gistfile1.txt

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model = pd.read_excel('ea_test_model_analysis.xlsx')\n",
    "fips_cd = pd.read_excel('fips_codes_website.xls')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Get the unique FIPS value for each state\n",
    "fips_state_cd_all = pd.concat([pd.Series(fips_cd['State Abbreviation'].unique()), \n",
    "                           pd.Series(fips_cd['State FIPS Code'].unique())], axis=1)\n",
    "\n",
    "# Rename columns\n",
    "fips_state_cd_all.columns = ['state_cd', 'state_num']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>id</th>\n",
       "      <th>phone</th>\n",
       "      <th>gender</th>\n",
       "      <th>age_5way</th>\n",
       "      <th>ethnicity_4way</th>\n",
       "      <th>q1_healthcare</th>\n",
       "      <th>married</th>\n",
       "      <th>modeled_income_bucket</th>\n",
       "      <th>uninsured_score</th>\n",
       "      <th>state_cd</th>\n",
       "      <th>...</th>\n",
       "      <th>party_id</th>\n",
       "      <th>hh_size</th>\n",
       "      <th>gen_vote_2012</th>\n",
       "      <th>gen_vote_2010</th>\n",
       "      <th>support_score</th>\n",
       "      <th>turnout_score</th>\n",
       "      <th>county_fips</th>\n",
       "      <th>zip</th>\n",
       "      <th>census_id</th>\n",
       "      <th>census_neighborhood_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MI-9738252</td>\n",
       "      <td>2316756588</td>\n",
       "      <td>M</td>\n",
       "      <td>a_18to26</td>\n",
       "      <td>W</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1-under 40k</td>\n",
       "      <td>48.88</td>\n",
       "      <td>MI</td>\n",
       "      <td>...</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>2</td>\n",
       "      <td>(null)</td>\n",
       "      <td>(null)</td>\n",
       "      <td>29.43</td>\n",
       "      <td>28.874</td>\n",
       "      <td>29</td>\n",
       "      <td>49712</td>\n",
       "      <td>260290013001040</td>\n",
       "      <td>(null)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>TX-000002087318</td>\n",
       "      <td>8323167566</td>\n",
       "      <td>F</td>\n",
       "      <td>b_27to34</td>\n",
       "      <td>W</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1-under 40k</td>\n",
       "      <td>25.06</td>\n",
       "      <td>TX</td>\n",
       "      <td>...</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>1</td>\n",
       "      <td>(null)</td>\n",
       "      <td>(null)</td>\n",
       "      <td>20.29</td>\n",
       "      <td>59.5919</td>\n",
       "      <td>201</td>\n",
       "      <td>77055</td>\n",
       "      <td>482015223022025</td>\n",
       "      <td>(null)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>IL-000005343750</td>\n",
       "      <td>7087473204</td>\n",
       "      <td>M</td>\n",
       "      <td>C_35TO49</td>\n",
       "      <td>B</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1-under 40k</td>\n",
       "      <td>40.19</td>\n",
       "      <td>IL</td>\n",
       "      <td>...</td>\n",
       "      <td>(null)</td>\n",
       "      <td>3</td>\n",
       "      <td>(null)</td>\n",
       "      <td>(null)</td>\n",
       "      <td>86.91</td>\n",
       "      <td>27.6557</td>\n",
       "      <td>197</td>\n",
       "      <td>60475</td>\n",
       "      <td>170318294021023</td>\n",
       "      <td>Urban</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>MO-000000184817</td>\n",
       "      <td>6365610888</td>\n",
       "      <td>F</td>\n",
       "      <td>C_35TO49</td>\n",
       "      <td>B</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3-over 80k</td>\n",
       "      <td>4.73</td>\n",
       "      <td>MO</td>\n",
       "      <td>...</td>\n",
       "      <td>(null)</td>\n",
       "      <td>1</td>\n",
       "      <td>(null)</td>\n",
       "      <td>(null)</td>\n",
       "      <td>81.41</td>\n",
       "      <td>83.4038</td>\n",
       "      <td>183</td>\n",
       "      <td>63367</td>\n",
       "      <td>291833119092020</td>\n",
       "      <td>(null)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>PA-9509497</td>\n",
       "      <td>4128269507</td>\n",
       "      <td>M</td>\n",
       "      <td>A_18TO26</td>\n",
       "      <td>W</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2-40 - 80k</td>\n",
       "      <td>17.93</td>\n",
       "      <td>PA</td>\n",
       "      <td>...</td>\n",
       "      <td>Democrat</td>\n",
       "      <td>8</td>\n",
       "      <td>(null)</td>\n",
       "      <td>(null)</td>\n",
       "      <td>36.39</td>\n",
       "      <td>42.0622</td>\n",
       "      <td>3</td>\n",
       "      <td>15024</td>\n",
       "      <td>420034150013034</td>\n",
       "      <td>Rural</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                id       phone gender  age_5way ethnicity_4way  q1_healthcare  \\\n",
       "0       MI-9738252  2316756588      M  a_18to26              W              1   \n",
       "1  TX-000002087318  8323167566      F  b_27to34              W              1   \n",
       "2  IL-000005343750  7087473204      M  C_35TO49              B              2   \n",
       "3  MO-000000184817  6365610888      F  C_35TO49              B              1   \n",
       "4       PA-9509497  4128269507      M  A_18TO26              W              1   \n",
       "\n",
       "   married modeled_income_bucket  uninsured_score state_cd  \\\n",
       "0        0           1-under 40k            48.88       MI   \n",
       "1        0           1-under 40k            25.06       TX   \n",
       "2        0           1-under 40k            40.19       IL   \n",
       "3        0            3-over 80k             4.73       MO   \n",
       "4        1            2-40 - 80k            17.93       PA   \n",
       "\n",
       "            ...             party_id hh_size gen_vote_2012 gen_vote_2010  \\\n",
       "0           ...              Unknown       2        (null)        (null)   \n",
       "1           ...              Unknown       1        (null)        (null)   \n",
       "2           ...               (null)       3        (null)        (null)   \n",
       "3           ...               (null)       1        (null)        (null)   \n",
       "4           ...             Democrat       8        (null)        (null)   \n",
       "\n",
       "  support_score turnout_score county_fips    zip        census_id  \\\n",
       "0         29.43        28.874          29  49712  260290013001040   \n",
       "1         20.29       59.5919         201  77055  482015223022025   \n",
       "2         86.91       27.6557         197  60475  170318294021023   \n",
       "3         81.41       83.4038         183  63367  291833119092020   \n",
       "4         36.39       42.0622           3  15024  420034150013034   \n",
       "\n",
       "  census_neighborhood_type  \n",
       "0                   (null)  \n",
       "1                   (null)  \n",
       "2                    Urban  \n",
       "3                   (null)  \n",
       "4                    Rural  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Get area code for each phone number\n",
    "model['area_code'] = model['phone'].astype(str).str.extract('^([0-9][0-9][0-9])', expand=False)\n",
    "\n",
    "# Remove irrelevant parts of phone number\n",
    "model.drop('phone', axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Make categories uniform. a: 18-26, b: 27-34, c: 35-49, d: 50-64\n",
    "model['age_5way'] = model['age_5way'].astype(str).str.lower().str.extract('^([a-z])', expand=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "model.modeled_income_bucket.unique()\n",
    "\n",
    "# Change the groupings to 1: Under 40k, 2: 40-80k, and 3: Over 80k\n",
    "model['modeled_income_bucket'] = model['modeled_income_bucket'].astype(str).str.extract('^([0-9])', expand=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model.census_id = model.census_id.astype(str)\n",
    "\n",
    "# Join state codes to remove any NaN values\n",
    "model = model.merge(fips_state_cd_all, how='left', left_on='state_cd', right_on='state_cd')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Regular expression that breaks the census tract id number into state fips, county fips, \n",
    "# census tract code, and tabular block code\n",
    "census_id_regex = '^([0-9]{2})([0-9]{3})([0-9]{6})([0-9]{4})'\n",
    "new_column_names = ['state_fips', 'county', 'census_tract', 'tabulation_block']\n",
    "model[new_column_names] = model.census_id.astype(str).str.extract(census_id_regex, expand=False)\n",
    "\n",
    "# Removes redundant columns\n",
    "model.drop(['county'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x9c6a6d8>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAGpCAYAAADY7qJlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noAQCgUCwKPn/AVle0fz8YWKPAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x9c6a160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.FacetGrid(model, hue='q1_healthcare', legend_out=False, \n",
    "              size=(6), palette= 'muted').map(sns.kdeplot, 'uninsured_score').add_legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To better compare how each prediction did, I used a KDE plot to make a more comparable looking plot. The values of 'q1_healthcare' mean that the customer: 1(Is insured), 2(Is not insured), 3(Unknown), and 4(Refuse to answer).\n",
    "\n",
    "It was expected to see that those with health insurance have a lower 'uninsured_score', while those who answered that they do not have health insurance have a higher 'uninsured_score.' It is also interesting to see that those who do not know their health insurance have the highest 'uninsured_score.' Keep in mind that because these are relative plots, it may not be indicative of the actual outcomes. The amount of people that answered 3/4 were significantly smaller than 1/2, as seen in the below table."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "background = ['ethnicity_4way', 'modeled_income_bucket']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>q1_healthcare</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ethnicity_4way</th>\n",
       "      <th>modeled_income_bucket</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">A</th>\n",
       "      <th>1</th>\n",
       "      <td>31</td>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>71</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">B</th>\n",
       "      <th>1</th>\n",
       "      <td>287</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>77</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>114</td>\n",
       "      <td>12</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">H</th>\n",
       "      <th>1</th>\n",
       "      <td>261</td>\n",
       "      <td>128</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>92</td>\n",
       "      <td>36</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>120</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">W</th>\n",
       "      <th>1</th>\n",
       "      <td>463</td>\n",
       "      <td>94</td>\n",
       "      <td>3</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>275</td>\n",
       "      <td>39</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>643</td>\n",
       "      <td>46</td>\n",
       "      <td>2</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "q1_healthcare                           1    2  3   4\n",
       "ethnicity_4way modeled_income_bucket                 \n",
       "A              1                       31   14  0   1\n",
       "               2                       18    1  1   1\n",
       "               3                       71    6  1   0\n",
       "B              1                      287   77  1   9\n",
       "               2                       77   11  0   2\n",
       "               3                      114   12  2   1\n",
       "H              1                      261  128  1   3\n",
       "               2                       92   36  1   0\n",
       "               3                      120   21  0   0\n",
       "W              1                      463   94  3   8\n",
       "               2                      275   39  2  10\n",
       "               3                      643   46  2  14"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab([model.ethnicity_4way, model.modeled_income_bucket], model.q1_healthcare)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Going forward we will compare between these 4 groups. The below plots are examples of the kind of demographic data the model is based on.\n",
    "\n",
    "The first plot shows the ethnicity of W(White), B(Black), H(Hispanic), and A(Asian). For the second plot, the three recorded group ranges for income include 1(Under 40k), 2(40-80k), and 3(Over 80k).\n",
    "\n",
    "I have split the below plots to demonstrate how the outcome may change due to these features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIcAAAEaCAYAAACLsTehAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDkmSJEmSJNXY/wJPL9FIiEjU2AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x9b780f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIcAAAEaCAYAAACLsTehAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAMCA/f82Daz7dFiGmwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xc4b8588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for x in background:\n",
    "    sns.factorplot(x, col='q1_healthcare', data=model, kind='count', sharey=True, palette='RdBu')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the 'ethnicity_4way' plot, we see that H make the majority of those that answered 'No' for their healthcare, despite having a population much less than W.\n",
    "\n",
    "We see in the second plot that those in the 'Under 40k' bracket make up a very large portion of the uninsured group. \n",
    "\n",
    "Just based on these plots, further investigation into the 'H/Under 40k' bracket may yield a higher 'uninsured_score.'"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}