOM7QBwrMdhYUWBg06g9paePvtCrp29YZlO36sMW04NlivLC90v7yE5BW+\niIiEl+pqyMiI4ehRCwsXVtK1a2SthxLKid4ftLSuiEiY8vdUuLlzHezebWPUqBpGj/YYGosRzNCG\n01HCFxEJQ/6eCrdpk40lS6Lp0sXL/PlumjKMygzT8szQhoYo4YuIRLiiIguTJjlxOHy4XJXExRkd\nkQSCEr6ISBjKS8und3Ifeif3adEIeY8HJkxwUlpq5ZFHqujRo+l1+7y0fPqWxtG3NC5sR+v76/0M\nZRq0JyISpvyRmBYsiGb79iiGD69h7NiaZh0jPjWFbbt+AKDmg5SwHfxm1kR/jK7wRUQi1NatNrKz\no+nUyUt2dtPq9hJ+dIUvIhKBiostTJzoxGYDl6uSNm2af6yyvPywmMMe6ZTwRUQijNcLd93lpKTE\nysMPu+nVq+Xz7ZXoQ5+69EVEIsyiRdFs2RLF4MEeJkxoXt1ewo8SvohIBNmxw8a8edGcdZaXhQvd\nWI/LAvGpKabdC16U8EVEIkZpad0UPJ8Pli1zk5Dwn61U4lNTsO/aiX3XTiV9k1LCFxGJAD4fZGY6\nOXjQyrRp1fTtW2t0SBJkGrQnIhIBcnLsbNhgZ8AAD1OmVP/k5xppb35K+CJiGsfWQDf7AipNNfAP\nd/PPJ5eTmOhl8WI3NtvJn9f/jrr/5wUvtJBnpt8pdemLiClEwuYnzXHdyyP4x7KH8NXaOPOWKSQn\n+076PL1/P2W290QJX0TEpHw+KHxlJnzfBa6cR5vuO0/53Kh/FJz0tpiHuvRFxBTy0vJN1f3qDytX\n2vn+b4OJ67KXC3/zl9O+L1s/+DkDe9Z9Idiy9+eUBSvIEGa23yklfBExDTP8UfaXggIrWVkOzjzT\nx+bXutChw8bTPr8sL58tGrT3E2b6nVLCFxExmfJySE934nZbcLkq6NDh5HX7HztZoo9PTQG7DXJP\n/4VBQp9q+CIiJjNzppN9+2xkZFQzZEjz59sfW4yHbdu0GI8JKOGLiJjI6tVRrFplp2fPWrKyqowO\nR0KIuvRFRExi/34L06Y5iYvz4XJV4nC07HjHFuOx222UqUs/7Cnhi4iYgNsNd94ZQ0WFhZycSjp3\nblzdviFlefkkJbWCkqOnfZ5W6Qt96tIXETGBWbMcFBTYGDOmmhEjPEE9tzbeCQ9K+CIiYW79+ihW\nrIime/da5sxR3d6fzLRlsBK+iEgYKyy0kJnpJDbWR06Om5iY4MdQlpeP94w4vGfEmapL32w9F0r4\nIiJhKnbIEP7r6q84etTCvHluunb1GhJHfGoK1vIfsJb/YIrEaFYatCciEobiU1OYuXsUH3Ixv03I\nY/ToAUaHZDpm2zJYCV9EJAzllV3BU9xHVz5jUacn8FiMS/hlefkkdD67/raZmKk96tIXEQkzRUUW\nbi9dgMNSxcs/fxTPxvWGxqMu/fCghC8iEkY8HpgwwUlpqZVH5vk4790lRockYUJd+iIiYWTBgmi2\nb49i+PAaxo6t8euxm1uvNlut26yU8EVEwsTWrTays6Pp1MlLdrYbi8V/x67fKOf/bjcn6UtoU8IX\nETFQ6pq6K+OG9l0vLrYwcaITmw1crkratGn8axur/x11/9+y1y+H83t80jKq4YuIGCR1TQq7Du1k\n16Gd9cnxZLxeuOsuJyUlVrKyqujVy9vo1zZW/ztg2zl1/x1L/C3h7/ik5ZTwRURC3KJF0WzZEsXg\nwR4mTPBv3V4ih8Xn8/lnSyUDlDSwe1M4SEpqpXaECDO0AczRDjO0ARrXjoa6vXfssHHTTTG0a+dj\n8+YKEhJ8jX5tU53qeI1px8kG7YVSl76ZfqeaSzV8EREDnS4ZlpbWTcHz+WDpUvcJyb6h1/o7ltM5\n1YC/D56v+3lZml/CO+l5QQMGG0td+iIiIcjng8xMJwcPWpk2rZp+/WqNDqlJTrbxjD93njPbxjbB\noCt8EZEQlJNjZ8MGOwMGeJgypdrocE6rMfPwWzrtT1pOCV9EJMTs2WNl9mwHiYleFi92Y7MZHVHD\njo3sz/u/+z/+EuDvq3At9tN0SvgiIiHkyBFIT4/B44HFi90kJ4f+uOpjU/CO3T42FuD4RByIBK1E\n3zRK+CIiIcLng6lTnRQWWpkypYqrrz593T7crnDDJU6z0qA9EZEQsXKlndxcO336eJg+/fR1+1Aa\ntJaXlk/v5D70Tu4TElPw5OR0hS8i0gSBuqouKLCSleXgzDN9LFvmJupHf51D/WpeiT706QpfRKSR\nAnVVXV4O6elO3G4LzzxTSYcOJ9btT3besrx8anr3oaZ3n5D9EiChRVf4IiIGmznTyb59NjIyqhky\npPHz7ZXopSmU8EVEGikQI81Xr45i1So7PXvWkpVVFbTzSuRRwhcRaQJ/Jtz9+y1Mm+YkLs6Hy1WJ\nwxGc80pkUsIXETGA2w133hlDRYWFnJxKOncO/fn2Et40aE9ExACzZjkoKLAxZkw1I0Z4jA5HIoAS\nvohIkK1fH8WKFdF0717LnDknr9uL+JsSvohICzVlF7jCQguZmU5iY33k5LiJiQlwcCL/RwlfRKQF\n4lNTGNhzJwN7Njw3v7oaMjJiOHrUwrx5brp29Tb6PKlrUkhdE7rbwIZ6fKKELyLSJAmdzyah89n1\n96/qX8Bt6aa3AAAgAElEQVS2c2DbOXW3T2fuXAe7d9sYNaqG0aMbX7c/tjnNrkM7QzKphnp8UkcJ\nX0SkkRI6n421/Aes5T/UJ31P95/X//z42z+2aZONJUui6dLFy/z5biyWgIdbryklBzEvTcsTEWmB\nvLT8+qvaU60nX1RkYdIkJw5H3Xz7uDj/n+NUji3Le+x2IObztyQ+CR5DEv4f//hHXnvtNXw+H6NG\njeK2226jrKyMe+65h6KiIjp06MDTTz9N69atjQhPROSkDn9ZVH9lf/jLovrHT5fkPB6YMMFJaamV\n+fPd9OjR+Lr98UI9kYZ6fGJAl/7nn3/Oa6+9xmuvvUZubi7vvvsuX3/9NS6Xi/79+7Nx40b69u2L\ny+UKdmgiIg06/GXRCcm+IQsWRLN9exTDh9cwdmxNACM7OW2yI8cEPeF/8cUXXHLJJTgcDmw2G5df\nfjkbN25k8+bNjBw5EoCRI0eSn69fTBEJb++8A9nZ0XTq5CU7O7h1++OV5eUr2UvwE/4FF1zArl27\nKCsro7Kykq1bt3Lo0CEOHz5MYmIiAImJiRw+fDjYoYlIgEXS1K3iYgu//S3YbOByVdKmjdERSaQL\neg2/S5cupKenc/vttxMbG0u3bt2wWk/83mGxWLAY9VVYRALi2NStY7fNXPP1euGuu5wcOgQPP1xF\nr17Nq9uL+FODCb+srIwFCxZQWFjIM888w5NPPsmMGTNo04KvqzfffDM333wzANnZ2SQnJ5OQkEBJ\nSQlJSUkUFxfTtm3bBo+TlNSq2TGEErUjdJihDRCa7bDbbSfcbijGkGhD//51///ggya97PHHYcsW\nGDYMHnzQidXqDEBwwRUSn0cLmaENLdFgwn/wwQe54oor2Lt3L2eccQbt2rXj97//fYsG1R0+fJiE\nhASKiop4++23Wb16NQcOHGDt2rWMHz+edevWkZLScLdfScnRZscQKpKSWqkdIcIMbYDQbUfujRvr\nu/Nzb9x42hib2wZ/7hl//HS2mst/2ehj7thh48EHYzjrLB8vvmjl8OHQ+yyaKlR/p5rCDG2Aln1p\naTDhHzhwgN/85jesWrUKh8PBPffcww033NDsEwJMnjyZsrIyoqKimDVrFq1atWL8+PFkZmayZs2a\n+ml5ImIugezGD8Z884aUltZNwfP5YOlSN4mJsZSUBD0MkZNqMOFHRUVx9Oh/vhV99dVX2Gy207yi\nYa+88spPHouPj+fFF19s0XFFRPylLC+/ST0GPh9kZjo5eNDKjBlV9OtX69d4/Nl7IZGpwYQ/adIk\nxowZw7fffsvEiRPZs2cPjz32WDBiExFptKYm6MYes7Fycuxs2GBnwAAPU6ZU++X8x4RC74WEvwYT\nfrt27XjhhRfYu3cvXq+XRx55hKSkpGDEJiLSJEYlwj17rMye7SAx0cvixW5a2AkqEhANJvzMzEw2\nbNjANddcE4x4RETCypEjkJ4eg8cDixe7SU72+f0cgei9kMjTYMK/4IILePbZZ+nZsydO53+mllx+\n+eUBDUxEJNT5fDB1qpPCQitTplRx9dX+rdsfT4leWqpR8/B37NjBjh07Tnh85cqVAQtKRCQcrFxp\nJzfXTp8+HqZP92/dXsTfGkz4xxL7Dz/8QG1tbYsW3BERMYuCAitZWQ7i430sXeomSpuNS4hr8Ff0\n66+/5t577+Xrr7/G5/PRoUMHsrOz6dy5czDiExEJOeXlkJ7uxO224HJV0LGj/+v2Iv7W4OY5Dz30\nEHfeeSc7d+7kww8/ZPz48Tz00EPBiE1EJOR0zjmbC9Jy2bfPRkZGNUOGBK5ubzaRtHlSKGow4X//\n/fcMGTKk/n5qaiplZWUBDUpEJBR1zjmb8l034dl9K9azd5OVVWV0SGHj2OZJuw7tVNI3SIMJ3+Fw\n8Omnn9bf/+STT4iJiQloUCIiochbcgH8ZQlEH8Hxm7E4HEZHJNJ4Ddbw77//fiZPnlw/WK+srIzs\n7OyAByYiEkrcbuj89ocU1Nhw/OZ3FM543+iQwkpeWn79lb2Zt0YOZQ0m/EsvvZSNGzfy5Zdf1g/a\ni4uLC0ZsIiIhY9YsBwUFNsaMqeapp54zOpywpERvrAa79PPy8vjVr35F165dcTqdpKamkp+vD01E\nIsf69VGsWBFN9+61zJnj37p9fGpK/Sp6IoHUYMJfsmQJK1asAODcc89l7dq1LFy4MOCBiYiEgsJC\nC5mZTmJjfeTkuPHnEKZjm+LYd+1U0peAa7BLv6amhsTExPr7CQkJAQ1IRCJHqNd0q6shIyOGo0ct\nLFxYSdeuXqND8ptQf+8bwwxtCKYGE36vXr249957ueGGG/D5fLz11ltceumlwYhNREzs2DStY7dD\n8Y/23LkOdu+2MWpUDaNHe/x+fKM2xQmH974hZmhDsDWY8GfNmsXKlSt59dVXiYqKonfv3txyyy3B\niE1ExDCbNtlYsiSaLl28zJ/vxmJp2uvjU1PAboPcjad9njbFkWCx+Hy+Rq0JWVtbS0FBAeeeey6t\nW7cOdFyNUlJy1OgQWiwpqZXaESLM0AYIr3acqkvW6DYUFVkYNCiW8nILeXkV9OjRtK78Y7V5gJre\nfUIyqTelO9zoz+NUzNCGpkpKatXs157yCr+wsJB77rmHyZMn079/f2655RZKS0vxer0sWLCA3r17\nN/ukIiIQmrVXjwcmTHBSWmpl/nx3k5N9uPjg+br/l6UZG0dLhOLvTyg75Sj9Rx99lDvuuIOBAweS\nm5tLRUUFb7/9Nq+88goLFiwIZowiIkGzYEE027dHMXx4DWPH1jTrGGV5+dT07gP9+oXk1b1mB0Sm\nU17hHzp0iGHDhgHwwQcfcP311xMVFUWHDh04ejT8u0VERH5s61Yb2dnRdOrkJTu76XX745Xl5dd1\nvwa5G9mIQYASHhqch+/1etm+fTv9+/cHwOfzUVlZGfDARESCqbjYwsSJTmw2cLkq+b/VxMNKY6/c\nj/VAhOr4AgmMU17hd+3aFZfLRVVVFQ6Hg8suu4zq6mpeeOEFTcsTEVPxeuGuu5yUlFh5+GE3vXqZ\ns25/PCX6yHPKK/xZs2Zx8OBBPvvsM5577jmsVitz5szh/fffZ+bMmcGMUUQkoBYtimbLligGD/Yw\nYULz6vahQFfucjqNnpYXiswyxULtCA1maAOYox3BbMOOHTZuuimGdu18bN5cQUKC//4kmuGzAHO0\nwwxtgJZNy2uwhi8iYlalpXVT8Hw+WLrU7ddkLxJqlPBFJCL5fJCZ6eTgQSvTplUz9NFrNEVNTE0J\nX0SAulXLjq1cdrrHGvvaUJeTY2fDBjsDBniYtelq08xLD8fPQoKjwbX0d+3axfLly6msrMTr9eL1\nevn222/ZvHlzMOITkSA42UYkjd2cJBw3Mdmzx8rs2Q4SE70sXuzGNs4co/LD8bOQ4GnwCv+BBx5g\n8ODB1NbWcuutt3Luuedy2223BSM2EQmSqH8UnPS2GR05AunpMXg8sHixm+Rkn0a3S0RoMOE7nU7S\n0tK4/PLLad26NXPmzGHjxtPv/iQi4WXrBz+n3zfQ75u621C3Tnnv5D70Tu5z2ivFxj4vFPh8MHWq\nk8JCK5MnV3P11bX1PyvLyw/7ZB9On4UEX4Nd+k6nk7KyMjp37szevXvp27cvpaWlwYhNRIKkLC+f\nLSdZkrWxSSNcksvKlXZyc+306eNh+vRqo8MJiHD5LCT4Gkz4Y8eOJTMzk2effZa0tDTeeOMNLrro\nomDEJiJBdLKrWzOty15QYCUry0F8vI+lS91ENfjXT8RcGvyVHzp0KEOGDMFisbB27Vq++uorunXr\nFozYRMRAx+/pHp+aEtZJv7wc0tOduN0WXK4KOnbUfHuJPA0m/AMHDvDKK69QVlZ2wuOPP/54wIIS\nEfGnmTOd7NtnIyOjmiFDaht+gYgJNZjwMzMzufzyy7n88svrH7O0ZM9IEQkLZXn5pujSX706ilWr\n7PTsWUtWVpXR4YgYpsGEX1tby/Tp04MRi4iEmMYm+pN9MWjMl4VAf6HYv9/CtGlO4uJ8uFyVOBwB\nOY1IWGhwWt5ll13GO++8Q3W1OUe0ikjLnGwP9sbsy97Yvduby+2GO++MoaLCQna2m86dVbeXyNbg\nFf6GDRt4+eWXT3jMYrHwj3/8I2BBiYi01KxZDgoKbIwZU82IER6jwxExXIMJ/7333gtGHCISpk5W\n629M/T+QYwTWr49ixYpounevZc4c1e1FoBEJv6KigmeffZbt27fj8Xjo27cvmZmZxMbGBiM+EQkD\nJ0vYjUnigajdFxZayMx0EhvrIyfHTUyM308hEpYarOE/+uijuN1uHnvsMebPn09NTQ2zZs0KRmwi\nYjLxqSkB3Y2uuhoyMmI4etTCvHluunY1x6Y4Iv7Q4BX+p59+yvr16+vvz5o1i6FDhwY0KBExn2As\n5DN3roPdu22MGlXD6NGq24scr8ErfIB///vfJ9yO0pqUIhJiNm2ysWRJNF26eJk/342WCxE5UaPW\n0h81ahSDBg3C5/OxefNmxo8fH4zYRMREAjlIr6jIwqRJThyOuvn2cXF+PbyIKTSY8NPS0rj44ov5\n8MMP8fl8PPvss1x44YXBiE1ETCYQ3fgeD0yY4KS01Mr8+W569FDdXuRkGuzSnzBhAjExMdx6662M\nGTOGCy+8kNtuuy0YsYmINGjBgmi2b49i+PAaxo6tMTockZDVYMLfs2cPd9xxB1u3bq1/7PiavoiE\nn9Q1KaSuCdxo+WDZutVGdnY0nTp5yc4OTt3eLO+dRJ4GE3779u15/vnnefLJJ1m2bBmgzXNEwlnq\nmhR2HdrJrkM7wzpxFRdbmDjRic0GLlclbdoE/pxmee8kMjVqlH6nTp347//+bz766CMmT56Mz6c1\nqUXEOF4v3HWXk5ISK1lZVfTqpbq9SEMaTPjx8fEAxMXFsWTJEs477zz++c9/BjwwEQmMvLR8eif3\noXdyH/LSwnPb20WLotmyJYrBgz1MmBC8ur0Z3juJXBZfMy7Xi4uLadeuXSDiaZKSkqNGh9BiSUmt\n1I4QYYY2gDnacbo27Nhh46abYmjXzsfmzRUkJIRuj6MZPgswRzvM0Aaoa0dznXJa3vjx43G5XAwa\nNOgnP7NYLLzzzjvNPqmISHOUltZNwfP5YOlSd0gne5FQc8qE/+ijjwLw9NNP07Zt26AFJCJyMj4f\nZGY6OXjQyowZVfTrV2t0SCJh5ZQJPzk5GYBp06axYcOGoAUkIuZwbBS7v2rdOTl2NmywM2CAhylT\nqv1yzOP5O16RUNPgoL3u3buzbt06vvjiC4qKiur/ExE5FX9PX9uzx8rs2Q4SE70sXuzGZvNDkMfR\ndDuJBA0urbt371727t17wmPV1dW89957AQtKROSYI0cgPT0GjwcWL3aTnKy6vUhzNJjwN2/eDEBN\nTQ1vv/02q1at4pNPPgl4YCISvvLS8v3SRe7zwdSpTgoLrUyZUsXVVwembu+veEVCWYMJ/5tvvmHV\nqlWsXbuWI0eOMGHCBJ555pkWnXTZsmW88cYbWK1WunbtyuOPP05FRQX33HMPRUVFdOjQgaeffprW\nrVu36DwiYhx/JM6VK+3k5trp08fD9On+r9sfT4lezO6UNfy3336b22+/nVGjRvHvf/+bJ598knbt\n2nH33Xe3aNT+gQMHWL16NWvXrmX9+vXU1tby5ptv4nK56N+/Pxs3bqRv3764XK5mn0MkFMWnptRv\nD2u0UIrlVAoKrGRlOYiP97F0qZuoBi9PROR0TpnwJ0+eTKtWrVi1ahVz5szhiiuu8MsJ4+LiiIqK\norKyEo/Hg9vtpl27dmzevJmRI0cCMHLkSPLz9W1bzCM+NQX7rp3Yd+00PNGGUiynUl4O6elO3G4L\nCxdW0rGj6vYiLXXK78xvvPEGr7/+Or/97W/p0KEDqamp1Na2vH4WHx/P7bffztVXX43T6eTKK6/k\niiuu4PDhwyQmJgKQmJjI4cOHW3wuEQlPd98N+/bZyMioZsgQzbcX8YdTXuF37dqVGTNmsGXLFsaP\nH8/OnTs5fPgw48eP59133232Cb/++mv++Mc/snnzZv76179SUVFBbm7uCc+xWCzakU9MpSwvn5re\nfajp3YeyPGN7r0IplpNZvTqKF1+Enj1rycqqMjocEdNo0lr6hw8frr/yX79+fbNOmJeXx/vvv8/c\nuXMBWLduHXv37mX79u289NJLJCUlUVxczO9+9zst+CMSYT77DC67DKxW+Ogj6NLF6IhEzKNJw2AS\nEhIYN24c48aNa/YJzz//fBYvXozb7cbhcLBt2zYuueQSYmJiWLt2LePHj2fdunWkpDRcWzTLRghq\nR2gwQxsgfNvhdsOvfhVLebmNV1+F1q2PUlJidFQtE66fxY+ZoR1maAMEaPOcQOnWrRsjRowgLS0N\nq9XKz3/+c379619TXl5OZmYma9asqZ+WJyKRY9YsBwUFNsaMqebXv44O+2QvEmqatT1uqDDLtzW1\nIzSYoQ0Qnu1Yvz6KO+6IoXv3WjZsqKBTp/Brw8mE42dxMmZohxnaAC27wm9wLX0RkUAqLLSQmekk\nNtZHTo6bmBijIxIxJy1lIUGjpUvlx6qrISMjhqNH6+bbd+3qNTokEdPSFb4EhXYjk5OZO9fB7t02\nRo2qYfRoj9HhiJiaEr6IGGLTJhtLlkTTpYuX+fPdaOkNkcBSl74EhXYjk+MVFVmYNMmJw+HD5aok\nLs7oiETMTwlfgkaJXgA8HpgwwUlpqZX589306KG6vUgwqEtfRIJqwYJotm+PYvjwGsaOrTE6HJGI\noYQvIkGzdauN7OxoOnXykp2tur1IMCnhi6mFw77vkaK42MLEiU5sNnC5KmnTxuiIRCKLEr6YVjjs\n++5v8akp0L+/0WH8hNcLd93lpKTESlZWFb16qW4vEmxK+CImcewLDtu2hdwXnEWLotmyJYrBgz1M\nmKC6vYgRNEpfTKssL78+8YXivu+RYscOG/PmRXPWWV4WLnRj1WWGiCGU8MXUIinRH/uCY7fbKMvd\naHQ4AJSW1k3B8/lg6VI3CQlhu1eXSNhTwhcxkbK8/LrdtEJgVzCfDzIznRw8aGXGjCr69as1OiSR\niKbONREJiJwcOxs22BkwwMOUKdVGhyMS8ZTwRcTv9uyxMnu2g8REL4sXu7HZjI5IRJTwRcSvjhyB\n9PQYamosPPecm+Rk1e1FQoESvoj4jc8HU6c6KSy0MmVKFddco7q9SKhQwpegiaRV7yKprcdbudJO\nbq6dPn08TJ+uur1IKFHCl6CIpFXvIqmtxysosJKV5SA+3sfSpW6iNAdIJKTon6SItFh5OaSnO3G7\nLbhcFXTsqLq9SKhRwo9AqWvqrjqDuT99JK16V5aXz41zzgbgjazgtjV1Td3CO7k3BnfhnZkznezb\nZyMjo5ohQ8xTt2/JvxWjPguRU1GXfoRJXZPCrkM72XVoZ/0fs2Apy8s3fbKHuvd4e9sf2N72h6C+\nx8c+220HtgX1vKtXR7FqlZ2ePWvJyqoK2nkDrSX/Voz6LERORwlfRJpt/34L06Y5iYvz4XJV4nAY\nHZGInIrF5/OFbbGtJASWD22ppKRWQW9HILr0jWiHv/mzDUaUTY6dN1jdyG43DBkSS0GBjZycSkaM\n8Pjt2KHy+6Qu/Tqh8nm0hBnaAHXtaC4lfIOZ6Zcw3NthhjZA8NoxfbqDFSuiGTOmmqee8m9Xvj6L\n0GKGdpihDdCyhK8ufRFpsvXro1ixIpru3WuZM8c8dXsRM1PCFwmS1DUpphjAVVhoITPTSWysj5wc\nNzExRkckIo2hhC8SBEbOjvCn6mrIyIjh6FEL8+a56drVa3RIItJISvgi0mhz5zrYvdvGqFE1jB7t\nv0F6IhJ4WnhHJAjy0vING7nvL5s22ViyJJouXbzMn+/GYjE6IhFpCiV8kSAJ10QPUFRkYdIkJw5H\n3Xz7uDijIxKRplLCF5HT8nhgwgQnpaVW5s9306OH6vYi4Ug1fBE5rQULotm+PYrhw2sYO7bG6HBE\npJmU8CNQpO7VLk23dauN7OxoOnXykp2tur1IOFPCjzCRule7NF1xsYWJE53YbOByVdKmjdERiUhL\nKOGLyE94vXDXXU5KSqxkZVXRq5fq9iLhToP2Ikwk7UsvzbdoUTRbtkQxeLCHCRNUtxcxAyX8CKRE\nL6ezY4eNefOiOessLwsXurGqH1DEFPRPWUTqlZbWTcHz+WDpUjcJCWG7maaI/IgSvogA4PNBZqaT\ngwetTJtWTb9+tUaHJCJ+pIRvIi3Zja2lr+3/fP9mvVZCR06OnQ0b7AwY4GHKlGqjwxERP1PCN4mW\n7Mbmj9duO7AtrHeBi3R79liZPdtBYqKXxYvd2GxGRyQi/qaELxLhjhyB9PQYamosPPecm+Rk1e1F\nzEij9E2iJbux+eO1druN3Bs3Num1YjyfD6ZOdVJYaGXKlCquuUZ1exGzUsI3kZbsxtbS1yYltaKk\n5GizjyHGWLnSTm6unT59PEyfrrq9iJmpS99EtEa+NEVBgZWsLAfx8T6WLnUTpa//IqamhG8SWiM/\ntIT6l6/yckhPd+J2W1i4sJKOHVW3FzE7JXwBQj9BhZNw+PI1c6aTfftsZGRUM2SI6vYikUCdeCbR\nkjXyjyWoY7e19K65rV4dxapVdnr2rCUrq8rocEQkSJTwTUSJOjSE8gZF+/dbmDbNSVycD5erEofD\n6IhEJFiU8CWkE1S4CsX30e2GO++MoaLCgstVSefOqtuLRBIlfAFCM0GJf82a5aCgwMaYMdXcdJPH\n6HBEJMg0aE8kAqxfH8WKFdF0717LnDmq24tEIiV8EZMrLLSQmekkNtZHTo6bmBijIxIRI6hLX8TE\nqqshIyOGo0fr5tt37eo1OiQRMYiu8EVMbO5cB7t32xg1qobRo1W3F4lkSvjSYvGpKdC/v9FhyI9s\n2mRjyZJounTxMn++G4vF6IhExEhB79L/4osvuPfee+vvf/PNN0yZMoUbb7yRe+65h6KiIjp06MDT\nTz9N69atgx2eNJEW7QlNRUUWJk1y4nDUzbePizM6IhExWtCv8M8//3zWrVvHunXreP3114mJiWHw\n4MG4XC769+/Pxo0b6du3Ly6XK9ihiZiCxwMTJjgpLbXyyCNV9Oihur2IGNyl/8EHH9CpUyfOOuss\nNm/ezMiRIwEYOXIk+fm6UgwHZXn51PTuA/366eo+RDzyCGzfHsXw4TWMHVtjdDgiEiIMHaX/5ptv\nMmzYMAAOHz5MYmIiAImJiRw+fNjI0KQJyvLySUpqBSVHg3bO1DV1KwPmpYXmlwyj4tu61cacOdCp\nk5fsbNXtReQ/DLvCr66u5n/+538YOnToT35msViw6C+VnELqmhR2HdrJrkM76xNrKDEqvuJiCxMn\nOrHZwOWqpE2boJ1aRMKAYVf4W7du5aKLLqJt27YAJCQkUFJSQlJSEsXFxfWPn05SUqtAhxkUakfT\n2O22E27787z+OFYg4zsVrxd++1soKYEFC+D6688I+DkDTf8uQosZ2mGGNrSExefzGbKDxj333MNV\nV11VX7d/4okniI+PZ/z48bhcLo4cOcJ999132mOUBLELOVCSklqpHc1w45yzAXgjq8hvx/RnG4Ld\npf/MM9HMnetg8GAPGzZEcfhweP9O6d9FaDFDO8zQBmjZlxZDrvArKir44IMPePTRR+sfGz9+PJmZ\nmaxZs6Z+Wp7IycSnprBt1w8A1HwQmlMBg1m737HDxrx50Zx1lpeFC91YrZqDJyI/ZUjCj42NZceO\nHSc8Fh8fz4svvmhEOKambW/NrbS0bgqezwdLl7pJSNCWtyJyclpL38TMuihOWV6+vsgAPh9kZjo5\neNDKjBlV9OtXa3RIIhLClPANlLomBbvdRu6NG40OJew0NtGH+vS9lsjJsbNhg50BAzxMmVJtdDgi\nEuK0lr5Bjk3d2nZgW8Cmbh1bFKemd5+IvBIO9el7LbFnj5XZsx0kJnpZvNiNzdbwa0QksukK3+Qi\nMdGb3ZEjkJ4eQ02NheeeqyQ5WXV7EWmYEr5B8tLy1aUfYMfe42O3zcDng6lTnRQWWpkypYprrlHd\nXkQaRwnfQHlp+aaZGxqqzJLoj1m50k5urp0+fTxMn666vYg0nmr4BtI+8tIUBQVWsrIcxMf7WLrU\nTZS+rotIEyjhG6R+yty2bfVTzEROpbwc0tOduN0WFi6spGNH1e1FpGmU8EXCwMyZTvbts5GRUc2Q\nIarbi0jTqVPQIMcWj7HbbZTlatCenNrq1VGsWmWnZ89asrKqjA5HRMKUEr6BjNhHXsLL/v0Wpk1z\nEhfnw+WqxOEwOiIRCVdK+CIhyu2GO++MoaLCgstVSefOqtuLSPOphi8SombNclBQYGPMmGpuuslj\ndDgiEuaU8EVC0Pr1UaxYEU337rXMmaO6vYi0nBJ+mEpdk9Ko9eEb+zwJHYWFFjIzncTG+sjJcRMT\nY3REImIGSvhhqLGbwph58xizqq6GjIwYjh61MG+em65dvUaHJCImoYQvEkLmznWwe7eNUaNqGD1a\ndXsR8R+N0g9Djd0Uxoybx5jZpk02liyJpksXL/Pnu7FYjI5IRMxECT9MNTaBK9GHh6IiC5MmOXE4\n6ubbx8UZHZGImI0SvojBPB6YMMFJaamVefPc9Oihur2I+J9q+CIGW7Agmu3boxg+vIZx42qMDkdE\nTEoJ30Cpa1Lo/7y2x41kW7fayM6OplMnL9nZqtuLSOAo4Rvk2JS5bQe2acpchCoutjBxohObDVyu\nStq0MToiETEzJXwRA3i9cNddTkpKrGRlVdGrl+r2IhJYGrRnkGNT5ux2G7k3anvcSLNoUTRbtkQx\neLCHCRNUtxeRwFPCN1BeWt32uCXaHjei7NhhY968aM46y8vChW6s6mcTkSDQn5owFZ+aQnyqav/h\nprS0bgqezwdLl7pJSNCWtyISHEr4YSg+NQX7rp3Yd+1U0g8jPh9kZjo5eNDKtGnV9OtXa3RIIhJB\nlEW/kKIAABMoSURBVPBFgiQnx86GDXYGDPAwZUq10eGISIRRDT8MleXl11/Zl+Vp6dxwsGePldmz\nHSQmelm82I3NZnREIhJplPDDlBJ9+DhyBNLTY6ipsfDcc5UkJ6tuLyLBpy59kQDy+WDqVCeFhVam\nTKnimmtUtxcRYyjhiwTQypV2cnPt9OnjYfp01e1FxDhK+CIBUlBgJSvLQXy8j6VL3USpgCYiBtKf\nIJEAKC+H9HQnbrcFl6uCjh1VtxcRY+kKXyQAZs50sm+fjYyMaoYMUd1eRIynhC/iZ6tXR7FqlZ2e\nPWvJyqoyOhwREUAJX8Sv9u+3MG2ak7g4Hy5XJQ6H0RGJiNRRwjdQfGoK9O9vdBjiJ2433HlnDBUV\nFv7wBzedO6tuLyKhQwnfIMfWw2fbNq2HbxKzZjkoKLAxZkw1N93kMTocEZETKOGL+MH69VGsWBFN\n9+61zJmjur2IhB5NyzPIsfXw7XYbZbkbjQ5HWqCw0EJmppPYWB85OW5iYoyOSETkp5TwDVSWl09S\nUisoOWp0KNJM1dWQkRHD0aMWFi6spGtXr9EhiYiclLr0RVpg7lwHu3fbGDWqhtGjVbcXkdClhC/S\nTJs22ViyJJouXbzMn+/GYjE6IhGRU1PCF2mGoiILkyY5cTjq5tvHxRkdkYjI6amGL9JEHg9MmOCk\ntNTKvHluevRQ3V5EQp+u8EWaaMGCaLZvj2L48BrGjasxOhwRkUZRwhdpgq1bbWRnR9Opk5fsbNXt\nRSR8KOGLNFJxsYWJE53YbOByVdKmjdERiYg0nhK+SCN4vXDXXU5KSqxkZVXRq5fq9iISXpTwRRph\n0aJotmyJYvBgDxMmqG4vIuFHCV+kATt22Jg3L5qzzvKycKEbq/7ViEgY0p8ukdMoLa2bgufzwdKl\nbhIStOWtiIQnJXyRU/D5IDPTycGDVn7/+2r69as1OiQRkWZTwhc5hZwcOxs22BkwwENmZrXR4YiI\ntIgSvshJ7NljZfZsB4mJXhYvdmOzGR2RiEjLKOGL/MiRI5CeHkNNjYXnnnOTnKy6vYiEP0MS/pEj\nR5g8eTJDhw4lNTWVvXv3UlZWxrhx47j++uu5/fbbOXLkiBGhSYTz+WDqVCeFhVamTKnimmtUtxcR\nczAk4c+dO5errrqKt956izfeeIPzzz8fl8tF//792bhxI3379sXlchkRmkS4lSvt5Oba6dPHw/Tp\nqtuLiHkEPeEfPXqUXbt2cfPNNwMQFRVFq1at2Lx5MyNHjgRg5MiR5OfnBzs0iXCffAJZWQ7i430s\nXeomSntJioiJBP1P2oEDB2jbti0zZ87kn//8JxdddBH3338/hw8fJjExEYDExEQOHz4c7NAkgpWX\nw69/DW63BZergo4dVbcXEXMJ+hW+5/+3d+9BUdVtHMC/uyyJDiiKCg5mJZkJJDCa7ohKrKgIrAIK\nU4p5Ga8jYuKoiyBCrhqDWcmMAt6SsIsKLhmllShoiOiUOmh4D1DEG2Doctuzz/sHL2fkomVe4MDz\nmdkZ9+zZs8+Xoz7nwv5+BgPOnz+PDz74APv27UPHjh2bXL6XyWSQ8TRk7CUKCzNDfj4wZ04NPD35\nvj1jrO156Wf4NjY2sLa2xsCBAwEAY8eORWJiIrp37447d+6gR48euH37Nrp16/aP2+rRw+JFl/tS\ncI6W9803dQ/glf8/pE3K+6JeW8gAcI7WpC1keBYv/Qy/R48e6NWrF65duwYAOH78ON588024u7tj\n3759AACdTgcPD4+XXRpjjDHWZsmI6KXfrMzPz0d4eDhqa2vRp08frFu3DoIg4KOPPsLNmzdha2uL\nzz//HJ07d37ZpTHGGGNtUos0fMYYY4y9XDzSHmOMMdYOcMNnjDHG2gFu+Iwxxlg7IImGX11djYCA\nAEyYMAFeXl749NNPAUCS4+8LggBfX1/MmzcPgDQzqFQqqNVq+Pr6iiMmSjGH1Od0uHr1Knx9fcXH\noEGDkJSUJKkM9RISEuDt7Q21Wo0lS5agpqZGcjl27twJtVoNHx8f7Ny5E4A0/l2EhYVh2LBhUKvV\n4rIn1Z2QkIAxY8bA09MTx44da4mSm9Vcjp9++gne3t4YMGAAzp0712D91pijuQwxMTEYN24cxo8f\nj+DgYFRUVIivPXUGkgi9Xk9ERLW1tRQQEEAnT56kmJgYSkxMJCKihIQEio2NbckS/5Xt27dTaGgo\nzZ07l4hIkhnc3d2prKyswTIp5li2bBnt2bOHiOr+Xv3999+SzEFEJAgCubq6UnFxseQyFBUVkUql\nourqaiIiWrRoEaWmpkoqx4ULF8jHx4eqqqrIYDDQ9OnTqaCgQBIZTp48SefOnSMfHx9x2ePqvnTp\nEo0fP55qamqoqKiIPDw8SBCEFqm7seZyXL58ma5evUpBQUGUl5cnLm+tOZrLcOzYMbG22NjYZ9oX\nkjjDB4COHTsCAGprayEIArp06SK58fdLSkqQmZmJgIAAcZnUMtSjRl/ukFqOtjanQ3Z2Nvr06YNe\nvXpJLoO5uTkUCgUqKythMBhQVVWFnj17SirH1atXMXDgQHTo0AEmJiZ49913cfDgQUlkGDx4cJOv\nQD+u7kOHDsHb2xumpqbo3bs3+vTpg7Nnz770mpvTXA47Ozu88cYbTdZtrTmay+Dq6gq5vK5VOzk5\noaSkBMB/yyCZhm80GjFhwgQMGzYMQ4cORb9+/SQ3/v7atWuxbNkycecBkFwGoG7o4xkzZsDf3x+7\nd+8GIL0cj87p4Ofnh4iICOj1esnlqJeeng5vb28A0tsXlpaWmDlzJt577z2MGDECFhYWcHV1lVSO\nfv364dSpUygvL0dlZSWysrJw69YtSWV41OPqvn37NmxsbMT1bGxscOvWrRap8VlINUdKSgrc3NwA\n/LcMkmn4crkcaWlpyMrKwqlTp5CTk9Pg9dY+/v7hw4dhZWUFe3v7JmfH9Vp7hnrffPMNdDodtm7d\nil27duHUqVMNXpdCjrY0p0NNTQ0OHz6McePGNXlNChkKCwuxc+dOZGRk4OjRo9Dr9UhLS2uwTmvP\nYWdnh9mzZ2PmzJmYPXs23n777QYH9kDrz/A4/1S3FDM1p7Xn2Lx5M0xNTRvc32/snzJIpuHXs7Cw\ngJubG86dOwcrKyvcuXMHAP71+Pst5Y8//kBGRgZUKhWWLFmCnJwcLF26VFIZ6vXs2RMA0K1bN4we\nPRpnz56VXI7m5nQ4f/68OKcDII0cAJCVlQUHBwexVqnti7y8PLi4uKBr165QKBQYPXo0Tp8+Lbl9\nMWnSJKSmpiI5ORldunTB66+/Lrl9Ue9xdVtbW4uXlIG625TW1tYtUuOzkFqO1NRUZGZmYv369eKy\n/5JBEg2/tLRU/C3RqqoqZGdnw97eHiqVSjLj74eGhiIzMxMZGRnYsGEDlEolYmNjJZUBACorK/Hg\nwQMAgF6vx7Fjx/DWW29JLkdbmtMhPT0dPj4+4nOp7Yu+ffvizJkzqKqqAhFJdl/UX/YuLi7Gzz//\nDLVaLbl9Ue9xdatUKqSnp6OmpgZFRUUoKCgQD5pbu0evrEopR1ZWFrZt24ZNmzahQ4cO4vL/kkES\nQ+teuHABGo0GRqNRvJc/a9YslJeXS3L8/dzcXGzfvh3x8fGSy1BUVITg4GAAdV8xVKvVmDt3ruRy\nAG1jTge9Xg93d3ccOnQI5ubmACDJfbFlyxbodDrI5XLY29tDq9Xi4cOHksoxZcoUlJeXQ6FQICws\nDEqlUhL7IjQ0FLm5uSgvL4eVlRVCQkIwatSox9YdHx+PlJQUmJiYIDw8HCNGjGjhBHUa51i4cCEs\nLS2xevVqlJWVwcLCAgMGDMDWrVsBtM4czWVITExEbW0tunTpAgBwdnZGVFQUgKfPIImGzxhjjLFn\nI4lL+owxxhh7NtzwGWOMsXaAGz5jjDHWDnDDZ4wxxtoBbviMMcZYO8ANnzHGGGsHFC1dAGNtkcFg\nwJYtW7B//37IZDIIggA/Pz/MnTv3ie+Li4uDTCYTxzqod+DAAWRmZmLdunUvsuwGdXz33XfieOoA\nYG9vj7Vr1zZYLy8vD99++y20Wm2z2ykqKkJ8fDzWrFnzzDXl5+dj3bp1KC8vhyAIcHZ2Rnh4uDix\nFmPsybjhM/YCREdHo7S0FLt374a5uTkePHiA4OBgmJubY8qUKY99X2sZz1smk+H9999vcuDRmKOj\n42ObPVA36lxhYeFzqWnx4sX45JNP4OTkBCJCdHQ0vvjiC2g0mueyfcbaOm74jD1nJSUl2L9/P44e\nPSqOfmdubo7IyEhcuXIFAHD37l2Eh4fj5s2bUCgUWLx4cZNRsr7//nts3rwZnTp1wmuvvSYOq6lS\nqeDl5YUjR47AxMQEoaGh2LZtGwoLC7F8+XKMGzcOFy9ehFarhV6vR2lpKWbMmIGpU6ciLi4Ot27d\nQkFBAYqLixEQEIB58+Y9VT6lUglHR0fcvXsXy5Ytw+bNm/HVV19hx44d4mh577zzDj7++GNotVpc\nv34dq1evxsqVKxEfH4/9+/dDLpdj+PDhWLp0KeRyOXQ6HZKSkmA0GuHg4IBVq1bhlVdeafC59+7d\ng16vBwDxKkhxcTEA4MaNGwgLC0NZWRnMzMyg1WrRv39/pKSk4MsvvwRQd3CycuVKdOrUScxw7949\n7NmzB9u3b8eBAwcgCIJYF2NtDjHGnquDBw+Sv7//E9cJCQmhHTt2EBFRYWEhDR8+nO7evUtxcXEU\nFxdHJSUl5OrqSnfu3CFBEGj+/Pmk0WiIiMjd3Z2SkpKIiEij0dDkyZNJEATKzc0lX19fIiJas2YN\nHT9+XNy+i4sLERFt3LiRAgICqLa2lu7du0cuLi5UUVHRpL6NGzeSq6srTZgwQXxcu3aNiIj69+9P\nubm5RESUk5NDQUFBZDAYSKlUksFgIKPRSKtWraKSkhI6ceIEBQUFERHRkSNHKDAwkKqrq8lgMND8\n+fMpOTmZLl68SJMnT6bq6moiIlq/fj1t2rSpSU2pqank7OxMY8aMoZUrV9KRI0fE12bPnk27du0S\nP2fRokV04cIFGj16NJWXlxMRUXR0NMXExDTJkJmZSSEhISQIAgmCQKGhoZSWlvbE/ceYFPEZPmMv\nwKOX5g8cOID4+HgIgoAOHTpg7969OHHihHhf+9VXX4WTkxPOnDkjvuf06dNwcXER76H7+/vj0KFD\n4usjR44EANja2sLGxgZyuRy9evXC/fv3AQAajQZZWVlITExEfn4+KisrxfcqlUooFAp069YNlpaW\nqKioEK9EPFr/ky7pOzk5NXhuYmICFxcXTJw4EaNGjcKUKVNgbW2Nv/76S1wnJycHPj4+4pn7xIkT\nodPpAAAFBQUIDAwEANTW1sLBwaHJZ/r5+WHs2LHIzs5GdnY2NBoN1Go1VqxYgZMnT+Kzzz4DALi5\nucHNzQ3JyclQqVTiGOSBgYFYsWJFkwzHjx/H2bNn4e/vDwCorq5G7969m83NmJRxw2fsOXNwcMCV\nK1fw4MEDmJubw9PTE56enrhx4wamTp0KoOHMXfXPBUEQn8tksgbrNJ5b3dTUVPyziYlJkxoWLVoE\nS0tLuLu7w8vLCz/++KO43caXyo1GIyIiIpCXlwcAT7wnX6/xNgBg06ZNOHPmDDIzMzFr1qwGU3nW\nZ3w0ExHBYDBAEAR4enoiIiICAPDw4cMGPwug7oDghx9+wIIFC+Dh4QEPDw9MmzYNvr6+WLFiBUxN\nTRts+/Lly4/9vMYZjEYjpk2bhunTpwMA7t+/D4WC/2tkbQ9/LY+x58zW1hbjx4+HRqNBRUUFgLqZ\nBQ8fPiw2kqFDh2Lv3r0A6n6T/ffff4eLi4vYoAYNGoTTp0+jpKQERIT09PSnqiE7OxsLFy6ESqVC\nbm4ugLrG1vhAA6g7CNBqtdDpdNDpdHB0dHzqzGVlZfDy8kK/fv0QEhICV1dXXLx4EQqFQmzeSqUS\n6enpqK6uhsFgQEpKCpRKJYYMGYJff/0VpaWlICJERUUhKSmpwfa7du2K5ORk5OTkiMsuXboEe3t7\nAMDgwYPFg5rffvsNkZGRGDJkCDIyMsSrHrt374ZSqWxSu1KpRFpaGvR6PQwGA4KDg/HLL7889c+A\nsdaOD2MZewGioqKwY8cOfPjhhyAi1NTUwNnZGVu2bAEAREREIDIyEikpKZDJZFizZg26d+8u3gqw\nsrJCVFQUZs6cCTMzM/Tv3/9ffW79+xcuXIjJkyeje/fuGDx4MOzs7HD9+vXn8i2AR7chk8kgk8nQ\ntWtXBAYGYtKkSTAzM4OtrS38/PxQU1ODiooKLF++HDExMfjzzz8xceJEGAwGjBgxAlOnToVcLseC\nBQswbdo0GI1G2NvbY86cOQ0+s3PnzkhISEBsbCwiIiJgamqKvn37YsOGDQCAyMhIhIeH4+uvv0bH\njh2h1WphZ2eHOXPmICgoCAaDAY6OjoiOjm6Swd3dHfn5+QgMDIQgCBg5ciR8fX2f+efEWGvD0+My\nxhhj7QBf0meMMcbaAW74jDHGWDvADZ8xxhhrB7jhM8YYY+0AN3zGGGOsHeCGzxhjjLUD3PAZY4yx\ndoAbPmOMMdYO/A9kwSTAVsJcKQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10f1633d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8,6))\n",
    "ax.plot(gf2_m6, m6.fittedvalues, 'r.', label='Predicted')\n",
    "ax.plot(gf2_m6, aaps_m6, 'g.', label='Observed')\n",
    "plt.plot([35, 110], [35, 110], 'blue', label='Identity');\n",
    "legend = plt.legend(loc='upper left')\n",
    "plt.xlabel('Goldman-Fristoe Score')\n",
    "plt.ylabel('Arizona Score')\n",
    "plt.title('Model 6')\n",
    "ax.set_xlim([30,120])\n",
    "ax.set_ylim([60,110])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Future Considerations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In an effort to increase sample size, we could model all observations with both scores (73) and use clustered robust standard errors or build a repeated measures model.  \n",
    "\n",
    "For robust standard errors see http://statsmodels.sourceforge.net/0.6.0/stats.html   \n",
    "`sandwich_covariance.cov_cluster(results, group)`"
   ]
  }
 ],
 "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": 1
}