{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Chapter 12: Insurance redlining example\n",
    "\n",
    "Load in the Python packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import statsmodels.api as sm\n",
    "import statsmodels.formula.api as smf\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ecological Correlation\n",
    "\n",
    "Read in the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>usborn</th>\n",
       "      <th>income</th>\n",
       "      <th>home</th>\n",
       "      <th>pop</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Alabama</th>\n",
       "      <td>0.98656</td>\n",
       "      <td>21442</td>\n",
       "      <td>75.9</td>\n",
       "      <td>4040587</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alaska</th>\n",
       "      <td>0.93914</td>\n",
       "      <td>25675</td>\n",
       "      <td>34.0</td>\n",
       "      <td>550043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Arizona</th>\n",
       "      <td>0.90918</td>\n",
       "      <td>23060</td>\n",
       "      <td>34.2</td>\n",
       "      <td>3665228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Arkansas</th>\n",
       "      <td>0.98688</td>\n",
       "      <td>20346</td>\n",
       "      <td>67.1</td>\n",
       "      <td>2350725</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>California</th>\n",
       "      <td>0.74541</td>\n",
       "      <td>27503</td>\n",
       "      <td>46.4</td>\n",
       "      <td>29760021</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             usborn  income  home       pop\n",
       "Alabama     0.98656   21442  75.9   4040587\n",
       "Alaska      0.93914   25675  34.0    550043\n",
       "Arizona     0.90918   23060  34.2   3665228\n",
       "Arkansas    0.98688   20346  67.1   2350725\n",
       "California  0.74541   27503  46.4  29760021"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eco = pd.read_csv(\"data/eco.csv\", index_col=0)\n",
    "eco.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(eco.usborn, eco.income)\n",
    "plt.ylabel(\"Income\")\n",
    "plt.xlabel(\"Fraction US born\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>income</td>      <th>  R-squared:         </th> <td>   0.334</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.321</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   24.60</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 25 Sep 2018</td> <th>  Prob (F-statistic):</th> <td>8.89e-06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>15:47:22</td>     <th>  Log-Likelihood:    </th> <td> -487.38</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    51</td>      <th>  AIC:               </th> <td>   978.8</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    49</td>      <th>  BIC:               </th> <td>   982.6</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     1</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>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th>  <td> 6.864e+04</td> <td> 8739.004</td> <td>    7.855</td> <td> 0.000</td> <td> 5.11e+04</td> <td> 8.62e+04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>usborn</th> <td>-4.602e+04</td> <td> 9279.116</td> <td>   -4.959</td> <td> 0.000</td> <td>-6.47e+04</td> <td>-2.74e+04</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 7.381</td> <th>  Durbin-Watson:     </th> <td>   1.635</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.025</td> <th>  Jarque-Bera (JB):  </th> <td>   6.433</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.727</td> <th>  Prob(JB):          </th> <td>  0.0401</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.956</td> <th>  Cond. No.          </th> <td>    35.8</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                 income   R-squared:                       0.334\n",
       "Model:                            OLS   Adj. R-squared:                  0.321\n",
       "Method:                 Least Squares   F-statistic:                     24.60\n",
       "Date:                Tue, 25 Sep 2018   Prob (F-statistic):           8.89e-06\n",
       "Time:                        15:47:22   Log-Likelihood:                -487.38\n",
       "No. Observations:                  51   AIC:                             978.8\n",
       "Df Residuals:                      49   BIC:                             982.6\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const       6.864e+04   8739.004      7.855      0.000    5.11e+04    8.62e+04\n",
       "usborn     -4.602e+04   9279.116     -4.959      0.000   -6.47e+04   -2.74e+04\n",
       "==============================================================================\n",
       "Omnibus:                        7.381   Durbin-Watson:                   1.635\n",
       "Prob(Omnibus):                  0.025   Jarque-Bera (JB):                6.433\n",
       "Skew:                           0.727   Prob(JB):                       0.0401\n",
       "Kurtosis:                       3.956   Cond. No.                         35.8\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lmod = sm.OLS(eco.income, sm.add_constant(eco.usborn)).fit()\n",
    "lmod.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(eco.usborn, eco.income)\n",
    "plt.ylabel(\"Income\")\n",
    "plt.xlabel(\"Fraction US born\")\n",
    "plt.plot([0, 1.0], [lmod.params[0] + lmod.params[1]*0.0, lmod.params[0] + lmod.params[1]*1])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Initial Data Analysis\n",
    "\n",
    "Read in and plot the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>race</th>\n",
       "      <th>fire</th>\n",
       "      <th>theft</th>\n",
       "      <th>age</th>\n",
       "      <th>involact</th>\n",
       "      <th>income</th>\n",
       "      <th>side</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>60626</th>\n",
       "      <td>10.0</td>\n",
       "      <td>6.2</td>\n",
       "      <td>29</td>\n",
       "      <td>60.4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.744</td>\n",
       "      <td>n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60640</th>\n",
       "      <td>22.2</td>\n",
       "      <td>9.5</td>\n",
       "      <td>44</td>\n",
       "      <td>76.5</td>\n",
       "      <td>0.1</td>\n",
       "      <td>9.323</td>\n",
       "      <td>n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60613</th>\n",
       "      <td>19.6</td>\n",
       "      <td>10.5</td>\n",
       "      <td>36</td>\n",
       "      <td>73.5</td>\n",
       "      <td>1.2</td>\n",
       "      <td>9.948</td>\n",
       "      <td>n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60657</th>\n",
       "      <td>17.3</td>\n",
       "      <td>7.7</td>\n",
       "      <td>37</td>\n",
       "      <td>66.9</td>\n",
       "      <td>0.5</td>\n",
       "      <td>10.656</td>\n",
       "      <td>n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60614</th>\n",
       "      <td>24.5</td>\n",
       "      <td>8.6</td>\n",
       "      <td>53</td>\n",
       "      <td>81.4</td>\n",
       "      <td>0.7</td>\n",
       "      <td>9.730</td>\n",
       "      <td>n</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       race  fire  theft   age  involact  income side\n",
       "60626  10.0   6.2     29  60.4       0.0  11.744    n\n",
       "60640  22.2   9.5     44  76.5       0.1   9.323    n\n",
       "60613  19.6  10.5     36  73.5       1.2   9.948    n\n",
       "60657  17.3   7.7     37  66.9       0.5  10.656    n\n",
       "60614  24.5   8.6     53  81.4       0.7   9.730    n"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "chredlin = pd.read_csv(\"data/chredlin.csv\", index_col=0)\n",
    "chredlin.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ((ax1, ax2, ax3), (ax4, ax5, ax6)) = plt.subplots(nrows=2, ncols=3)\n",
    "ax1.scatter(chredlin.race, chredlin.involact)\n",
    "ax1.set(title=\"Race\")\n",
    "ax1.set_ylabel(\"Involact\")\n",
    "\n",
    "ax2.scatter(chredlin.fire, chredlin.involact)\n",
    "ax2.set(title=\"Fire\")\n",
    "\n",
    "ax3.scatter(chredlin.theft, chredlin.involact)\n",
    "ax3.set(title=\"Theft\")\n",
    "\n",
    "ax4.scatter(chredlin.age, chredlin.involact)\n",
    "ax4.set(title=\"Age\")\n",
    "ax4.set_ylabel(\"Involact\")\n",
    "\n",
    "ax5.scatter(chredlin.income, chredlin.involact)\n",
    "ax5.set(title=\"Income\")\n",
    "\n",
    "ax6.scatter(chredlin.side, chredlin.involact)\n",
    "ax6.set(title=\"Side\")\n",
    "\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A more compact way of achieving a similar result:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3UAAACsCAYAAADc14VaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzt3X+YZFV95/H3t/rX9PS0TDP0sIQZgbA4ZjQj0A2Ik8dFTVxgWQkPEyVmQAT5ITGabJZoNmvisz7ZxCUJUSPMiAIOxJ8gSlw3ShB1M1G0B3FEcBR/IAMs0zQz0NPT093V9d0/qqqp7q6qvlV9b917635ez1PPdP269b3nfM+pe6buOdfcHREREREREUmnXNwBiIiIiIiISPM0qBMREREREUkxDepERERERERSTIM6ERERERGRFNOgTkREREREJMU0qBMREREREUkxDepERERERERSTIM6ERERERGRFNOgTkREREREJMVSN6g7++yzHdBNt6C32ClndWvwFjvlrG4N3GKnfNWtwVvslLO6NXgLJHWDumeeeSbuEEQaopyVtFHOSpooXyVtlLMShdQN6kREREREROQFGtSJiIiIiIikmAZ1IiIiIiIiKaZBnYiIiIiISIp1xh2ASLMKBWdsYprp/CzdnR2s6esml7O4w8oc1YOEQXkk0j7UnltL5S2gQZ2kVKHg7Hl6nCt2jLB3/yTrBnq56ZJhNhzdr46shVQPEgblkUj7UHtuLZW3lOn0S0mlsYnpuQ4MYO/+Sa7YMcLYxHTMkWWL6kHCoDwSaR9qz62l8pYyDeoklabzs3MdWNne/ZNM52djiiibVA8SBuWRSPtQe24tlbeUaVAnqdTd2cG6gd55j60b6KW7syOmiLJJ9SBhUB6JtA+159ZSeUuZBnWSSmv6urnpkuG5jqx8Dvmavu6YI8sW1YOEQXkk0j7UnltL5S1lWihFUimXMzYc3c9d12zWak8xUj1IGJRHIu1D7bm1VN5SpkGdpFYuZwz298QdRuapHiQMyiOR9qH23FoqbwGdfikiIiIiIpJqkQ3qzGy9md1nZo+Y2Q/N7F1VXmNm9iEze9TMdpvZqVHFIyIiIiIi0o6iPP0yD/yxuz9gZv3ALjO7x90frnjNOcBJpdsZwI2lf0VERERERCSAyH6pc/en3P2B0t/jwCPAsQtedj6ww4u+Daw2s2OiiklERERERKTdtGROnZkdD5wC3L/gqWOBxyvu72XxwE9ERERERERqiHxQZ2argDuBP3T35xc+XeUtXmUbV5rZiJmNjI6ORhGmSKiUs5I2yllJE+WrpI1yVqIW6aDOzLooDuj+0d0/X+Ule4H1FffXAU8ufJG7f9Tdh919eHBwMJpgE6hQcEbHp3hi/yFGx6coFBaNd2OV9PjilNWcheTlRdLiSaqoczZoPai+JIgs97HtrFb7b4d+oZU5m88XePLAJI+NTfDkgUny+UKknyfJENlCKWZmwMeBR9z972q87G7gHWb2aYoLpDzn7k9FFVOaFArOnqfHuWLHCHv3T7JuoJebLhlmw9H9ibigZNLjk3gkLS+SFk9WBa0H1ZdIdtVq/ycNruInowfVLwSUzxf40dPjXH37rrny2rZ1iJce3U9np65k1s6irN3NwMXAa83swdLtXDO72syuLr3my8DPgEeBm4BrIownVcYmpuc6MIC9+ye5YscIYxPTMUdWlPT4JB5Jy4ukxZNVQetB9SWSXbXa/76DU+oXGrDv4NTcgA6K5XX17bvYd3Aq5sgkapH9Uufu/0r1OXOVr3Hg96OKIc2m87NzDbJs7/5JpvOzMUU0X9Ljk3gkLS+SFk9WBa0H1ZdIdtVq//nZgvqFBszUKK/8rE7BbHf6HTahujs7WDfQO++xdQO9dHd2xBTRfEmPT+KRtLxIWjxZFbQeVF8i2VWr/Xd25NQvNKCrRnl1duiQv92phhNqTV83N10yPNcwy+eQr+nrjjmyouXG18ik53aYIJ0VScvbVsSj/FxakHooFBzHuf3yM7jl0tM4Zf3qea9TOYskR7Ptsd77avUTa1f1JOp7JQ6NLCCzdlUP27YOzSuvbVuHWLuqJ85dkBaw4hmQ6TE8POwjIyNxh9EShYIzNjHNdH6W7s4O1vR1J2pScLPxNbIYQggLJ8ReYFnL2V+MTfDY2CFWdndwaHqW49as5Pg1fbHlbpTtKKKFPdoyZ+vVQ7Vy3L51iGNWr2B1b/HATQuoJFbsFZClPjYJmu33gryvVj8Rcj+eqpxtZgGZQsHZd3CK/GyBzo4ca1f1aJGUdAt2wKtBnbTa6PgUF9ywc9453+sGernrms0M9vc0/doaUtV5p10I9ZUqEe1v5nJ2qXLMWl6lTObyNeuabY8Jasepytla5fbZq87kjdu/lYTylOgFylkN26XlGlkMQQsnpEvW6itr+xuVpcpR5SySHM22R7Xj5mgBGQlKgzppuUYWQ9DCCemStfrK2v5GZalyVDmLJEez7VHtuDlaQEaC0qBOWq6RxSuStvCG1Je1+sra/kZlqXJUOYskR7PtUe24OVpARoLSnDqJRSOTnpc5QTpV5863g6Qv8BO2CPY39sKKI2eXKses5VWKxF4JWetjk2A5C6UloB2nLmdbtICMJFegSo3s4uMi9eRyFngibyOvlfhlrb6ytr9RWaocVc4iydFse1Q7bk6tclN5SiWdfikiIiIiIpJiGtSJiIiIiIikmAZ1IiIiIiIiKaZBnYiIiIiISIppUCciIiIiIpJiGtSJiIiIiIikWGSDOjO72cz2mdlDNZ4/y8yeM7MHS7c/jyoWERERERGRdhXldepuBf4B2FHnNf/X3c+LMIZM00UpJY2Ut9mi+haRSuoTwqFyzJ7IBnXu/k0zOz6q7Ut9hYKz5+lxrtgxwt79k6wb6OWmS4bZcHS/GrUklvI2W1TfIlJJfUI4VI7ZFPecujPN7Ptm9n/M7GUxx9JWxiam5xozwN79k1yxY4SxiemYIxOpTXmbLapvEamkPiEcKsdsinNQ9wBwnLu/Avgw8IVaLzSzK81sxMxGRkdHWxZgmk3nZ+cac9ne/ZNM52djiihblLPNUd7GJ46cVX1Ls9THtqd27hNambPtXI5SW6BBnZltDvJYI9z9eXc/WPr7y0CXmR1V47Ufdfdhdx8eHBxczsdmRndnB+sGeuc9tm6gl+7OjpgiyhblbHOUt/GJI2dV39Is9bHtqZ37hFbmbDuXo9QW9Je6Dwd8LDAz+3dmZqW/Ty/FMracbWZJoeCMjk/xxP5DjI5PUSj4vOfX9HVz0yXDc426fD71mr7uOMIVCSTMvF2qjUi8CgXHcW6//AxuufQ0Tlm/Wv2USIapTwhPM9+l+s5Mv7oLpZjZmcCrgEEz+y8VT70IqDvcN7NPAWcBR5nZXuAvgC4Ad98GbAHebmZ5YBK4yN2VQQEEmQCbyxkbju7nrms2a+UjSY2w8laTxJOtWv1s3zrEMatXsLpX/ZRI1qhPCFej36X6zmwPS/1S1w2sojj466+4PU9xUFaTu/+uux/j7l3uvs7dP+7u20oDOtz9H9z9Ze7+Cnd/pbv/2/J3JxuCToDN5YzB/h6OHVjJYH+PGqakQhh5q0niyVatfq66fRezBdRPiWSQ+oTwNfJdqu/M9lD3lzp3/wbwDTO71d0fa1FMsgRNgBWpT20k2VQ/IlJJfUK8VP7tIeicuo+Z2eryHTMbMLOvRBST1FA+3xnQBFhpC1Gdw69J4uELs65UPyLZVKsfUZ8QjaD9tsq/PQQd1B3l7gfKd9x9P7A2mpCkmvL5zhfcsJN3fPJ7XLdlkxZBkVSrzOnNH7iPC27YyZ6nx0MZ2GmhoHCFXVeqH5HsqdePqE8IXyP9tsq/PViQtUnMbBdwgbv/snT/OOAudz814vgWGR4e9pGRkVZ/bOxGx6e44Iadcz+Pn7J+Ne983UmcuHYVvV1aBKWO2Aslqzm7lIU5DcUvkruu2cxgf8+yt18oOGMT02lcKCj2IBfmbBR1leL6kflirzT1semwVD/Swj4hEznbaL+tPjnRAlVE3Tl1Ff4M+Fcz+0bp/quBK5uJSpqz8Hzn7z1+gLfe+l12vvs1oRwAi7Ra1OfwlyeJy/JFUVeqH5FsWaofUZ8Qrkb7bZV/+gU6/dLd/xk4FfgM8FlgyN01p66FdL6ztBvldHqorkRkudSPtJbKO3uCzqkDmAX2Ac8BG83s1dGElDxJuCCjzneW5UhCDi+knE6POOoqiTkrIs2r1o9sv3iIgd6umCNrT8322+p70yvQ6Zdm9jbgXcA64EHglcC3gNdGF1oyJOWCjLqYuDQrKTm8kHI6PVpdV0nNWRFpXi5nnDS4ik++7Qz2jU8xNjHNB//lx/zRb21Q245AM/22+t50C/pL3buA04DH3P01wCnAaGRRJUiSLsioi4lLM5KUwwspp9OjlXWV5JwVkebtn5zhzR+7ny3bvsVVt+3iqw/vU9uOUKP9tvredAs6qDvs7ocBzKzH3X8EbIgurOTQBRkl7ZTDkjbKWZH2pLadbKqfdAs6qNtbuvj4F4B7zOyLwJPRhZUcmmgqaacclrRRzoq0J7XtZFP9pFvQ1S8vcPcD7v4+4L3Ax4HfjjKwpKicaHrK+tXcculp3H75GTiuyaOSCstd5EKTpiVMQfJJi+iItKeo27a+r5pXKDiOc/vlZ3DLpadxyvrV6ntTpu5CKWZ2ZJWHf1D6dxXwbOgRJUx5ound79jMUwcOc9XtuzR5VFJlOYtcaNK0hCloPmkRHZH2FGXb1vdV86qV3fatQxyzegWre9X3psVSv9TtAkZK/y68jUQbWnLkcsZsgbkBHWjyqKRLs4tcaNK0hKmRfNIiOiLtKaq2re+r5lUru6tu38VsAfW9KVL3lzp3P6FVgSSdJo9KFinvJUzKJxGJivqX5qns2kPgi4+b2RvM7G9Kt/MCvP5mM9tnZg/VeN7M7ENm9qiZ7TazUxsJvNU0eVSySHkvYVI+iUhU1L80T2XXHgIN6szsryleq+7h0u1dZvZXS7ztVuDsOs+fA5xUul0J3BgklriENblXk3glTbRQkIRpYT/6+o1r+eTbzmA6P6v+UCRB0nisogWWmtds2aUxT9pZ3dMvK5wLnOzuBQAz+wTwPeBPa73B3b9pZsfX2eb5wA53d+DbZrbazI5x96cCxtRSYUzu1SReSRstFCRhquxHC4UCz0xM8+aP3a+cEkmQtB6raIGl5jVTdmnNk3YW+PRLYHXF30eE8NnHAo9X3N9beiyxlju5V5N4JY20UJCEqdyP5nI5rrpNOSWSNGk+VtECS81rtOzSnCftKugvdX8FfM/M7gMMeDV1fqULqFq2VP3d1syupHiKJi9+8YuX+bHx0UTU7GiXnC1T7ra/VuesckqWo9362CRR24xGu+Ws8iR5gl58/FPAK4HPl25nuvunl/nZe4H1FffXAU/W+PyPuvuwuw8PDg4u82Pjo4mo2dEuOVum3G1/rc5Z5ZQsR7v1sUmithmNdstZ5UnyBF0o5W7gLOBed/+iu/+/ED77buCS0iqYrwSeS+p8urBUm4i6fesQHTmanlzaykmqcUyIXeozNUk3mELBeXaiWE6/fHaCfeOHyecLgcsuignoWay7LO5zLVX7w4uHMJwnD0xWLaOF5ddIDktzlLPZUa7rQqHA9ouHAvX3rcqPds/D5exflGVTa9uFgtPVAdu3BssTaY2gp1/+LfAm4K/N7DvAZ4AvufvhWm8ws09RHAgeZWZ7gb8AugDcfRvwZYoLsDwKHALe2uQ+pEZ5Iurnr3kVh6Zm+fkzE/z3LzzE6MGppiaXtnKSahwTYpf6TE3SDaZQcH4xNsHTzx/m2jt2z5XVtq1DfOjeH/PVh/ctWXZhT0DPYt1lcZ/rWZhTswXnH7/9C1694WjefefuRWUELCq/RnJYGqeczY7Kuh5c1cP73rCR95//clZ2d3BoepaezsW/AbQqP9o9D5ezf1GWTa1tnzS4il/uP8TTzx/mlp0/573nbWRNXzdr+3v4lSN626JO0sqKi08GfLFZB/Ba4ArgbHd/UVSB1TI8POwjIyOt/thQjY5PccENO+edi7xuoJe7rtnMYH9Py7eTtM8K+pkBY4q9d4k7Z0fHp3joied47xcfWlRW7z1vI1fdtmvufpT1uTCmVudT3BrY58zlbLls3nveRt7/pYerlhFQtfziyuEsUB+bHZV1vf3ioZrtsLJttaofj+BzEpWzy9m/KOug1rY/e9WZ7Pl/41WPKdT/RiZQzgb9pQ4z6wX+M8Vf7E4FPtFcXBLW5NJWTlKNY0LsUp+pSbrBTOdnWdndUbWsVvd2zbvfqrLLYt1lcZ+DKpfN6t6uumWUpBzOAuVsdlTW9VLtsNp76r0uzNii/Jy4LGf/oiybWtvOzxZqHlO0S52kVdA5dZ8BHqH4K91HgBPd/Q+iDKydhTW5tJWTVOOYELvUZ2qSbjDdncXTZ6qV1YHJmXn3W1V2Way7LO5zUOWyOTA5U7OMapVfXDmcBcrZ7Kis63rtsNZ76r0uzNii/Jy4LGf/oiybWtvu7MjVPKZolzpJq6DXqbuF4kDuanf/Wvki5O2i1RNww1p0opntNLuvUSyUsVR8HTnqfmarYkq7NX3dHLdmJddt2TSvrLZtHeLOXY/P3W9kIvxy20wW6y6L+xzUQG8X2y8u5uPf/s4rqpZR+TXN5LA0RzmbHZV1ve3rP130fVGt3ps9Blm4aNdS3x+NfE4aF1RZTjtb09fNjstO55ZLT+MzV76SWy49jR2Xnd50G13qGGz71iFWdueqHlOob4hf4Dl1ZvYq4HgqTtl09x3RhFVb2OfOxzUBt1Bwxiaml73oRCPbWe6+hhVzve0vjG/HZaezakUnM/lC1c8MEFOizp2PS6HgHJicZnJ6lnzBeeq5w3x+117O+fVjOOGoPlb2dHBU3+KLjdaqk6l8YdltJup8SqKA+xx7IbQyZ8s5dv09e3jr5hM4alU3e/cfnlug4bg1K3nxwEp+MnqQ6+/Zw4VD6+cm5R/zohUcOJzPVA61mvrY7Kis666OHPnZAnmHFV25qt8PC98T5Bik2qJdQb4/gnxOA8c4icvZZr8PwzyGrXkM1tPJoeni4n4fuvcnjB6cYsdlp3PEyk4OTxeYXSJHJBSBCjbQoM7MbgNOBB4EyifMuru/s+nwmhR2552lBRuSvq8RxRd7D5OkA45Gy7ja62+59DRNkI5WpnI2yAINn73qTN64/VvKuWTKVL5K8+ot2hXlwh7tvBhVmMdN9RZGUf8bu1AXShkGNnojS2WmRLtPwK2U9H1NenztoNEyrvZ6TZCWMAVZoCE/W1DOiaRcvUW7olzYo537iTD3ud7CKFkr17QKOqfuIeDfRRlIXJYzyTRt524nfbJx0uNLuiD52GgZV3u9JkhLmIIs0NDZkWso59LWN4tkQb1Fu8Joy1k8hghzn7s6q/ezjfa/Ep+gg7qjgIfN7Ctmdnf5FmVgrdLsBNXyuccX3LCTzR+4jwtu2Mmep8cTffCQ9EnvSY8vyYLmY6NlXO31x61ZqXqS0ARZoGHtqp6GFkpIW98skgW1Fu0Kqy1n8RgirH0uFJyDh/PL7n8lXkHn1P2Hao+7+zdCj2gJUZw738wE1aTPT6sl6YtTRBBf7DvXivkejeRjo2Vc7fVAovMo5WIvyFbPUarMsd7uDvIFX7Q4UtC8TWvfnGKZy1dpXuWiXUstsNFMW87iYlRhHDeVy3pwVQ9Xn3Uiq3u7ODQ9yyvWH8GRfT2JP3bMgPDm1MUxeGulXM4a/rJP67nbzexrKyU9vqRqJB8bLeNar1c9SViC5GTQvE1r3yySBbmccWRfD/Qt/dpm2nIWjyHC2OdyWe/dP8lVt+2ae3znu18Dfdks1zSqe/qlmf1r6d9xM3u+4jZuZs+3JsRkyuK525JcykeRIrUFkfagttw6Kuv2UHdQ5+6/Ufq3391fVHHrd/cXtSbEZGqHc7e1mED7WM7cUOWApFW1/G2HvlkkLGnu49WWl6eRuldZt4eglzSQBXI5Y8PR/dx1zeZUnmMc10XXJRrN5KNyQNKsXv6muW8WCUva+/i0H2fFqdG6V1m3h6CrX0oV5XOMjx1YyWB/9Ym+STU2MT3X2KF4nvoVO0YYm5iOOTJpVqP5qByQNKuXv2num0XC0g59vNpyc5qpe5V1+mlQl1FaTECUA5Jmyl+R+tRGskt1n02RDurM7Gwz22Nmj5rZe6o8f6mZjZrZg6Xb26KMR16gSbGiHJA0U/6K1Kc2kl2q+2yKbFBnZh3AR4BzgI3A75rZxiov/Yy7n1y6fSyqeBoV9+TiqD9fk2LbQ9A80YISkhSN9G31Xqv8lSwL0o5a3UbiPm6SF0RR92HVr/IkOlEulHI68Ki7/wzAzD4NnA88HOFnhiLuycWt+HxNik2/oHmiBSUkKRrp25Z6rfowyaqg7aiVbSTu4yaZL+y6D6t+lSfRivL0y2OBxyvu7y09ttCFZrbbzO4ws/URxhNY3JOLW/X5mhSbbkHzRAtKSFI00rcFea3yV7KokXbUqjYS93GTLBZm3YdVv8qTaEU5qKuWPQt/Y/0n4Hh33wT8C/CJqhsyu9LMRsxsZHR0NOQwF4t7gmncny/L14qcDZonyicJIkk52+hrJXtafVyQJElsG0mMKWnSnLNh1a/yJFpRDur2ApW/vK0Dnqx8gbuPuftU6e5NwFC1Dbn7R9192N2HBwcHIwm2UtwTTOP+fFm+VuRs0DxRPkkQScrZRl8r2dPq44IkSWLbSGJMSZPmnA2rfpUn0YpyTt13gZPM7ATgCeAi4M2VLzCzY9z9qdLdNwCPhPXhhYIzNjHd1LnE5QmmC8/5bdUE/Dg/P2i5Lad8JRwDvV188m1nsG98irGJae7c9Th/9FsbGOjtYnR8aq5uBnq7FuXT9ouHGOjtinsXJGPq9W0L+5TKvB1c1cM7X3cSJxzVRy7n7Bs/zEy+oL5HMinuY5R6MV1/zx4uHFrPmr5u1vb31P2eieo4Qscniy23TNb0dbPjstN5bOwQR63qpre7k84Ow3EKBU/N8XW7M/foVp0xs3OBvwc6gJvd/S/N7H8AI+5+t5n9FcXBXB54Fni7u/+o3jaHh4d9ZGSk7ueGMREz7k4hjs8PY+GNBHacsQcUJGcbVa0Otl88xEsGV/HoMxOL6ubfH9XHU88fXjQATGidZV3sFRJFzpZV69uAqn3KSYOreH5qhqcOHOaq23cxuKqHPzl7A9fesTsNfU9WxF7wUeZrUsV9jFJNPl9gz75xrrpt17IXQmpWwO1mKmfDOibe8/Q419+zh7e86gTefWfzfXASczcFAhVQpIO6KARpCKPjU1xww8555+2uG+jlrms2M9jfE3WIqRW03FJWvrH3FFF03rXq4LNXnckbt38r8OMJrbOsa8ucradenwLMPbf94iHe/6WHlcfJkrl8leoaOTaI6jgi4HYzlbNhlHV5G+89b6P64HgEytkoT7+MjSZiNkcLb6RHrTqYmS009LjqTJJgqT6l/Nzq3i7lsUhCJWEhJB2fLBZGmZS3oT442aJcKCU2mojZHC28kR616qCrI9fQ46ozSYJ6fUrlcwcmZ5THIgmVhIWQdHyyWBhlUt6G+uBka8tBXXkBiTuuPpPtFw/x+o1rEzERs1BwRseneGL/IUbHpygUojn1tdnPKU9gLTfYWhNYg75OwlGtPmvl+NpVPVXrptbjqrOltardZkE+X+DJA5M8NjbBkwcmyecLQP0+pfK5bV//Kddt2RRJHquepZ1Vy++wc75eO174WeWFkMJuyzo+eUG5zAuFAtsvHlpWmZTL9c5dj/OBC+f3wbe+9TTyswX1nQnQdnPqai0gsWFtP52d8Y1hW7W4yHI/pw1Xv4w9qOWcO1+tPndcdjpT+ULNHK9VNymqs8SIaVGg2Cslivke+XyBHz09ztW3v7CIwratQ7z06Pp5C/P7m97uDvIFD3X1y5Qt/pQ0sReQ5tTVVyu/ezpzXHLzd0JfqKSRhZD2T87Esfpl2+fswjp//ca1/Pf/tJGOnDVd1uVyLRQKFBym8oXiIM6dP/7c99V3RitQYbbdL3XVrlZ/1W272D85k7i4rtgxwtjEdKI+J5czBvt7OHZgJYP9PTUbZdDXyfJUq8/Hxg7VzfFadaM6a1yr2m0W7Ds4NTegg2JZXn37LvYdLF6qtF5+Vj53ZF8Pa/tXhJrHqmdpZ7Xy+7GxQ6HnfLV2XOvz90/ORPKdpO+6xXX+1Yf38eaP3U93Z0fTZVIu16OP6KWzI8fWj9/Ps4em5wZ0oL4zbm03qEvqJNlWxZXU/ZfmVKvPld0dquMWUXsKT63FevKzhZgieoHqWdpZrfxe2d2x6LEocl7tq/WiLnMtnJJMbTeoS+ok2VbFldT9l+ZUq89D07Oq4xZRewpPrcV6Ojvi/xpSPUs7q5Xfh6ZnFz0WRc6rfbVe1GWuhVOSKf5v05CEOSE0CmFN3l1qYrMmCbeXavV53JqV3HTx/Me2XzzEQG9XnKG2JbWnojAWVFi7qodtW+f3zdu2DjHY1x37AiWqZ4H2XSynVn4ft2ZlS3I+zPbVrnUUtuWW+VLlPNDbxfaLh6ounKK+Mz5tsVBKFBNCo7DchSqCTubXghjzxL7jy50QvbA+B3q7+OX+Qzw2doiV3R0cmp7lyL4u+ld0cfyavizXdSRiaE+xV2Blzoa5iEg+X2DfwSnyswU6O3IM9nXz6DMTiVigRP1m02IvpDAWnWj3xXJqLWDSqpwPo32FWEexV2grFvdptsyXKufy89ffs4cLh9azbqCXI3q7MFDfGZ1ABdoWg7ryle7b/Qr3WdnPkMXes4TdedfKg/ef/3JefuwRyoX0S1TORtnvqE9rC4nK12YpF5MvxDpqi5yNylLlrLYSi+ysfpmVSbhZ2U+pr96kd+WChC3Kfkd9miSFcjH5VEetsVQ5qx6Sqy0Gda2chBvn+dyabCxQf9J7ZS5o7oGEIcp+p9ltK7clbPp+jU/Q9qw6Cl+1sl+qnFUPydUWg7pWTXIvn0d8wQ072fyB+7jghp3seXq8ZQcUmswvUD0PrtuyiePWrJzLhbhzVdq34xhdAAAUaklEQVRHlP1OM9tWbksU9P0aj0bas+ooXLXKfqC3q245qx6Sqy3m1EFrJrkn4TxiTeZvWOyFE8W584WC88zEFIdnCnQY9HZ3sLr3hVxIQq5K0xKXs1H2O41uW7mdOInL12bp+7X1Gm3PIdVR7JWahDl19cp+TV933XJWW2m5QIXbGXUUrVK+0n2UknAecSv2U5IvlzPW9q+o+XwSclXaR5T9TqPbVm5LVPT92nqNtmfVUXjqlf1S5ax6SKZIT780s7PNbI+ZPWpm76nyfI+Zfab0/P1mdnyU8SyXziOWtFCuSrtSbou0D7Xn+Kjs209kv9SZWQfwEeC3gL3Ad83sbnd/uOJllwP73f3fm9lFwAeANzX6WYcP5xmbnMYM3KHgTk9nB50dxky+wKyDu89d4+v5qRkmp2eZdWdFVwdH9fUE+tm4fB7xwmt36DxiaUSh4OyfnOLwdIF8wenMGSu6ckzNFJgp3e/t7mBVVwejE9MvvKY7x+R0ga6c0dOVY3rWmckXqp76UCtXB3q7GB2f0ikTsmzl02+m8rMYYAaduRyz7uBOwSFnUFjQJ0+WFvRZvaKT0YlpZmYLpZzPcXjG5/rqWrnZ6n5Ypxml01L11ki9hrWtsHKpcjtdnTk6c8V21dvdQb5Q/F6o/HvhZzUTR/k9hcILx1SVn93s/ixsz+XrDE/nZxkdn6q6TbXJcCynL612/dxnJ6c5PDNLT0eOAjCTL5DLGV05w8zmcqYyLwd6u9g/OcN0fpYVXTmm8870bIGOnNHblWNgZbDj8yAxJjVPwowzytMvTwcedfefAZjZp4HzgcpB3fnA+0p/3wH8g5mZNzDR7/DhPD8Zm+DD9/6Yt7zqBN595+655PzgRSfTkTPe8cnvzT1261tPY//ENH/02e83fPHKXM7YcHQ/d12zOfFJIslUKDi/GJtg7ODUXA6+fuNa/uC1J/H2f3ygIk+HeWLGefvtu+Yeu3HrELt+/gwbjjmCNau6+Zuv7OGrD++rmsPVcnWgt4ufjB5s24vrSutUuzjtP7z5FGbyBW76vz/jLa86gU/8288X9cnXbdnE//rnPZx+/GrOO3ndovz+8L0/rpnTZa3sh9v9gtTtKujFk4PUa1jbCiuXqm3nui2buOuBJ7jg1GO59o7dDK7q4U/O3sC1d+xe9FlAw3FUXmy6VpsePTjV1P5UtudCocAzE9O8+WP314xNbTI8zfalC+vg9RvX8s7XvYSrb9/F4Koe/tu5L513jH3dlk2s7O6gpyvH+GR+3rFP5fsW5ux1WzZxdH+e44/qi/NC9ZEKO84oT788Fni84v7e0mNVX+PueeA5YE0jHzI2Oc3bb9/FhUPr5zoaKJ4X/K5PP8j+iZl5jz3+7ORcQpUfu2LHCGMT04E+r3we8bEDKxnsb/5/ECSbxiameWzs0LwcvHBo/dyADij9m5s74C0/9vbbd/Hajcdw7R27eWL/YS4cWj/3XLUcXpir+ydn5jqOeu8TWcrYxPSiXNo/McMfffb7c31xtT752jt2c/VZJ7Jl+MVV83upnC5rVT9cbT/VZpJvqXprpF7D2lZYuVRtO9fesZsrXv2rcwfEV5914tzfCz+rmTjK76nXppfTNsrtOZfLcdVtu+rGpjYZrmb60oV1cOHQeq4u9edXn3XiomPsa+/YzbMTM3TmOhYd+1S+b2HOXnvHbh579lBTdZuWPAk7zigHddUyY+EvcEFeg5ldaWYjZjYyOjo677l8wdm7f5LVvV01L8hcaWV3hybZS+Rq5ex0fnZRDlbL3ZxRNU/dfS6vV/d2zXtuqRzWAhNST71+dqFquVTO63I+1+qTV/d20ZGzms9V3o87N9VmkqtevoZ58eSwthVWLtXaTmWbqtX2pvOzTcVRfk+9Nt3s/gTZt8ptprlNNtLHJtnCOqjMi3rH4guPa4K+r5m6TUuehB1nlIO6vcD6ivvrgCdrvcbMOoEjgGcXbsjdP+ruw+4+PDg4OO+5zpyxbqCXA5MzNS/IXOnQ9KwmhkrkauVsd2fHohyslrsFp2qemtlcXh+YnJn33FI5rEnRUk+9fnaharlUzutyPtfqkw9MzjBb8JrPVd6POzfVZpKrXr6GefHksLYVVi7V2k5lm6rV9ro7O5qKo/yeem262f0Jsm+V20xzm2ykj02yhXVQmRf1jsUXHtcEfV8zdZuWPAk7zigHdd8FTjKzE8ysG7gIuHvBa+4G3lL6ewvwtUbm0wGs6e3mxq1D3LnrcT5w4aZ5F0P84EUnM9DXNe+x9Uf2cv0bX6GLJkos1vR1c9yalfNy8M5dj3Pj7506LyehwI1bh+Y9duPWIb728FNct2UTxw6s4M5dj889FySHdcFQCUu1XBro6+L6N75iri+u1idft2UT277+U+4Y+WXV/G40p6OmNpNOS9VbI/Ua1rbCyqVq27luyyZu+ubPuG5Lsb1t+/pP5/5e+FnNxFF+T702HUbbCBKb2mT8FtbBnbseZ1upP9/29Z8uOsa+bssmjuzrIl+YXXTsU/m+hTl73ZZNHHfkyqbqNi15EnackV583MzOBf4e6ABudve/NLP/AYy4+91mtgK4DTiF4i90F5UXVqml2gUbm1/9ElZ05QKvfimpFHvFVruQc3n1y9mC01Gx+mW+dH+5q1/WkpbVoDIu9goJcmHcylyC4inDHTVXv4SezlzV1S/z5ZXOAq5+2WpqM0uKvTCq5atWv0zH6pf19q3WNkMox0TmbJrUXv2yQE+HFVe/nC2QM61+uZSAccZ/8XF3/zLw5QWP/XnF34eB31nu56xY0cmxK4LvypGdPdC33E8VaU4uZ6zpWxEoB4/tXpDXy8xbXTBUwtJ0LlXk8K+s7q39uoRQm0mnMC+eHNa2wsqlqttp4LuhmTjqvifE46kgsalNxq9aHaztX9HUtqKqy7TkSZhxRnrxcREREREREYmWBnUiIiIiIiIppkGdiIiIiIhIimlQJyIiIiIikmKRrn4ZBTMbBR5b8PBRwDMxhJMkWS+DWvv/jLuf3epgyszsSuAvWZyzWZX1PK2UyJyFRf1s2upM8UavMuak5Sskr0wVT32tjidJOZu0ulgoyfElOTYIN75AOZu6QV01Zjbi7sNxxxGnrJdBUvc/qXHFReXxgrSURVriLFO80Ut6zEmLT/HUl7R4Winp+57k+JIcG8QTn06/FBERERERSTEN6kRERERERFKsXQZ1H407gATIehkkdf+TGldcVB4vSEtZpCXOMsUbvaTHnLT4FE99SYunlZK+70mOL8mxQQzxtcWcOhERERERkaxql1/qREREREREMin1gzozO9vM9pjZo2b2nrjjaQUz+4WZ/cDMHjSzkdJjR5rZPWb2k9K/A3HHGSYzu9nM9pnZQxWPVd1nK/pQKSd2m9mpMcWcudwsM7P1ZnafmT1iZj80s3eVHm/rPF2KmXWY2ffM7Eul+yeY2f2l8viMmXXHHWOlNORw0vvDtPVdNeJ9n5k9USrjB83s3Irn/rQU7x4z+4+tjrdS3Plap9+rWX4tiCkx7cPMNlSUwYNm9ryZ/WGc5ROXavUSczyB+6kExZeYvEnKMU+qB3Vm1gF8BDgH2Aj8rpltjDeqlnmNu59csVzqe4B73f0k4N7S/XZyK7DwGh219vkc4KTS7UrgxhbFOCfjuQmQB/7Y3X8NeCXw+6X9b/c8Xcq7gEcq7n8AuL5UHvuBy2OJqoqU5XCS+8NbSVHfRfV4oZinJ5duXwYo5cNFwMtK77mhlDctl5B8rdXvQZXya6FEtA9331MuA2AIOATcVXo6zvKJy8J6idOtBO+n4nArAfulmCTimCfVgzrgdOBRd/+Zu08DnwbOjzmmuJwPfKL09yeA344xltC5+zeBZxc8XGufzwd2eNG3gdVmdkxrIp2T6dx096fc/YHS3+MUBzLH0uZ5Wo+ZrQP+E/Cx0n0DXgvcUXpJ0sojzTmcmDxLW99VI95azgc+7e5T7v5z4FGKeROH2PO1Tr+XNEloH68Dfurujy35Solcg/1UyzXYL7VcUo550j6oOxZ4vOL+XpLZgYbNga+a2S4zu7L02NHu/hQUkwtYG1t0rVNrn5OQF0mIIRHM7HjgFOB+spmnZX8P/AlQKN1fAxxw93zpftJyJC05nMb+MMl9Vy3vKJ0SenPFKURJijdJsSzs96B6+bVCUtvHRcCnKu7HVT5xqVYvSZOEPFlK4vImzmOetA/qrMpjWVjOc7O7n0rxNJPfN7NXxx1QwiQhL5IQQ+zMbBVwJ/CH7v583PHExczOA/a5+67Kh6u8NEk5kvT4ytqpP0xqmd8InAicDDwF/G3p8STFm5hYqvR7tcqvFRLXPqw4d/gNwOdKD8VZPnFJXL2kUOLyJu5jnrQP6vYC6yvurwOejCmWlnH3J0v/7qN4PvrpwNPl03RK/+6LL8KWqbXPSciLJMQQKzProti5/aO7f770cBbzFGAz8AYz+wXF08JeS/GXu9Vm1ll6TdJyJBU5nNL+MMl91yLu/rS7z7p7AbiJF06xTFK8iYilWr9Xp/wil9D2cQ7wgLs/XYottvKJS416SZq486SupOVNEo550j6o+y5wkhVXkOum+HP+3THHFCkz6zOz/vLfwOuBhyju91tKL3sL8MV4ImypWvt8N3CJFb0SeK7883cLZS43K5Xmi30ceMTd/67iqSzmKe7+p+6+zt2Pp5gLX3P33wPuA7aUXpa08kh8Dqe4P0xy37XIgnl9F1AsYyjGe5GZ9ZjZCRQXePlOq+MriT1fa/V7dcov6niS2j5+l4pTL+Mqn7jUqZekiTtP6kpS3iTmmMfdU30DzgV+DPwU+LO442nB/v4q8P3S7YflfaY4P+de4Celf4+MO9aQ9/tTFH9en6H4P7KX19pniqfhfKSUEz8AhmOKOVO5uWDff4PiqU+7gQdLt3PbPU8Dls1ZwJdKf/8qxYPgRymeitQTd3wLYk10DqehP0xb31Uj3ttK8eymeJByTMXr/6wU7x7gnJjzIdZ8rdPv1Sy/iONJXPsAVgJjwBEVj8VSPnHdatVLzDEF7qcSFF9i8qZO229pGVopGBEREREREUmhtJ9+KSIiIiIikmka1ImIiIiIiKSYBnUiIiIiIiIppkGdiIiIiIhIimlQJyIiIiIikmIa1IlIZMzsnWb2iJntN7P3xB2PSCUzW21m15T+PsvMvtTg+19qZg+a2ffMbKi8LRGRrDGzf4s7hqzToC5FShekVZ1JmlwDnOvuA+7+1wufNLPOGGISKVtNMUeb9dvAF939FIrX3tKgTkQyyd1fFXcMWafr1CWcmR0P/B/gPuBMihc0/HWgF7jD3f+i9LrTgA8CfcAU8DrgEPDXFC923AN8xN23t3QHJLPMbBtwGcWLEt8MnOju7zCzW4FngVOAB4A/Bz5MMa87gfe5+xdjCVoyxcw+DZxPMUdngAngGeDlwC5gq7u7mQ0BfwesKj1/KcX8vRmYpXjB66crtnWPu1/b0p2RzDGzLwDrgRXAB939o2Z2OfBu4EmKFzyeKvW7g8A24MWlt/+hu++MI25pT2Z20N1XmdlZwPuo3pdWO1adAW4EhoE88F/c/T4zu5Tif5x1lLbzt0A3cHHpvee6+7NmdiLwEWCQ4nHvFe7+o5bsdMLof8nTYQPwVne/xsyOLCVxB3CvmW0CfgR8BniTu3/XzF4ETAKXA8+5+2lm1gPsNLOvuvvPY9sTyQx3v9rMzgZeA5y34OmXAL/p7rNm9j+Br7n7ZWa2GviOmf2Lu0+0OmbJnPcAL3f3k0sHIl8EXkbxgHgnsNnM7qf4nw7nu/uomb0J+MtSvm4DDrr735T+A+7l7n5yHDsimXRZ6XigF/iumf1v4L3AqcA48DXg+6XXfhC43t3/1cxeDHwF+LU4gpZMOIXFfel3qH6s+i4Ad/91M3sp8FUze0lpOy8vbWsF8Cjwbnc/xcyuBy4B/h74KHC1u//EzM4AbgBe26odTRIN6tLhMXf/dunvN5rZlRTr7hhgI+DAU+7+XQB3fx7AzF4PbDKzLaX3HgGcBGhQJ3H7nLvPlv5+PfAGM/uvpfsrKP5v8iOxRCZZ9h133wtgZg8CxwMHKB5Y3GNmUPxf46fiClCkwjvN7ILS3+sp/oLxDXd/FsDMPkfxP9AAfhPYWMphgBeZWb+7j7cyYMmMan3pc1Q/Vv0Niv9xhrv/yMwe44W8va+Uo+Nm9hzwT6XHf0Dx+HYV8CrgcxW53RPxviWWBnXpMAFgZicA/xU4zd33l05jWwEYxYHdQgb8gbt/pVWBigRU+SucARe6+564ghEpmar4e5bid6QBP3T3M+MJSWSx0i/Lvwmc6e6HzOzrFE/9rfXrW6702snWRCgZV6svrXWsGmQ7hYr7hdI2c8ABnSFRpEU30uVFFA+GnzOzo4FzSo//CPiV0rnKmFl/aQGKrwBvN7Ou0uMvMbO+GOIWqecrwB9Y6b/ZzOyUmOOR7BgH+pd4zR5g0MzOBDCzLjN7WZPbEgnLEcD+0oDupcArgZXAfzCzgdIxwIUVr/8q8I7yHTPTQbC0Wq1j1W8Cv1d67CUUz9QJ9J+8pV/7fm5mv1N6v5nZK6IIPg00qEsRd/8+8D3ghxQn6O8sPT4NvAn4sJl9H7iH4i94HwMeBh4ws4eA7ejXWUme9wNdwO5Snr4/5ngkI9x9jOJc44eA62q8ZhrYAnyg1L8+SPF0n5rbMrOq2xIJ0T8DnWa2m2Kf+W3gCeB/AvcD/0Lx+/+50uvfCQyb2W4zexi4uvUhS5bVOVa9Aegwsx9QnHN3qbtP1d7SIr8HXF7a5g8pLliVSVr9UkRERKQNmNkqdz9Y+gXkLuBmd78r7rhEJHr6pU5ERESkPbyvtDDFQxQXRftCzPGISIvolzoREREREZEU0y91IiIiIiIiKaZBnYiIiIiISIppUCciIiIiIpJiGtSJiIiIiIikmAZ1IiIiIiIiKaZBnYiIiIiISIr9f4nvssTL/ZvpAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 900x180 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.pairplot(chredlin, x_vars=[\"race\",\"fire\",\"theft\",\"age\",\"income\"], y_vars=\"involact\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>        <td>involact</td>     <th>  R-squared:         </th> <td>   0.509</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.499</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   46.73</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 25 Sep 2018</td> <th>  Prob (F-statistic):</th> <td>1.78e-08</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>15:47:22</td>     <th>  Log-Likelihood:    </th> <td> -28.016</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    47</td>      <th>  AIC:               </th> <td>   60.03</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    45</td>      <th>  BIC:               </th> <td>   63.73</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     1</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>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th> <td>    0.1292</td> <td>    0.097</td> <td>    1.338</td> <td> 0.188</td> <td>   -0.065</td> <td>    0.324</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>race</th>  <td>    0.0139</td> <td>    0.002</td> <td>    6.836</td> <td> 0.000</td> <td>    0.010</td> <td>    0.018</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 2.747</td> <th>  Durbin-Watson:     </th> <td>   1.590</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.253</td> <th>  Jarque-Bera (JB):  </th> <td>   2.582</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.549</td> <th>  Prob(JB):          </th> <td>   0.275</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 2.666</td> <th>  Cond. No.          </th> <td>    70.2</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:               involact   R-squared:                       0.509\n",
       "Model:                            OLS   Adj. R-squared:                  0.499\n",
       "Method:                 Least Squares   F-statistic:                     46.73\n",
       "Date:                Tue, 25 Sep 2018   Prob (F-statistic):           1.78e-08\n",
       "Time:                        15:47:22   Log-Likelihood:                -28.016\n",
       "No. Observations:                  47   AIC:                             60.03\n",
       "Df Residuals:                      45   BIC:                             63.73\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const          0.1292      0.097      1.338      0.188      -0.065       0.324\n",
       "race           0.0139      0.002      6.836      0.000       0.010       0.018\n",
       "==============================================================================\n",
       "Omnibus:                        2.747   Durbin-Watson:                   1.590\n",
       "Prob(Omnibus):                  0.253   Jarque-Bera (JB):                2.582\n",
       "Skew:                           0.549   Prob(JB):                        0.275\n",
       "Kurtosis:                       2.666   Cond. No.                         70.2\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lmod = sm.OLS(chredlin.involact, sm.add_constant(chredlin.race)).fit()\n",
    "lmod.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.subplot(1, 2, 1)\n",
    "plt.scatter(chredlin.race, chredlin.fire)\n",
    "plt.xlabel(\"Race\")\n",
    "plt.ylabel(\"Fire\")\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.scatter(chredlin.race, chredlin.theft)\n",
    "plt.xlabel(\"Race\")\n",
    "plt.ylabel(\"Theft\")\n",
    "\n",
    "plt.tight_layout()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Full model with diagnostics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>        <td>involact</td>     <th>  R-squared:         </th> <td>   0.752</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.721</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   24.83</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 25 Sep 2018</td> <th>  Prob (F-statistic):</th> <td>2.01e-11</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>15:47:22</td>     <th>  Log-Likelihood:    </th> <td> -12.014</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    47</td>      <th>  AIC:               </th> <td>   36.03</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    41</td>      <th>  BIC:               </th> <td>   47.13</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>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>      <td>   -1.1855</td> <td>    1.100</td> <td>   -1.078</td> <td> 0.288</td> <td>   -3.408</td> <td>    1.036</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>race</th>           <td>    0.0095</td> <td>    0.002</td> <td>    3.817</td> <td> 0.000</td> <td>    0.004</td> <td>    0.015</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>fire</th>           <td>    0.0399</td> <td>    0.009</td> <td>    4.547</td> <td> 0.000</td> <td>    0.022</td> <td>    0.058</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>theft</th>          <td>   -0.0103</td> <td>    0.003</td> <td>   -3.653</td> <td> 0.001</td> <td>   -0.016</td> <td>   -0.005</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age</th>            <td>    0.0083</td> <td>    0.003</td> <td>    3.038</td> <td> 0.004</td> <td>    0.003</td> <td>    0.014</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>np.log(income)</th> <td>    0.3458</td> <td>    0.400</td> <td>    0.864</td> <td> 0.393</td> <td>   -0.462</td> <td>    1.154</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 2.243</td> <th>  Durbin-Watson:     </th> <td>   2.161</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.326</td> <th>  Jarque-Bera (JB):  </th> <td>   1.355</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.170</td> <th>  Prob(JB):          </th> <td>   0.508</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.759</td> <th>  Cond. No.          </th> <td>2.01e+03</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 2.01e+03. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:               involact   R-squared:                       0.752\n",
       "Model:                            OLS   Adj. R-squared:                  0.721\n",
       "Method:                 Least Squares   F-statistic:                     24.83\n",
       "Date:                Tue, 25 Sep 2018   Prob (F-statistic):           2.01e-11\n",
       "Time:                        15:47:22   Log-Likelihood:                -12.014\n",
       "No. Observations:                  47   AIC:                             36.03\n",
       "Df Residuals:                      41   BIC:                             47.13\n",
       "Df Model:                           5                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==================================================================================\n",
       "                     coef    std err          t      P>|t|      [0.025      0.975]\n",
       "----------------------------------------------------------------------------------\n",
       "Intercept         -1.1855      1.100     -1.078      0.288      -3.408       1.036\n",
       "race               0.0095      0.002      3.817      0.000       0.004       0.015\n",
       "fire               0.0399      0.009      4.547      0.000       0.022       0.058\n",
       "theft             -0.0103      0.003     -3.653      0.001      -0.016      -0.005\n",
       "age                0.0083      0.003      3.038      0.004       0.003       0.014\n",
       "np.log(income)     0.3458      0.400      0.864      0.393      -0.462       1.154\n",
       "==============================================================================\n",
       "Omnibus:                        2.243   Durbin-Watson:                   2.161\n",
       "Prob(Omnibus):                  0.326   Jarque-Bera (JB):                1.355\n",
       "Skew:                           0.170   Prob(JB):                        0.508\n",
       "Kurtosis:                       3.759   Cond. No.                     2.01e+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, 2.01e+03. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lmod = smf.ols(formula='involact ~ race + fire + theft + age + np.log(income)', data=chredlin).fit()\n",
    "lmod.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(lmod.fittedvalues, lmod.resid)\n",
    "plt.ylabel(\"Residuals\")\n",
    "plt.xlabel(\"Fitted values\")\n",
    "plt.axhline(0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Trick needed to get statsmodels plots not to appear twice:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda/lib/python3.7/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig=sm.qqplot(lmod.resid, line=\"q\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Make the partial residual plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pr = lmod.resid + chredlin.race*lmod.params['race']\n",
    "plt.scatter(chredlin.race, pr)\n",
    "plt.xlabel(\"Race\")\n",
    "plt.ylabel(\"partial residuals\")\n",
    "xl,xu = [0, 100]\n",
    "plt.plot([xl,xu], [xl*lmod.params['race'], xu*lmod.params['race']])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pr = lmod.resid + chredlin.fire*lmod.params['fire']\n",
    "plt.scatter(chredlin.fire, pr)\n",
    "plt.xlabel(\"Fire\")\n",
    "plt.ylabel(\"partial residuals\")\n",
    "xl,xu = [0, 40]\n",
    "plt.plot([xl,xu], [xl*lmod.params['fire'], xu*lmod.params['fire']])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Sensitivity Analysis\n",
    "\n",
    "Generate all combinations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(),\n",
       " (1,),\n",
       " (2,),\n",
       " (3,),\n",
       " (4,),\n",
       " (1, 2),\n",
       " (1, 3),\n",
       " (1, 4),\n",
       " (2, 3),\n",
       " (2, 4),\n",
       " (3, 4),\n",
       " (1, 2, 3),\n",
       " (1, 2, 4),\n",
       " (1, 3, 4),\n",
       " (2, 3, 4),\n",
       " (1, 2, 3, 4)]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import itertools\n",
    "inds = [1, 2, 3, 4]\n",
    "clist = []\n",
    "for i in range(0, len(inds)+1):\n",
    "    clist.extend(itertools.combinations(inds, i))\n",
    "clist"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = chredlin.iloc[:,[0,1,2,3,5]].copy()\n",
    "X.loc[:,'income'] = np.log(chredlin['income'])\n",
    "betarace = []\n",
    "pvals = []\n",
    "for k in range(0, len(clist)):\n",
    "    lmod = sm.OLS(chredlin.involact, sm.add_constant(X.iloc[:,np.append(0,clist[k])])).fit()\n",
    "    betarace.append(lmod.params[1])\n",
    "    pvals.append(lmod.pvalues[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Construct list of model names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['race',\n",
       " 'race+fire',\n",
       " 'race+theft',\n",
       " 'race+age',\n",
       " 'race+logincome',\n",
       " 'race+fire+theft',\n",
       " 'race+fire+age',\n",
       " 'race+fire+logincome',\n",
       " 'race+theft+age',\n",
       " 'race+theft+logincome',\n",
       " 'race+age+logincome',\n",
       " 'race+fire+theft+age',\n",
       " 'race+fire+theft+logincome',\n",
       " 'race+fire+age+logincome',\n",
       " 'race+theft+age+logincome',\n",
       " 'race+fire+theft+age+logincome']"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vlist = ['race']\n",
    "varnames = np.array(['race','fire','theft','age','logincome'])\n",
    "for k in range(1, len(clist)):\n",
    "    vlist.append('+'.join(varnames[np.append(0,clist[k])]))\n",
    "vlist"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>beta</th>\n",
       "      <th>pvals</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>race</th>\n",
       "      <td>0.0139</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+fire</th>\n",
       "      <td>0.0089</td>\n",
       "      <td>0.0002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+theft</th>\n",
       "      <td>0.0141</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+age</th>\n",
       "      <td>0.0123</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+logincome</th>\n",
       "      <td>0.0082</td>\n",
       "      <td>0.0087</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+fire+theft</th>\n",
       "      <td>0.0082</td>\n",
       "      <td>0.0002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+fire+age</th>\n",
       "      <td>0.0089</td>\n",
       "      <td>0.0001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+fire+logincome</th>\n",
       "      <td>0.0070</td>\n",
       "      <td>0.0160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+theft+age</th>\n",
       "      <td>0.0128</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+theft+logincome</th>\n",
       "      <td>0.0084</td>\n",
       "      <td>0.0083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+age+logincome</th>\n",
       "      <td>0.0099</td>\n",
       "      <td>0.0017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+fire+theft+age</th>\n",
       "      <td>0.0081</td>\n",
       "      <td>0.0001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+fire+theft+logincome</th>\n",
       "      <td>0.0073</td>\n",
       "      <td>0.0078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+fire+age+logincome</th>\n",
       "      <td>0.0085</td>\n",
       "      <td>0.0041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+theft+age+logincome</th>\n",
       "      <td>0.0106</td>\n",
       "      <td>0.0010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race+fire+theft+age+logincome</th>\n",
       "      <td>0.0095</td>\n",
       "      <td>0.0004</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 beta   pvals\n",
       "race                           0.0139  0.0000\n",
       "race+fire                      0.0089  0.0002\n",
       "race+theft                     0.0141  0.0000\n",
       "race+age                       0.0123  0.0000\n",
       "race+logincome                 0.0082  0.0087\n",
       "race+fire+theft                0.0082  0.0002\n",
       "race+fire+age                  0.0089  0.0001\n",
       "race+fire+logincome            0.0070  0.0160\n",
       "race+theft+age                 0.0128  0.0000\n",
       "race+theft+logincome           0.0084  0.0083\n",
       "race+age+logincome             0.0099  0.0017\n",
       "race+fire+theft+age            0.0081  0.0001\n",
       "race+fire+theft+logincome      0.0073  0.0078\n",
       "race+fire+age+logincome        0.0085  0.0041\n",
       "race+theft+age+logincome       0.0106  0.0010\n",
       "race+fire+theft+age+logincome  0.0095  0.0004"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame({'beta':np.round(betarace,4), 'pvals':np.round(pvals,4)}, index=vlist)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calculate the dfbetas - note these are not the same as leave out one coef changes as they have been scaled to standard error changes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.3995427771699168"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lmod = smf.ols(formula='involact ~ race + fire + theft + age + np.log(income)', data=chredlin).fit()\n",
    "diagv = lmod.get_influence()\n",
    "min(diagv.dfbetas[:,1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since the t-statistic is 3.82, a reduction of 0.4 is not going make a difference to the conclusion."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(diagv.dfbetas[:,2], diagv.dfbetas[:,3])\n",
    "plt.xlabel(\"Change in Fire\")\n",
    "plt.ylabel(\"Change in Theft\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sm.graphics.influence_plot(lmod)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>race</th>\n",
       "      <th>fire</th>\n",
       "      <th>theft</th>\n",
       "      <th>age</th>\n",
       "      <th>involact</th>\n",
       "      <th>income</th>\n",
       "      <th>side</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>60607</th>\n",
       "      <td>50.2</td>\n",
       "      <td>39.7</td>\n",
       "      <td>147</td>\n",
       "      <td>83.0</td>\n",
       "      <td>0.9</td>\n",
       "      <td>7.459</td>\n",
       "      <td>n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60610</th>\n",
       "      <td>54.0</td>\n",
       "      <td>34.1</td>\n",
       "      <td>68</td>\n",
       "      <td>52.6</td>\n",
       "      <td>0.3</td>\n",
       "      <td>8.231</td>\n",
       "      <td>n</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       race  fire  theft   age  involact  income side\n",
       "60607  50.2  39.7    147  83.0       0.9   7.459    n\n",
       "60610  54.0  34.1     68  52.6       0.3   8.231    n"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "chredlin.loc[[60607, 60610],:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Leave out these two case. Could be a more elegant way?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>        <td>involact</td>     <th>  R-squared:         </th> <td>   0.786</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.770</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   50.14</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 25 Sep 2018</td> <th>  Prob (F-statistic):</th> <td>8.87e-14</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>15:47:23</td>     <th>  Log-Likelihood:    </th> <td> -8.9369</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    45</td>      <th>  AIC:               </th> <td>   25.87</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    41</td>      <th>  BIC:               </th> <td>   33.10</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>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>      <td>    0.7533</td> <td>    0.836</td> <td>    0.901</td> <td> 0.373</td> <td>   -0.935</td> <td>    2.441</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>race</th>           <td>    0.0042</td> <td>    0.002</td> <td>    1.848</td> <td> 0.072</td> <td>   -0.000</td> <td>    0.009</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>fire</th>           <td>    0.0510</td> <td>    0.008</td> <td>    6.038</td> <td> 0.000</td> <td>    0.034</td> <td>    0.068</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>np.log(income)</th> <td>   -0.3624</td> <td>    0.319</td> <td>   -1.135</td> <td> 0.263</td> <td>   -1.007</td> <td>    0.282</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 5.938</td> <th>  Durbin-Watson:     </th> <td>   2.390</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.051</td> <th>  Jarque-Bera (JB):  </th> <td>   5.224</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.531</td> <th>  Prob(JB):          </th> <td>  0.0734</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 4.288</td> <th>  Cond. No.          </th> <td>    947.</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:               involact   R-squared:                       0.786\n",
       "Model:                            OLS   Adj. R-squared:                  0.770\n",
       "Method:                 Least Squares   F-statistic:                     50.14\n",
       "Date:                Tue, 25 Sep 2018   Prob (F-statistic):           8.87e-14\n",
       "Time:                        15:47:23   Log-Likelihood:                -8.9369\n",
       "No. Observations:                  45   AIC:                             25.87\n",
       "Df Residuals:                      41   BIC:                             33.10\n",
       "Df Model:                           3                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==================================================================================\n",
       "                     coef    std err          t      P>|t|      [0.025      0.975]\n",
       "----------------------------------------------------------------------------------\n",
       "Intercept          0.7533      0.836      0.901      0.373      -0.935       2.441\n",
       "race               0.0042      0.002      1.848      0.072      -0.000       0.009\n",
       "fire               0.0510      0.008      6.038      0.000       0.034       0.068\n",
       "np.log(income)    -0.3624      0.319     -1.135      0.263      -1.007       0.282\n",
       "==============================================================================\n",
       "Omnibus:                        5.938   Durbin-Watson:                   2.390\n",
       "Prob(Omnibus):                  0.051   Jarque-Bera (JB):                5.224\n",
       "Skew:                           0.531   Prob(JB):                       0.0734\n",
       "Kurtosis:                       4.288   Cond. No.                         947.\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ch45 = chredlin.loc[~chredlin.index.isin([60607, 60610]),:]\n",
    "lmod45 = smf.ols(formula='involact ~ race + fire + np.log(income)', data=ch45).fit()\n",
    "lmod45.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Discussion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>        <td>involact</td>     <th>  R-squared:         </th> <td>   0.743</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.683</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   12.31</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 25 Sep 2018</td> <th>  Prob (F-statistic):</th> <td>6.97e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>15:47:23</td>     <th>  Log-Likelihood:    </th> <td> -5.3371</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    22</td>      <th>  AIC:               </th> <td>   20.67</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    17</td>      <th>  BIC:               </th> <td>   26.13</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>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th> <td>   -0.2344</td> <td>    0.238</td> <td>   -0.986</td> <td> 0.338</td> <td>   -0.736</td> <td>    0.267</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>race</th>      <td>    0.0059</td> <td>    0.003</td> <td>    1.814</td> <td> 0.087</td> <td>   -0.001</td> <td>    0.013</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>fire</th>      <td>    0.0484</td> <td>    0.017</td> <td>    2.866</td> <td> 0.011</td> <td>    0.013</td> <td>    0.084</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>theft</th>     <td>   -0.0066</td> <td>    0.008</td> <td>   -0.787</td> <td> 0.442</td> <td>   -0.024</td> <td>    0.011</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age</th>       <td>    0.0050</td> <td>    0.005</td> <td>    0.993</td> <td> 0.335</td> <td>   -0.006</td> <td>    0.016</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 6.437</td> <th>  Durbin-Watson:     </th> <td>   2.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.040</td> <th>  Jarque-Bera (JB):  </th> <td>   4.177</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.806</td> <th>  Prob(JB):          </th> <td>   0.124</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 4.399</td> <th>  Cond. No.          </th> <td>    277.</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:               involact   R-squared:                       0.743\n",
       "Model:                            OLS   Adj. R-squared:                  0.683\n",
       "Method:                 Least Squares   F-statistic:                     12.31\n",
       "Date:                Tue, 25 Sep 2018   Prob (F-statistic):           6.97e-05\n",
       "Time:                        15:47:23   Log-Likelihood:                -5.3371\n",
       "No. Observations:                  22   AIC:                             20.67\n",
       "Df Residuals:                      17   BIC:                             26.13\n",
       "Df Model:                           4                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "Intercept     -0.2344      0.238     -0.986      0.338      -0.736       0.267\n",
       "race           0.0059      0.003      1.814      0.087      -0.001       0.013\n",
       "fire           0.0484      0.017      2.866      0.011       0.013       0.084\n",
       "theft         -0.0066      0.008     -0.787      0.442      -0.024       0.011\n",
       "age            0.0050      0.005      0.993      0.335      -0.006       0.016\n",
       "==============================================================================\n",
       "Omnibus:                        6.437   Durbin-Watson:                   2.000\n",
       "Prob(Omnibus):                  0.040   Jarque-Bera (JB):                4.177\n",
       "Skew:                           0.806   Prob(JB):                        0.124\n",
       "Kurtosis:                       4.399   Cond. No.                         277.\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lmods = smf.ols(formula='involact ~ race + fire + theft + age', data=chredlin.loc[chredlin.side == 's',:]).fit()\n",
    "lmods.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>        <td>involact</td>     <th>  R-squared:         </th> <td>   0.756</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.707</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   15.45</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 25 Sep 2018</td> <th>  Prob (F-statistic):</th> <td>6.52e-06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>15:47:23</td>     <th>  Log-Likelihood:    </th> <td> -5.9264</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    25</td>      <th>  AIC:               </th> <td>   21.85</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    20</td>      <th>  BIC:               </th> <td>   27.95</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>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th> <td>   -0.3186</td> <td>    0.227</td> <td>   -1.403</td> <td> 0.176</td> <td>   -0.792</td> <td>    0.155</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>race</th>      <td>    0.0126</td> <td>    0.004</td> <td>    2.806</td> <td> 0.011</td> <td>    0.003</td> <td>    0.022</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>fire</th>      <td>    0.0231</td> <td>    0.014</td> <td>    1.655</td> <td> 0.114</td> <td>   -0.006</td> <td>    0.052</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>theft</th>     <td>   -0.0076</td> <td>    0.004</td> <td>   -2.069</td> <td> 0.052</td> <td>   -0.015</td> <td> 6.13e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age</th>       <td>    0.0082</td> <td>    0.003</td> <td>    2.369</td> <td> 0.028</td> <td>    0.001</td> <td>    0.015</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 1.447</td> <th>  Durbin-Watson:     </th> <td>   2.306</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.485</td> <th>  Jarque-Bera (JB):  </th> <td>   0.375</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.068</td> <th>  Prob(JB):          </th> <td>   0.829</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.584</td> <th>  Cond. No.          </th> <td>    278.</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:               involact   R-squared:                       0.756\n",
       "Model:                            OLS   Adj. R-squared:                  0.707\n",
       "Method:                 Least Squares   F-statistic:                     15.45\n",
       "Date:                Tue, 25 Sep 2018   Prob (F-statistic):           6.52e-06\n",
       "Time:                        15:47:23   Log-Likelihood:                -5.9264\n",
       "No. Observations:                  25   AIC:                             21.85\n",
       "Df Residuals:                      20   BIC:                             27.95\n",
       "Df Model:                           4                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "Intercept     -0.3186      0.227     -1.403      0.176      -0.792       0.155\n",
       "race           0.0126      0.004      2.806      0.011       0.003       0.022\n",
       "fire           0.0231      0.014      1.655      0.114      -0.006       0.052\n",
       "theft         -0.0076      0.004     -2.069      0.052      -0.015    6.13e-05\n",
       "age            0.0082      0.003      2.369      0.028       0.001       0.015\n",
       "==============================================================================\n",
       "Omnibus:                        1.447   Durbin-Watson:                   2.306\n",
       "Prob(Omnibus):                  0.485   Jarque-Bera (JB):                0.375\n",
       "Skew:                          -0.068   Prob(JB):                        0.829\n",
       "Kurtosis:                       3.584   Cond. No.                         278.\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lmodn = smf.ols(formula='involact ~ race + fire + theft + age', data=chredlin.loc[chredlin.side == 'n',:]).fit()\n",
    "lmodn.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/json": {
       "Software versions": [
        {
         "module": "Python",
         "version": "3.7.0 64bit [Clang 4.0.1 (tags/RELEASE_401/final)]"
        },
        {
         "module": "IPython",
         "version": "6.5.0"
        },
        {
         "module": "OS",
         "version": "Darwin 17.7.0 x86_64 i386 64bit"
        },
        {
         "module": "pandas",
         "version": "0.23.4"
        },
        {
         "module": "numpy",
         "version": "1.15.1"
        },
        {
         "module": "matplotlib",
         "version": "2.2.3"
        },
        {
         "module": "seaborn",
         "version": "0.9.0"
        },
        {
         "module": "scipy",
         "version": "1.1.0"
        },
        {
         "module": "patsy",
         "version": "0.5.0"
        },
        {
         "module": "statsmodels",
         "version": "0.9.0"
        }
       ]
      },
      "text/html": [
       "<table><tr><th>Software</th><th>Version</th></tr><tr><td>Python</td><td>3.7.0 64bit [Clang 4.0.1 (tags/RELEASE_401/final)]</td></tr><tr><td>IPython</td><td>6.5.0</td></tr><tr><td>OS</td><td>Darwin 17.7.0 x86_64 i386 64bit</td></tr><tr><td>pandas</td><td>0.23.4</td></tr><tr><td>numpy</td><td>1.15.1</td></tr><tr><td>matplotlib</td><td>2.2.3</td></tr><tr><td>seaborn</td><td>0.9.0</td></tr><tr><td>scipy</td><td>1.1.0</td></tr><tr><td>patsy</td><td>0.5.0</td></tr><tr><td>statsmodels</td><td>0.9.0</td></tr><tr><td colspan='2'>Tue Sep 25 15:47:23 2018 BST</td></tr></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{|l|l|}\\hline\n",
       "{\\bf Software} & {\\bf Version} \\\\ \\hline\\hline\n",
       "Python & 3.7.0 64bit [Clang 4.0.1 (tags/RELEASE\\_401/final)] \\\\ \\hline\n",
       "IPython & 6.5.0 \\\\ \\hline\n",
       "OS & Darwin 17.7.0 x86\\_64 i386 64bit \\\\ \\hline\n",
       "pandas & 0.23.4 \\\\ \\hline\n",
       "numpy & 1.15.1 \\\\ \\hline\n",
       "matplotlib & 2.2.3 \\\\ \\hline\n",
       "seaborn & 0.9.0 \\\\ \\hline\n",
       "scipy & 1.1.0 \\\\ \\hline\n",
       "patsy & 0.5.0 \\\\ \\hline\n",
       "statsmodels & 0.9.0 \\\\ \\hline\n",
       "\\hline \\multicolumn{2}{|l|}{Tue Sep 25 15:47:23 2018 BST} \\\\ \\hline\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "Software versions\n",
       "Python 3.7.0 64bit [Clang 4.0.1 (tags/RELEASE_401/final)]\n",
       "IPython 6.5.0\n",
       "OS Darwin 17.7.0 x86_64 i386 64bit\n",
       "pandas 0.23.4\n",
       "numpy 1.15.1\n",
       "matplotlib 2.2.3\n",
       "seaborn 0.9.0\n",
       "scipy 1.1.0\n",
       "patsy 0.5.0\n",
       "statsmodels 0.9.0\n",
       "Tue Sep 25 15:47:23 2018 BST"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%load_ext version_information\n",
    "%version_information pandas, numpy, matplotlib, seaborn, scipy, patsy, statsmodels"
   ]
  }
 ],
 "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": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}