{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "import statsmodels.api as sm\n",
    "plt.style.use('ggplot')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data are available in the MOOC repository. The CSV file contains data for the 1314 women that were polled in Whickham, England, in 1972-1974 and were categorized as \"currently smoking\" or \"never smoked\". Each line is related to a person and contains whether she smokes or not, whether alive or dead twenty year after the survey and her age at the time of the survey."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of rows: 1314\n"
     ]
    },
    {
     "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>Smoker</th>\n",
       "      <th>Status</th>\n",
       "      <th>Age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Yes</td>\n",
       "      <td>Alive</td>\n",
       "      <td>21.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Yes</td>\n",
       "      <td>Alive</td>\n",
       "      <td>19.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>No</td>\n",
       "      <td>Dead</td>\n",
       "      <td>57.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>No</td>\n",
       "      <td>Alive</td>\n",
       "      <td>47.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Yes</td>\n",
       "      <td>Alive</td>\n",
       "      <td>81.4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Smoker Status   Age\n",
       "0    Yes  Alive  21.0\n",
       "1    Yes  Alive  19.3\n",
       "2     No   Dead  57.5\n",
       "3     No  Alive  47.1\n",
       "4    Yes  Alive  81.4"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_url = \"https://gitlab.inria.fr/learninglab/mooc-rr/mooc-rr-ressources/-/raw/master/module3/Practical_session/Subject6_smoking.csv\"\n",
    "data = pd.read_csv(data_url)\n",
    "print(\"Number of rows:\", len(data))\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The CSV file does not contain any missing data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of rows with missing values: 0\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of rows with missing values:\", data.isnull().any(axis=1).sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "Let's visualize the number of women alive and dead after twenty years, according to their smoking habits. A heatmap is effective in this case."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApgAAAH0CAYAAACZ7111AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzs3Xd4FNX+x/H3SQ+hVynSO9KbSFMRKyigrgU7Fq6KYO9ef/Zyrw2xFxBQXK9iRRELIIiKIKAUkd47BBJSSDK/P85s2Gx2U2BIKJ/X8+TJ7syZM2f6d86cM2scx0FERERExCtRpV0AERERETm6KMAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTx0xAaYxxjHGXFba5fCCMeZhY8yy0i6HRGaMGWaMWWeMyTHGPHwQ+eTZb40xq4wxD3hSyOKX5SpjTFYJzWu0Mea7kpjXkc4Yc7Ix5i9jzD5jzNTSLo8cedx9yDHG1An3XQpmjKnvrq8eBaQpsXOaMWaqMeYtD/Ip9Hrj1bzCKTTAdFeqY4x5Psy4oyboE+8ZYy4zxhxxL1o1xtQCXgCeBGoD//Ew+85AvmPJC8aYU4wxW40x0YcifzlkXgXmAg2BQcaYHu65tf6hmqExprUxZqx7AUo3xqw0xrxgjKkYki7WGPOMMWajMSbNGDPDGNPxUJVLPPMzUBPYcKhndKSe5w/AcODC0i7EITAIuC3wxRjznTFmtBcZF7UGMw24yRjT1IuZHk6MMXGlXYaSdKwt7wFqiD02PnccZ6PjOCleZew4zlbHcVK9yi/EQGyZsw9R/nJoNAGmOI6z1nGcHV5mXMDx3gFIAa4FWgJDgf7AByHpngWGADdgb45WAN8ZY47zspxeOZrPb8aKLUpax3EyHcfZ5DhOzqEu17HCcZxkx3F2lnY5vOY4zg7HcXYfqswL/ANGA98Bs4DPQsY5wGWRvrvDvgNGB31fBTyKvWtPBrYANwPxwEhgJ7AeuDnMvIYDHwOp2Duz20LSlAVedKffC/wBDAoaX9/NZzAwyc3nPxGWuwPwtVu+FGA2cGZImlXAI+48dwCbsbVd0UFp4oOWdaf7+UlgWSHr/ZAuL9AI+Mgt915gAdAvaNqOwLfusm8FPgHqBY1/GFgGnAcscfP+EWjkjj/ZnXfw32h3XF9gqjvvZGAa0CVk2Rq4808H1gA3udO8FZQmxi3HSjfdQuCGIuzTZwNzgAx3+74CJAUtV2i560fI51LgV3cZtgFfAU0LOUZWAQ+4nx8H/g6T76vAL0XdFkHp1gS2IWCwx1lg/50A3ApkhUzTF5iJvYlcD7wLVCnmcVAJ+NDdBzYDjwFjgO8K2AbjgHFB369219W1QcPGAP6ibLeQc9UwYJ1b3reAWGwAtRp7DL4BxIWUZxh2P04H/gHuB2KKc6xHWM7HgcXYY2wt8BpQoYBj5Koww6YG5XcxMM8t5yrguZB1MBV42932G4GthR0PQdOeD+QA5d3v5dz5XB+UJhrYBDwcIY9ywB7g0pDh9d28Ty7qsYs9/81zt+Mm7D5cM2h8YP2dA8xw87kZKI/djze5+8pa4Lmirodi7PcxwEPAcnc+64GRQePLYp+ErHXHrwLuCxrfDHvOSHH/vgAaB42/CsgCTsGe2zPZf3wH9vG9wGTgCndd1AlZN6Hf+wLT3ekWAWeELFN74Bd3XS4FLiDonBVmPQXyDXeejwWectdLpju/SwtZ7wVuO/bv34+522YX9hiLcrfFZuw58vEw++Xr7rh04Hfg9JD90wF6BA27G3u+COyzowk6p7H/fHM99tyyG/gMqBYy7xEh2+ry4G0TYT1MxZ67HnTXxQ53fsHHelFjlMfdvHZjr1NPA1Gh8wpaptDtGVj++7A3mBnuepwMJBa4PYtwoAVW4onYE8QpQeMONMDcha2SbQw84OY7KWjYve6wliF578AeWE2xJ58s3IAKe0H90V1ZPbC1UNdjd+w+ITvROuAyN02DAg6cK7F3902xO3QmQQGEuyw7gXuwtRAXuWW6OijN8+4OcB7QHHtR2k3RAsxDsrzAcdgD8Tt32kZu+c52p2uJ3WH/zy1za2wwuhRIcNM8jA0ovsEGQG2xJ8Fp7vg4bFDouPM7jv0X1oHYRw1NgVbYnX8HbmDjLts8bPDWBWiH3T+SyRtgjsYGxqe7y3URdt8aUsB6beOux+eBFsBZ2MBsbNBFYZBb7vZuucMGEdigqJ+7/toDn2ODk7iQ7RgpwGzqju8WND4O2A78q6jbwk3XGXthD2yf4e72udKdz13uuskKmuZU7ElvGHb/7Yzdp6YDphjHwUTszcap7vYch93HCwowrwE2BH0fiz1OPggatgYYWpTtFrQ/JGMD0xbAudiLySTgPXcZ+mGD6X8FTfcw9gIx0N2PznbzfrQ4x3qE5XwA6Ik9Fvtgg9gxQdv6OHcfuMn9nOSW23G3x3FAZTf9VW4ZLscey72w+3/wOpjq7gevucvburBzfMg2ScUNrLGBjQPUDUk3tpBt+zrwY8iwR7H7bGC/Gk0hxy52Hz7NHd8N+9h3Wsg52nHX6bluujrAS8B8oCtQFzgJuK6o66EY+/0Y7D57OfYccCJwa9A5bCr2gjwgaHtd545PdPe577Hnz47YY28Z7vnD3d452MDhVDePathzdRb2etkUW8O8maIFmPOBM7H78HvuOq/opimDvSn5Anu8neiu871EDjALOs8/iz2XBc7197nL06eA9V7gtnPXaTI2SGqK3Wcd7DH+jDvsSnfYWUHTfYQ9hs/AnhtedLdn85BrZQ9ssDoSW6nTJuT8EhpgJmNr/U9wy7oa9/h20wxyt9Vwd51f5eZblABzF/Z819zdZruA/yvmProKey5+BHtDczn2GL8tZF6BALMC9vz/YdD2jHOXYzf2KUdd7DV5BF4FmO7nD7BBRJT7/UADzE+Dvke5Bf8iZNhOgmox3bzHhuT9PjAjaGWn4+7cQWneCcwvaCd6sDgnm6C85gP3hyzL5yFpvsG9SGIvFumEnNywd09FCTAPyfJiT/abCLobCrPNJ4QMi8eeaAa43x/GHjjVgtJcjD2BBIKcywCnCOs1sL0Hu9/7uuUOvpuv7M4/cCA0cOfVPCSvh4B5BcxrLPBbyLDz3LzqBa3bAk8AEfKu7E7XPdIxQUhtALa24NWg74Owd4iBoKLQbeEOe4K8tX3ryH8X/z/yBphTgadC0tR1y9yuKMcB9obQAfoGjY/D1loUFITUc6drGVTe24HN7vcm7vimxdhuo7EX/OAA/yvsXXt80LDPgP+5n8u46zL0zv8KYFfIdot4rBdjHxnobt/gGoTQfaQHYWrO3TIMDRnWy01bKWibLg3Ov4jlOg5bW/SfoGGXunmH1vY+CywsIK8O7nRN3O/Rbt53ut8P9Nht7+ZbO+Q4vTwk3WcEXXO8+ouw318QIW0fd3ynCOOHuPtd1aBhNbA3P1e4369y8+gZMu0MYHzIsP9QtAAz+AlX4ObmDPf7ddib2QpBaZq7acIGmG6afOd57HGVAdwYMnwi8EMBeRW47dz9e17IsIXAn2G2VeBpXWBbnR2SZi7wjvu5vpumDzYYXULIUyLCB5hbyXtuuQfYGPR9Jvmv409RyPXFXc4FIcNeA2YVdR91v68CfgpJ8wSwLmRewRU3eWI2d9it2PNKbHGOmeL2Ir8Hu8NdVczpQs0PfHBsG5Gt2LvZ4GFbgOoh080K+T4TG72DvduPA9YbY1ICf9idv0nIdL8VVkBjTDVjzCvGmCXGmF1uXq2wF8Zg80K+r8eeKMDe1cZj7wKDzShs/q5DtbwdgZ+dyG0BOwMDQ/LdDiSE5L3BcZytQd/XY+/cQ7dbHsaYBm4Hg2XGmN3YG4wK7F+3LYFtjuPk9rR3bNu0v4Oy6eTO6/eQct4XZvmDtcLeoQWb5ubVMn/yApejnTFmottBYg+21gvy7yMFeQ+4KKjt2OXYm61AW7yibouB2JM3xpjy2M5Jhe13nYERIXkvcsc1cfMq7DgIrLPceTmOk4mtdYnIcZzV2MejpxpjmgEVsY+8E4wxJ2BrbNY7jrPUnaSo222xO/+ATdhmCBkhwwL7aCtsbdLHIevhdaCCMaZa0HQFHethGWMGGWOmG2M2uPmOZ3/NZZG55agHPBdSzq/dJI2Dks9xitH2zhhTHdsEYwH26VFROBFHOM5c7E30te6gs7DraYz7vUjHrtsTerIxZq17fAX239DjK/T89gpwgdsz/0VjzFnGmGJd64qw33dw/38bIYuOwE7HcX6PML4VsMhxnG2BAY7jbMae41qFpA09llpy4NeU3H3YcZxNQDb79+GW2OMnOSjNEmzNWXE1xu7n4Y7Z0OULVpRtNz/k+yaC4oegYYFjPHB+CC3L9DBleRf7lKi7e44qzOKQc0voOaElthIhWOh1PZICzzfFiFHCxRG13etEUfmxTR5Wux2/LzfGlCtsophizADHcVa7vckfM8b4wyXBnjiChWuUvC/MdOGGFXZSCJ5XFLa6unOYdJkh34vSyWI0tjbnLuyFMA3bBii0EXlo3sHlNkHDvODl8hZUpihsjdFTYcZtL2A+gTwL225fYmuVbsLWbGRiT5DB67awdRaYx0nYmoBw5Ygk0vgibydjTBnsxWUG9jHNJnfUQvLvIwWZgH0M0t8Y8yP28WxwT8VCt4UxpgX28dmkQPHc/0VZh0+7+YcKLM9oCj4OQo/34vgBW2OQja2ZTzPGTHeHneSOD1aU7Vbcc0vg/4XYO/RQwZ1uCjrW8zHGdMXWhjwJ3ImtpT8RG2gVtzNKYD7DsY9SQ60L+lzkTmTGvsZmCvbR7AWO4wSvq43u/+PYf/ME9iK3iYK9Bjxh7CtSrsU+Vdnijiv02DXG1MXuz2Oxj/e2YR9/f0f+dZdneR3HmexOfwa25m4c8Kcxpo9T9A5woyna+b8gB3IeMiHDsx3HST+AvCMJ3Ych7z7s1bUqUn6hy5c3cdG2nZfxQ2hZvsKez8/E3gwWJtw5IfSc6NW2Cl2m0RzYPlrsc7bjOOuNMc2xzWZOxbYNfdoY09VxnLWRpjuQ92A+6U53d5hxW4BagS/GmHiKWStUiBNDvnfDNqAHe8dcEft4dlnI3xqKrxfwiuM4nzuO8yf2ZNuwmHksw+4k3UOGn1TE6Q/V8s4BuhtjkiKM/x3bBmd5mLyL04suE8AEvTbHGFMFu0885TjOZMdxFmEf9QfXei4CqhljGgdNVwnbziR4GcC2Dwst4/ICyrQQ6B0yrDf24F2UP3lELbDtoe53HOdHx3EWYzu7FOvgdWsqv8Q+kr0Ye9PwdVCSomyLQdjHTslunsnYu93Q/S70++9AqzD5LnP295wv7DhY6P7P3afd2thwNz6hfsCu+9OwbdECw/pgLy7BAaZX2y3UQuz+1zDCejiYHvk9sDXxDziO86tbG1uU9xIGLiy5x41bu7UWaBahnOGCkAIZYxoBP2HX36CQmhjY36HqjKBporDbq7AaswnYWvYbsJ1w3gzJFwo+djtja5ZHOI4z03GcvymktjiYY3vGfuA4TmD+vSnetaiw/X6u+//0CNPPASobYzpFGL8QaGWMqRoYYIypgT3HLYwwTcAiCj+2D8QioIUxpkJQmQJPFwqS7zyPvfZlkP+Y7UUhy+fBtgsVmF+vkOE9w5RlPLZd4zvGmCsPYp4Bi7DX7WCh1/UDVdQYJVwcscGJ3HM8k6BzT4DjOBmO43zjOM5d2FreMtj2xREVqwbTnckeY8yD2Eayob4Dhrq1EHuwPTG9fG1EP2PMzdjeS2diG4Zf7I77wZ3/J8aYu7HV6JWwF750x3HeDJNfQf4GBhtjZmBX9iOEWekFcRwn1RjzGrbGN/D4Ywi2mcGWAie2DtXyvoI98X9mjPk3ttFxK+zd8tfYNhq/AeOMMS9imzDUx+5MLzqOs6KIq2Cl+/9cdz2mYWtxtgLXGWOWA1WwjbPTgqb7zl2e94wxw7E7/OPYNp8OgOM4y4wx7wBvGmPuwj4GSMI+mqrmOM7TEcr0LDDXGPMctjdxfWyD7vHFvBFZjT2BDjPG/NfNJ9C2prjGYNtHNsK26Qu+Gy/KthjoLkuw/wKPGmOWYB/RnIsNDII9BHzrPpUYgz1mm2Br8252HCeNQo4Ddzt8DowyxtyA7WxwD7bXZmF+wO6z57rLGRj2DPbcFBxgerXd8nAcJ8UY8wS2tg1sbV4M9gTa3nGccDfSRfU39kZpCLbWsQdwYxGmW41to3i2MeZDIMO9abgfeNsYswv4FFtr0wLbmeGG4hTMGNMSe5wtAG4BqrjLD7bnebbjOLvd89cTxpiN2OP5Tmzg93pB+bvnvnHY/XCNO6/AuKIcu/9gj6XbjTHjsZ0IHyrisj2ODfAWYtfjYGzbwuLsJ0XZ78cDrxhjEtxlqAyc5DjOi9h99yfgQ2PMbdj1XAto4TjOW9j29A+54+/E3pj+B3tj+GEhZfsv8JEx5jdsLW8PbNOagzXeXc733Gt8ojuvNAo+r+U7z7vH1UvYc9BW7OPeC7HtpvtGysijbZeH4zjLjTEfYbfVDdjj61/YjjmXhkk/wRizDxhvjIk7gNgh2H+x2/g3bMXBSdjKBDj42uKixijtjP2xkPexzVOGY/tQRLISOMW9AU12/67AViz+hm0y0Qd7ji/45t4pvGHzaEIa67szmk/+xunHYXug7cbebf+L8J18HgjJbxkhr73ANrJ9LOi7g+219Cn2scpG3EbjQWkSsRf5ldigZBO2If6pTt6GvD2KsNytse1c0twy31jEZXmLvK8VCZyMAxvqDYr+mqJDtrzYO+WJbpn2utvz7JDl/wwbEKa52+gN9nc+eTh0GQjTOQH7mo7N2JPFaHdYb3d+6diD5PzQfQDbEWCKm2Yt9nH6b+R9DUg09vHAEnf5t2Hb+FxYyLoNft3NVuxrgYJf/3AyRejkg32Fxz9uGf9wlysLuCpkO0bs5OMOi8XecDhAxwj7YthtARyP244qzDH6hLtOUrEBbLjXFPXE7td73HSL3W0WEzTvwo6DKtg2Oqnu+nySQl5TFDTtQuxj6EDHQePmke/4KMJ2Gx06T0KOR3fYa7id5YKGDWH/6392Yt9gENzTPNx2y5d3mDI/it3/U7HBwCXkP0by7CPusLuwwUY2ec8nA7DBzF7seXYe8FDQ+KkENdgvoFwPu/MN9xdctlhswL/JXTczidBxJcw82rr53RtmXKHHLvub0KRha0zPdPM7uaDjFPv47i9sYBJ4DVrw62euCl3OCMdcYft9rLt9V7nLsA54IWh8OexN0EZ3/ErgnqDxzdx9IvCaoi8J85qiCOUb7u4faW65rgxeF6HrpoB1FXq+CrymKAN7brsAe266vZBtHe48fyCvKSps200lZP8O3S7usG/I+xq08ux/TVEGRXtN0bnu+r0x3Pkl9Ls7LFyHp1uDttVk7NteHIJeBxdmPYRbzgeAVcXcR1dhK2fexZ4vdmBv1qMjzQtbCzrd3QaOu+8Mcue1E3vu+YsC3tYS+Au8MkIOM8b+MsLljuOMK+2yHA7cBsXrsBf5kaVdnsOFW8N7oeM4EX/iTKQ0GGPOxt4g13Vsh5LDgjHmEexNbVvHcUrkp1OPVMaYetgg5VzHcb4o5eIcFYwxDwHDHcepUtplOdSK/YhcpCQYY87F3l0vxrbP/Df2bipc57Jj2UYKftwhUqLcDnB1sY+A3z+cgktXP2wTEAWXIYz96ef12NrWetja69VE7i0vBTD2l5duZ/8PnZyCbWYyqjTLVVIUYMrhqgz2AlUfe2DOwT6+2FyahTrcOI6jgFsON3dhH+f95n4+rDiO06HwVMesKtgfdaiNfZw6E/uEJLQDmBRN4BHz7dgmEyuxTZeeLcUylRg9IhcRERERTx3Ia4pERERERCJSgCkiIiIinlIbzKNIYvub1d5B5BD64aPHSrsIIke9bo0rHswvhMlhQjWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4ikFmCIiIiLiKQWYIiIiIuIpBZgiIiIi4qmY0i6AyJHisv5defORywtNd/bQkfz4699hx1WvXI7brjqNs3qewPHHVSItYx+Ll29k3Je/MnrirIh5Njy+Kv16t6FXpya0blqbGlXKkZWdw4Ytycz8Yxlv+H/ij8VrD3jZRA4HKbuT+ePX6Sya9zurly9h25ZN5GRnU65CReo3aUGPPufQ8aSTw0675M+5/DX3V1YtW8zWTevZk5xMRvpeypQtT+26Deh40sn0PuM84uITCizDlo3rmPS/sfz1x6/s2rGdxDJJ1GvUlN5nDqBz91MPwVKLHJ2M4zilXQbxSGL7m7UxD6FAgJmdncPWnSmR0931NjP/WJ5vePsWx/P5qJuoWqksAHtS00mIiyU2NhqAKT8v5oIRr5O5LyvPdN3aNuSH0bflGbY7JY34uBji42IByM7O4em3J/Poq18d1DJKwX746LHSLsJRbci5J5GdnZ37PTYunqioKDLS03KHtenUjZvufYr4hLyB4vMP38b82TNzv8cnJALkmbZajVrc/uiLHFe7btj5z589k1FP3kdmRjoAiWWSSE9Pw8nJAaBn335cM/wBjDEHuaRSkG6NK2oFHwVUgylSTOs276T5Of8u1jTlyybw8YtDqVqpLEtWbGLIg+8xd9EaYmOiuWZQd565YxB9T2rBM3cMYsST/jzTxsREk5WVzaTpfzHh69lMm/0PO5JTiYoytG9+PE/fPojuHRpz3/VnsWbjDsZ8GrkmVORwlp2dTcOmrehx2jmc0OFEqtesDcDWzRv4YsK7TP/2cxb8PovRLz/JDXf8X55pW7brzAkdTqRpq7ZUr1mHxDJJgK0VnTV1Mh+Nfpmtmzfw0mN38dio94mKyttCbOumDbzy1P1kZqTTpGUbhox4kONq1yU9bS9ffzyOzz54m5+mfEnNOvU5+4LCn2SIHOvUBlOkBIy4og81q1Vgb1omA4a9ytxFawDYl5XN6/7pPPraJACGDOpO47rV80y7Yu1W2p3/GBfd/iYTv5vHjuRUAHJyHOYsWsNZN4xkwdJ1ANx59ekluFQi3rr7iVE89Pw7nHrO+bnBJdiax2uG38/JZw0EYNaP37B96+Y8054x4BL6nuujXqNmucElQNnyFeh7ro9Lr7sVgA1rVrJ8yZ/55v3JuNfJSE+jQqUqjPj3c7m1nAmJZRh42fWcfOYAAL748F1S9+z2dsFFjkIKMEVKwOB+XQH4aPIcVm/Ynm/8qx9MZU9qOjEx0Vx8dqc849Zv2cXyNVsj5r0vK5sJX80GoFHdalQsl+hhyUVKTou2nQoc3+v0c3M/r/pncbHybtT8hNzPO7ZtyTMuIz2NOT//CMCpZw8iqWy5fNOf47sSgLS9qcz9ZVqx5i1yLFKAKXKINalXnbo1KwPw7cxFYdOkpmXmtts8rVuLYs8jPXN/u83oaB3WcnSKjYvL/ZyTk11Ayvz+Xjgv93P1mnXyjFu6cD6ZGRkAtO50Utjpq9WoRa3j6wPw19xfizVvkWOR2mCKFFPVSmWZOf4umtavQXSUYdO23fwyfwXvTpzFT3P+yZe+VeNauZ8XLt8QMd9FyzZwZo9WNG9wXLHL1KtTEwA2bk1m+67UYk8vciRYsmBu7uc69RsXmj4zI50d27Ywe8YPfPbB2wA0O6E9DZrkvYlbt3p/p7zadRtGzK92vUZsWLuK9WtWFrfoIsccBZgixZSUGE+HlnXZkZxKUmIcDepUpUGdqlxyThfGfDqLmx77gOzsnNz0NatVyP28YUtyxHwD4yqUSyQpMY7UtMwiladrmwb0P7kNAO9O/PlAFknksJeasoevPhoDQNNW7ahZp17YdLt2bGfE5WeHHdeuS0+uve3BMNNsAyCpbPl8vdODVapSzU0fucmKiFgKMEWKaOPWZB57bRKf/TCPpau2kLkvi6goQ5fW9Xlg6Dn0ObE5Vw7oxt70TG57+qPc6cqWic/9vDc9ctAYPK5cUkKRAsyqlcoy5smriI6O4p/VW3hu9JQDXDqRw1dOTg5v/Pdhdu3YRkxsHJcNvT1i2qjoKMpXtE1S0vamsi/TPvru3KMPAy+7nrLlKuSbJn2vrfWPi4/PNy5Y4B2a6Xv3HtByiBxLFGCKFNH3vyzh+1+W5BmWk+Pwy/yV9L9xFB/+91r6n9KW6y/syagPphbYMccLSYlx/O+FG6hXqwq7U9IYfNfbRa71FDmSjH/9Oeb/NgOAK268k7oNm0ZMW75CJV4a/zUAjuOwc/sWfpw0kW8mvs/cX6Zx+dA/h90qAAAgAElEQVQ7cnuji8iho94AIh5wHId7np8I2E425/RqnTsuZW9G7ucyCXH5pg03bk9qeoHzK5MQx8SR/6JrmwbsSU1n4LBX+XPp+gMtvshha8JbL/L9l/aJwCXXjcjTk7wwxhgqV63B+VcM5YY7/o/srCzGvPIMa1YszZMuwX2tUaCjTySBF7AnlClTnEUQOSYpwBTxyIq129i6cw8ADWpXyR2+cev+dpe1qud/PBc6LnlPWoE1kYHgsmfHJqTszWDgLa/y87wVB1t8kcPOh++M5JuJ7wNw0TXDOGPAJQecV6fup1Clek2cnBymf/tFnnEVK1cFIDVlNxnpkW/udm7f6qavdsDlEDlWKMAUOcQWLtvfc7xVo1oR07V0e5svWbkpYppAcNmrUxNS02xwOXNu/p+lFDnSTXj7Jb7+eBwAvmtu5qzzLzvoPCu5geSWjWvzDK9Tr1Hu5/VrIt+srXd7m9eu2+CgyyJytFOAKeKRBnWqUq2SfUHzqqCXqf+zegtrNu4AoG/38O+4LJMQR/f29iL33azwL5AukxDHpy/b4DJlbwYDhr3KjDnLvFwEkcPChLde5JtPxgM2uDz7/IP/aUbHcdi62d7sJSQm5RnXtFXb3A4+f84J/1Or27ZsZMPaVQCc0KHrQZdH5GinAFPEI0+OsD8ll52dw6Tpf+UZN/5L+2LmC8/omPvS9WBDL+pFuaQEsrKymTDp93zjA8Fl4LH4gGGvKLiUo9KEt17c/1h8yC1FCi6zs7MKTfPTlC9I3mlv/Jq37pBnXHxCIh1POgWAHyd9wt7UlHzTT/poLGB/OrLDib0LnZ/IsU4BpkgR1K1ZmZ/G3sGQ87tTP6h9pTH2NUWfvXwj5/VpB8BbH8/gn9V5f4ruhfe+Z+PWZJIS45k48l+0b3E8ALEx0Vx3YQ8euvEcAN7+ZCbL1uSdNjEhlk9eGkrPjk3Yk5rOgJtf0WNxOSr533k5N7i85NoRnDVocJGmW7pwPk/cdQMzf5jEjm15f6N80/o1+N8dxZiXnwLsr/j0OO2cfHkMuuwG4hMS2bVjGy/83+1sWr8GsD8j+dn7b/Hj158AcO7F15BUrvwBL6PIscI4jlPaZRCPJLa/WRvzEKlbszJ/T3ok93t6xj727E2nXJkEEuJjc4eHe9F6QPsWx/P5qJuoWqksALtT0kiIjyUu1r4tbMrPi7lgxOtk7stbG3Npvy68/egVAKSlZ5KcUnAP80vueJNf5uuXRg6FHz56rLSLcNTavmUTt199HgAmKopy5SsWmP6sQYNz22UuXjCHp++9MXdcbFw8CQmJZGSk5ekZfnyDJtzy4DNUqxG+LfT82TMZ9eR9ub3FE5PKkpGWlvuzlD1OO4chIx7EGHPgCyqF6ta4olbwUUDvwRQpgi079nDrU366tmlAm2Z1qFqpLJXKlSE9cx+r1m/nlwUreO/TX5g1P3IHgT8Wr6XjBY9z+9V9OavnCdSpUZHUtExm/7mKcV/+yphPfyHcDV9U0MUsMSGOxAJedQQQG6PDWo48Oc7+mzInJ4fdu3YUmD49PS33c/3Gzbnu9n+zZMFcVi1bQvLO7aTuSSYmNo7qNetQr1EzOnU/hc7dTyUqOjpinm07d+exUeP56qP3WDjvN3Zt30aZsmWp16gZJ581kM7dTz34BRU5RqgG8yiiGkyRQ0s1mCKHnmowjw5qgykiIiIinlKAKSIiIiKeUoApIiIiIp5SgCkiIiIinlJ308OMz+erA4wEegA5wAxguN/vX1eqBRMREREpItVgHn7eBT4HagK1gS/cYSIiIiJHBNVgHn6q+f3+4IBytM/nGxEpsc/nux643n6rfmhLJiIiIlIECjAPP9t8Pt9lwAfu90uA7ZES+/3+N4A3QO/BFBERkcODHpEffq4BfMAmYCNwgTtMRERE5IigGszDjN/vXwOcW9rlOBK1a16Hs3u1pn3LujSpW42qlcpRPimB3anpLF21iW9mLOLNj35i5+69EfMYeFo7BvfrSoeWdalSMYl9Wdms37yLGXOX8fqH01mwdP0Bla3h8VXp17sNvTo1oXXT2tSoUo6s7Bw2bElm5h/LeMP/E38sXlukZbzlslPp1akJVSuVZUfyXmb/uYpXJkxj2uylEafr0LIuD/7rHLq1bUhsTDSLlm/g+THf88l3f0ScplenJkx+czhfTfuTC0a8fkDLLUeXjPR0/v7L/hzj6mV/s2rZErZv3QTAeZdey8DB14WdbuvmDdx5zcAiz6fHaedw7a0PFatsbz73CDO//6rQdG9/PpPo6MiXvlXLljB54gcs+XMOe5J3kVSuPI2an8Bp/X20bNsp4nQr/1nMxHGv88+iBWRnZ1G7XiPOGnQZXXr2iTjN4vm/8/R9N9G2Sw9u/fd/Cy27yJFEAeZhwufzFXQ2dfx+/6MlVpgj1JXndWPoxb1zv6elZ5KWkUmVikl0a9eIbu0acfPgk7lwxBv8umBlnmnjYmMY/+wQ+vVunTtsT2o6cbHRNK1fg6b1a3Dled249/mJjBz/Y7HK1a1tQ34YfVueYbtT0oiPi6FJveo0qVedy/ufyNNvT+bRVyNfIK8a2I2X7r2Y2Fj7W8q79uylRpVynHtqW849tS2PvTaJx1+flG+6zifUY/Kbw0lMiCMrK5t9Wdl0OqE+458dwogn/bzun55vmrjYGEbefzF7UtO59Sl/sZZXjl4rli7kuX/fWuzpoqKiKV+xcoFp9u3LJC01BYAGTVoeUPkAYuPiSSyTFHG8IfKvEE6b/BnvjXqa7OxsABKTyrJ71w7mzprG3FnTIgbRy5f8xVP33si+zAyioqKJjolh5dJFvPLUfexOvoPT+l2Yb5p9+zIZM+ppEhLLcMW/7jyAJRU5vCnAPHykhhmWBAwBqgAKMAsxe+FqVj83kZ/nLefvlZtJTkkDICkxjgGnteeJEQOoXrkcHz53HW0GPMLulPTcae8acnpucPnah9N59u3JbNiajDGGds3r8Owd59O9Q2Oeum0gM/9YztxFa4pcrpiYaLKyspk0/S8mfD2babP/YUdyKlFRhvbNj+fp2wfRvUNj7rv+LNZs3MGYT2fly6NrmwaMvO9iYmKi+fyH+dz29Ees37KLyhWSePjm/lx3QQ8eGHo2S1Zs5OMpeWsln7h1IIkJcbz/5W8Me3wC6Zn7uOmSk3nmjvN5ZFh/xn/5Kyl7M/JMc/e1Z9C0fg3ufPZ/rN20s8jLKke/pLLlqdeoGfUaN6Neo2Z88OYLJO+M2EwcgCrVavDS+K8LTDP21f/w/ZcfERcfz4knn3HA5evS8zSuu614tZ8Ayxb/yZiXnyYnJ5sO3Xpz2dDbqVy1Bim7k/nfe68y9euJfPb+W9Su24AuPU/LM+2H74xkX2YG3U45k6tuvofY2DimfO7ng7de4H+jX6H7qWfnC3q/mPAum9av4ZLrRlCl+nEHvLwihyu1wTxM+P3+/wb+sJ12EoGrgQlAw1It3BHi/S9/44Wx3/Pbn6tyg0uA1LRMxn/xK9fcPwaAGlXKc3bPE/JMO7hfFwCm//4Ptz7lZ8PWZAAcx+GPxWsZNPw19qSmExUVxYA+7YpVrhVrt9Lu/Me46PY3mfjdPHYk23uJnByHOYvWcNYNI1mw1L7m9M6rTw+bx+PDzyMmJpo/l65n8N1vs37LLgB2JKdyy+MT+HbmIptuxACiovbX0CQmxNKtbUOysrIZ8ZSfvemZ5OQ4jBz/I38sXkv5sol0bdMgz7yaNajB7VedxpxFa3hlwrRiLasc3Zq1aseoD6dw1xMvc9E1wzix9+nExMYddL6ZmRn8MnUyAB1POoWksuUOOs/i8r87kpycbOrUb8yN9zxB5ao1AChbvgJX3XwPJ3Q4EYAP33mZHLeGE2yzgX8WLyAqKporbryT+IREoqKjOWPgJdRr1Iz0tL0sX/JnnnltWLOSSf8bS/3Gzenb31dyCylSghRgHkZ8Pl9ln8/3GLAAW7vcwe/33+33+7eUctGOCr/9uSr3c+0aFfOMO65qBYCINZO7U9L5Z7XdDGXLxBdrvuu37GL5mq0Rx+/LymbCV7MBaFS3GhXLJeYZX792Fbp3aAzAC2O/JysrJ18ez777LQD1alWhh5sWoFK5MkRHR7FtVyp7UtPzTLN8jV2eapXK5hn+8gOXEB0Vxc2Pvk9Ojl5MIPtFRUcfknzn/DyV1JTdAPQ+47xDMo+CbNm4nqUL5wNw1qDBxMTkf7jXz3clANu3bOTvv/Y/JdibshsnJ4ey5SuQWCbvsVSj1vEA7E7elTvMcRxGv/wUOTk5XDXs3kO2TkVKmwLMw4TP53sWmA3sAVr7/f6H/X6/nk16qHv7RrmfV6zblmfcyvX2e/sWx4edtnzZBJrUs+8ZnVOMx+NFlZ6Zlfs5OjrvYdnnxOa5n6e4NZWhfv5jObvdWtvTurXIHb5zz16ys3OoWjGJckkJeaZpcHw1ALbuTMkddvXAk+jRoTGjPpjKvCX68SgpGdO//RywAVnz1h1KfP4L5/2a+7l1xxPDpmnasi0JiWUA+OuP/enLlC2PiYoiZXcyaXtT8kyzZaM9hspX2H9DO23yZyxdOI++511E/cbNETlaKcA8fNwO1AIeADb4fL7d7t8en8+3u5TLdsSKi42hbs3KDL2oF28/dgUAy9Zs4atpf+VJ9+ZHMwDo3bkpz9/jo1a1Crnj2jWvwycvDqVcUgK/LljJhEmzPS9nr05NANi4NZntu/I2x23VuBYAm7fvzhMMBsvJcVi6ajMALRrVzB2elr6PX+avICYmmufvuZDEhFiiogw3XtKbji3rsic1Pbdmt3rlcjw2/DzWbNzBI68U3htXxAtbNq5nyYI5APQ6/eBfoLF4/mzuvu4Crh3Qk6EXnMIDN17K+DeeY9P6yDeG61atAKB8xUoROyNFRUdT8/j6AKxfvb+TYHxCAk1atCYnJ5uxr/6HjPR0crKzmfL5h6xatoSExDI0am6b5CTv3M5H746iSrXjGHTZ9Qe9rCKHM3XyOUz4/X4F+x7a+cvzJMTH5hv+8x/Lueq+0WTuy8oz/LUPp1O7ekVGXNGHoRf1YuhFvXJ7kcfHxbJxazLPvvMtT7zxNdnZ+R9RH4yubRrQ/+Q2ALw78ed842u6we6GLckF5hMYXzMoOAa4/8XP+OaNWxjcrysXndmJfVnZJCbYdnP/N+rL3Efnz955PpUrJDHkgffYm555cAslUkQ/TfkCx3GIjo6me59zDjq/Hdu2EBUVTWKZJNL2prJu9XLWrV7Oj199wqXX38qp55yfb5pdO2wTlkpVCv41tEpVqrEyKH2A7+phPHXvjfz8w9f8MvVbomNi2JdpO84NuvyG3Efn77/xPKkpu7nu9n8Tn5AYmr3IUUUBphyVNm/fTXxcLGXLxOe2mZz629/c/+JnYXtFO47DgyM/Z/HKTTx/94WUS0rI80g5IT6W8mUTSEqMIz1jn2flrFqpLGOevIro6Cj+Wb2F50ZPyZcmUP60QoK+QFBYLqSN6K8LVtJ3yAs8dGM/urapT2xMNHMWreHF977no8m25qjvSS3wndmJj7+dyzczFgLwr4t7c92FPWhYpyrbd6Uy8bs/eHjUl/l6nIscqJzsbGZ89yUAbTp3p2LlKgecV/3GzWjQtAXtuvSgcpXqREVHk5Gezp9zZuF/92W2bFzHe688Q7kKFencI++7KdPT7Ltx4+ILbl8dF5+QJ31A4xatue/p1/hk3OssW/wn2dnZ1G/cnDMHDebE3rbj3oLfZ/Hr9Cl07tGHdl16ADDlcz8/TvqYLRvXU7ZCRTp3P4VBlw8t8DVLIkcKBZhyVGp+zr9zP1erVJZL+3XhriFn8NPYO3jqrfzvm6xSMYnxzwyhd+emfDdrMU+8/jULl28gMT6Wrm0a8NjwAdzg60Xfbi3oO+SF3F7mByMpMY7/vXAD9WpVYXdKGoPvepvUtENTc/j7wtWce9OosOMSE2J58d6L2LVnL3c8+z8AnhgxgFuvPI31m3fi/2YObZrV4aZLT6FDy7r0vfZFz2tx5di0YM4sdm63tYEH27mn77kX5RsWn5BAp+6n0Lx1Bx4ecSXbNm9kwtsj6dT9VIyJ/D7MA9GwWSvuePSlsOMy0tMZ+8ozJCaVZfAN9p24E95+iW8+GU+lKtXo2vt01qxYypTP/az8Zwn3Pv1qgS+DFzkS6LGsHPW27kzhxbE/cN5Nr+A4cN/1Z3FWyGuK3nzkcnp3bsr03/+h/42jmDV/BbtT0tm8fQ+f/7iAPlc/x9ade2h4fDUeveXge7mWSYhj4sh/0bVNA/akpjNw2Kv8GeFXggI1hoHH2gXlCbCnmDWM919/Ng3qVOWBFz9j07bdNKlXneGXn8qmbbvpdsnTXP/vcfS47BmmzV5Kt3aNuOLc8J0gRIor0LmnUpVqtOnY7ZDNp2z5CvS/6GrA9gJfvfzvPOMDnXcyMwo+djIz0vOkL6rP3n+LrZs34Lv6JipWrsrGdauZPPF9yleszP+99B7X3fYQ/35hNM1bd2DZ4gX8NOXLYuUvcjhSgCnHjN8XrubnecsBuOb87rnDmzWokRtwvjj2+7DTbt2Zwvtf/gbAeX3aHlQ5AsFlz45NSNmbwcBbXuXneSsipt/o1pbWql4hYprg8RuLUbt6QpNa3HLZqcyat5y3P54JQP+T2xAVFcWESbNzOxVlZeXw8vv2F4z6n9KmyPmLRLJ71w7mz7b7XI/T+h3y1/U0br7/V7q2bsp7M1exsn2jws7tBb8RLlDbGkhfFGtX/sPkT9+ncYs2nHym/bnMP36ZjuM4nHTKmbmdimJiYjj9vIsBmDtL75+VI58CTDmmBDrCNDq+au6wFg3397oOfX1RsGXuuyyTEuOpXvnAXgQdCC57dWpCapoNLmfOXV7gNAuXbQDsC+KrhryzMiAqytC0vn0x9OLlG4tUFmMMox64BAeHmx79IHd4/Tp23axYl7cjQ2D5G9SuisjBmvH9V2RnZWGMoWff/qValjr17W9Z7N61k93J4d8Ol5Odzca1qwCoXa9B2DT5psnJ4d2RT4IxXD3s3tzH8lvcALd6zTp50teobV+TtnXThmIvg8jhRgGmHFMa1LadCFJS9z8Ky8nZ356wbs3Iv5ccHFQeSEeXMglxfPqyDS5T9mYwYNirzJizrNDpvv9lSe7nvie1CJumW7uGlC9re6V+N2txkcpzg68nXdo04LnR37F4xaZ84xPj40K+2175Dnr5uhy86d9+AUDzNh2pXrP2IZ/fsiX7X01WrUatPONateua+/nPOfl/qhXgn8ULcjv3nNC+a9g0oX746mNW/L2Qs8+/nNr18v8gW2Zm3vPIPvcRvcfNQ0VKhQJMOSoE/zxiJCd3aUqnE+oB9ichA+YtWZv7+boLe4adtkxCXO7PSS5Yuq7Yr/EJBJeBx+IDhr1SpOASYNX67cyca9MOv7wPMTH5D9s73J+YXL1hOzPmFp5vrWoVePim/ixbs4Wn3vomz7jV7kvnA+sqoEvrBrnlETkYSxfOY9O61QD09uDdl45T8E1Pyp5kvvSPBqBy1erUbdQsz/jqNWvTtJVt+vLNJ++TlZUVmgVffWR/arZK9Zo0O6F9oWXauW0LH7/3KjVq1aH/xVfnGRcIcFcsXZhn+LK/bRBcNSQAFjkSKcCUo0KdGpX4ZcI9DDm/O/VrVwkZV5E7ru7LR8/fQFRUFNt3pTJy/A+549ds3MmX0+xvBffr3Zq3H72CBu5j4piYKE5s24Bv3xpOQ/eXb14c+wOh3vi/y0j742XS/ng537jEhFg+eWkoPTs2YU9qOgNufqXQx+Kh7n/xM7KysmnbrA5jn7om90XwlcqX4YV7fZzZo5VN98KnRfp5x+fuvpAK5RIZ9vgEMjLzXkwD6+K8U9ty3qn2otui4XHceU1fO35q3t9VlmNL6p7d7EnelfvnOPYJQGZGep7hoa/yCTZtsu3ck1SuPB27n1Kk+b753CNcdU5Xrjonf+3hzz9+zcjH7mb2zB/YvWtH7vDMjHTmzJrGo7cNyW13edGQW4iKyn/p8109jKioaNau/IdXn76fndtse8yUPcm8N+oZFvxuazYvuubmIrUXHffaf0nbm8qVN91DXFze1x+172pvZOf8PJXfZ/6I4zisX72Cr/w2iO1wYq8irBGRw5vegyBHjbbN6vDyA5cAkJG5j92p6STGx+X57fCV67ZxyR1vsXn7njzTDn14HJ+NuomOLetyab8uXNqvC6lpGcTFxBAbu/9i8vyY73I7+xTVwNPa07tzUwBioqMY98yQAtNfcseb/DJ/ZZ5hvy5YybAnJvDSvRczoE87BvRpx87de6lQNiH3YvnYa5P4eMof4bLMo9/JrTmvTzvGffErU39bmm/83ys3M+r9H7np0lOY8N/r2JuWSZlE+7j8twUree/z8I8Q5djw0C1XsH1L/na+X388jq8/Hpf7vXufc7jutofypUvbm8rsGbYz3UmnnElsbMFvRyiKnOwc5syaypxZUwGIT0gkNi6OvSkp5ORkAxATG8cl1w6na6++YfNo3KI1V958N++Nepo5P09lzs9TKZNUjrS9Kbk1pOddei1dep5WaHnmzprGnFlT6d7nbFq265xvfK26Deh7ro8pn/t5+Yl7iIuPz+3B3rBZK3qUcptUES8owJSjwsatyQy+8216dmpC59b1qVm1PFUqJpGd7bBm4w4WLF3Pl1MX8OHXv4d9Ufr2Xan0vuI/XNa/K4NOa0+bZnWoXKEMWdnZrN20g1/mr+Ttj2cU2Ns7kqigBlWJCXGFvm4oNib8YTl64izmLV7L8Mv70LNjY6pWKsuWHSn8tmAlr0yYxrTZ+YPFUEmJcTx314Vs25nCPc99EjHdHc9+zOoNOxhyfnca1KnKxq3JfPr9PB4e9QVZWXoHphy4X6d9m/u6n14H+e7LgBZtO3L+Ff9i2ZI/2bh2lf1d8NQUEsskUb1WHVq06cQpZw2k2nEFP3rufcZ51GvUjG8mvs/ff85lT/IuylWoROMWrTmtv4+WbTsVWpb0tL2Me/2/lC1fgYuHDI+Y7tLrb6Nq9Vr8+M1Etm5aT4VKVejU/VTOv2IoMRHOASJHElNY2xU5ciS2v1kbU+QQ+uGjx0q7CCJHvW6NK6qb01FAbTBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTCjBFRERExFMKMEVERETEUwowRURERMRTMaVdgKOJz+eLBpYCLf1+f0Zpl0dERESkNKgG00N+vz8byAYSSrssIiIiIqVFNZjeewHw+3y+J4B1gBMY4ff7V5RaqURERERKiAJM773s/u8bMtwBoku4LCIiIiIlTgGmx/x+v5odiIiIyDFNwdAh4vP5jvf5fCeWdjlERERESppqMD3m8/nqAh8A7bCPxcv6fL4LgDP9fv+1pVo4ERERkRKgGkzvvQ58BZQD9rnDppC/TaaIiIjIUUkBpve6AE/5/f4c3B7kfr8/GahQqqUSERERKSEKML23GWgcPMDn87UE1pROcURERERKlgJM7/0H+NLn810NxPh8vkuAD4GnS7dYIiIiIiVDAabH/H7/O8BdwIXAWuAK4EG/3z++VAsmIiIiUkLUi9xjPp+vnN/v/xT4NGR4Xb/fr8fkIiIictRTDab3vvL5fPHBA3w+X0NgWimVR0RERKREKcD03q/Apz6fLwbA5/M1BaYCj5dmoURERERKigJMj/n9/juBdcAHPp/vBOAH4AG/3/9W6ZZMREREpGQowDw0rgeygd+A2/1+/3ulXB4RERGREqNOPh7w+Xw/4b5UPUgcsBe4yefz3QTg9/t7lXTZREREREqaAkxv6PG3iIiIiEsBpgf8fv+Y0i6DiIiIyOFCAeYh4P6Kz+VAbWA9MNbv9797qOcb3/LEQz0LkWNa+/oVS7sIIiJHBHXy8ZjP57sfuAeYANzi/r/LHS4iIiJy1FMNpveuBU72+/2rAwN8Pt9kYDp6F6aIiIgcA1SD6b0kYGvIsO1AYimURURERKTEqQbTe98A430+3z3AGqAetuZycqmWSkRERKSEqAbTezcDe4D5QAowD0gFhpVmoURERERKinGc0PeDixd8Pl8UUBXY5vf7c0pinhUHj9PGFDmENo25rLSLIHLUS4jBlHYZ5ODpEfkh4PP5ygCNgbJAY5/PB4Df7/+5NMslIiIiUhIUYHrM5/NdAbwMZAJpQaMcoG6pFEpERESkBCnA9N4zwPl+v39KaRdEREREpDSok4/3MoGppV0IERERkdKiANN7DwLP+Xy+qqVdEBEREZHSoEfk3lsKPALcGOjcAxjA8fy5ubEAACAASURBVPv90aVWKhEREZESogDTe2OB94APydvJR0REROSYoADTe1WAh/x+v95JKSIiIscktcH03rvA5aVdCBEREZHSohpM73UBbvb5fPcDm4NH+P3+XqVTJBEREZGSowDTe2+6fyIiIiLHJAWYHvH5fB2BDL/fP8b9Xh14AWgF/ALcXorFExERESkxaoPpnReA44K+vwE0cf+3wv7Cj4iIiMhRTwGmd1oAPwH4fL6KwDnAYL/fPwq4BOhfimUTERERKTEKML0Tg/2ZSIATgY1+v38pgN/vXwtULK2CiYiIiJQkBZjeWQhc6H6+GPguMMLn89UGkkujUCIiIiIlTZ18vHM38IXP53sNyAZ6BI27CJhZKqUSERERKWGqwfSI3++fAdQF+gIN/X7/30GjvwJuLZWCiYiIiJQw4zj6RcOjRcXB47QxRQ6hTWMuK+0iiBz1EmIwpV0GOXiqwRQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERTynAFBERERFPKcAUEREREU8pwBQRERERT8WUdgFEjmQj+rfi4Yvb536vOHhckad97pouXNOnKQBrtqbQZsSnYdOd2b423VvUoF2DytSpkkSV8vEkxEazfU8Gf63Zyaf/396dx9lUN3Ac/8zc2ccwthm7sTP2fUmpp+zRQid7SkipaFU8WkSSJMnylJAljiQl2VJZkn1fsu8Mg2H2/fnjXpdphlkczRjf9+vl1b3nnN85v0N35nt/2/nrGHPWHCExKfnWbkYkm0VHR7N50wb27N7N3r172Lt7N2fOnAbguef70++FF29YdtPGDaz7cy27d+3i1MkTXAq7RHRUFH5581KuXHn+81BzOnQ08PLyuuE5kpKS+GXxIn5c+AN/791DREQE/vnzU7t2HTp16UbdevUtv2eR3EoBUySLyhfNy5uPV89S2aZVAun5QIUMHTv0ydoEl/R3vr8SHUdiUjJF8/tQNL8PzWsWp3eLSjwx6jfOX4nJUn1EcoJdO3fwwnN9slR22tQprP7jd+d7b28f3N09uHTxIpsubmDTxg3MmjGdCZO/IiioTKryUVFRvPLyi6z7cw0ANpsNX988XAgNZdnSJSxftpS+/V64acgVkWsUMEWywMUFPu/dCG8PN9bvP0/DioUzXNbbw8a4ZxuRkJTEriNh1ClX8KbH/7jhOJOW7mP9/vMcOxdBTHwiAEX8venxQHnefLw6tcoUZOJzTeg4auUt3ZdIdsubNx9VgoOpUiWYysFVGf3Rh4SGnk+3XKNGjWlyT1Nq16lLqVKl8PXNA0BY2CUW/7yIz8aM5tTJkwx8qT/zf/gJV9eUI8SGvTeUdX+uwdXVlf4vDaBTl674+ubhypUrfDN1Cl/+bxKTJoyndOkg2jzc7rbcu0huooApkgV9W1SicaUA5q45wpGQ8EwFzP8atShbxI+Pf9hJ8QI+6QbMkd/vSHP72bBoRi3Yiae7jVcfqcZDNYtRrIAPpy9GZepeRHKKOnXrsXrdhhTbxn36SYbKduvRM83t/v756dK1Ox7uHgx7byiHDx1k+7at1K5T13nMgQP7WbzoJwC6dO1Or959nfvy5s1L/5cHEhISwo8LFzDmk1E0b9ESdw+PTN6dyN1Fk3xEMql0YV+GGLW4EB7D2zM3ZapsvfKF6NuyEgdOX2H0Dzstqc+mg6HO10Xze1tyTpHsYLPZbtu5a9Ss6XwdEhKSYt/1XetPPdMrzfI9n3kWgPPnzvHnn2utr6BILqOAKZJJnz3biDxe7gyeuZkL4bEZLufh5sr4Po1wwYWBX68nNj7Jkvo0rhTgfH30XIQl5xTJbbZs3ux8XbJkyRT7zpy2TyTy8/MjICAwzfKlSpfGzc3e6bd2zerbVEuR3ENd5CKZ0OOB8txfrSi/7TzDnDVHMlX2jceqU7m4P9N/O8CavSHpF7gJX083ShX2pdO9ZenfpgoA3646lKnAK5LbxcTEEBJyluVLlzB54hcA1K1Xn6rV0p6cl5iYeMNzJScnk5Rk/1J48MB+6ysrkssoYIpkUNH83gzrXIeo2AQGTFmfqbI1Sufn5YerEhIWzdDZW7N0/XrlC7HivVaptickJjHr90O8Pm1jls4rkpuEnj/Pg/c3TXNfs/sfYNjwkam2FyteHLDPJD99+hTFihVPdczhQ4ecAfP8uXMW1lgkd1IXuUgGfdqrIfl8PRj5/Q6Onc94V7TN1YXxfRrj7ubKm99s5HJUXJauH5+QREhYNCFh0cQlXGtpmbryAB99v8M5u1zkbuZqs1GwYCEKFiyEp6enc3uLlq0Y+Orr5PP3T1Wm6X3NnK+/nDQxzfN+9b9r2yMiNRRFJD1qwRTJAOOeMrSqXYIdRy/yxeK9mSo7sH1VagQVYMmWk/yw/niW67D96EUqvTAfsC+TFBSQh+dbV+GZByvQqWlZ+kxYyy9bTmb5/CK5QYECBVi5yj4JJzk5mXMhIcyb+y3fTJ/Kyl9/5a3B/6Wj8WSKMhUqVKRV67Ys+eVnvp8/Dx8fH7p070FAQCBnTp9m2tQpLF+2FDc3dxIS4nF1UduMSHr0KRFJR6G8nnzYvS4JiUm89NVfmXpiTqXi+Xj90eqER8fz6tQN6RfIoORkOBISwevTNjJ09lb8vN353/P3EOivWeQiV7m4uBBYpAj9Xx7IiI9Gk5AQz/Bh7/L3vn2pjn3nvWE0anIPADNnTKdNiwepV6sa7dq0YP68udSsVZvWbdoC9qWLROTmFDBF0vFepzoU9PNi2soDHDh9BV9PtxR/3N2ufYyc22z2baN71sfT3cYnC3cRFhmXqqzNsdizi8u1sm42l0zV76sVfxMTl4iftzsdGwdZdt8iuclDzVtQrFhxkpKSWPD9d6n2+/j6MnHyV3w8Ziz/eag5pYOCKFasOPXqN+CtwUP5evpMQkPtS4KVDgr6l2svcudRF7lIOkoXtj8R5NnmlXi2eaWbHnvq604ATPxlL2/N3Ows+26n2imeWf5PJQvlcZZ9a8YmJi5J3cJyI7HxSVyKjKWohw9lAvNkuJzI3aZwQACnT5/ixPFjae53dXWlRcvWtGjZOtW++Lg4du7YBkCt2nVuaz1FcgO1YIrc4fJ4uVHIzwuAiJiEbK6NSM6UnJzMqZP2Mco+vr6ZLr9s2RIiIiJwc3OjdVs9KlIkPWrBFEnHw8OX33T/oMdrMKhDDQD8u85Msa/GgB9uWnZC38Z0ua8cx89HpHmszdUl3TGfL7UNdnbT3+r6miJ3ooSEBOci6Dfyw4L5zmea16/fIFPnP3/+HGPHjAbg0cc7EBiY9mLsInKNAqZIDmbcU4b29Usxe9Uh1u0/R+gV+0LqLi4QXMKf3i0q0fM/FQBY9/c5Vmw/nZ3VFbllVy5fJjHp2pJbV9eejImO5tKli87tnh6ezpbIrVs2M2H8OB7v+AQNGjQisEgR53HHjh1lwfzvmDF9KgAlS5ai/aOPp7ruqj9+58TxY9zX7AGKFS+OzWYjKiqKP35byWeffsK5kBDKlC3LK6+9cVvuWyS3UcAUycFcXKB13RK0rlsCgIiYeOeEHk/3a89t/mPXGZ4ap8fXyZ3vyY6Pcfr0qVTbp02dwrSpU5zv2z/yGMNGXFs0fcvmTWzZvAkAT09PfHx8iI6OJiYmxnlMpUqV+fTzL/Dy8kp1/uPHjvLxRx8yauQI3Nzc8PHxJTz8CsnJ9h6E6jVq8tnnE/D11ThnkYxQwBTJwZZuPcVLX/1F0yqBVCuVn4B8Xvj7ehAdl8jRcxFsPXyB+euOslwtl3IXC65alQ8+/IhNGzewd/duQkNDuXw5DHd3D0qWLEWV4GAebN6S5i1aYrPZ0jxHoyb30KVbd7Zu2cLZM6cJD4+gUKFCVA6uSus2bWnd5mFcXTVtQSSjXK5+O5M7n3/XmfrHFLmNzk7vlt1VEMn1vNzI3FptkiPp65iIiIiIWEoBU0REREQspYApIiIiIpZSwBQRERERS2kWeQ5iGEZ/4BvTNK8YhjEZqA28ZZrmr9lcNREREZEMUwtmztLHES5bAMWBfsCobK6TiIiISKaoBTNnubrMUGtgqmmamw3DuOmXAMMw+gB9AHBvf3trJyIiIpIBWgczBzEM4xugEFARqIG9hXmVaZp1MlJe62CK3F5aB1Pk9tM6mLmDushzlqeBd4EGpmlGAV5Ar2ytkYiIiEgmqYs8BzFNM9EwjLJAc2A44I2+BGRYzaACtKpTnFpBBSlX1I9Cfl74ebsTHh3P/jOXWb7tNFNW7CcsMi5V2XsqB/BA9aLULluQoIA8FPTzxNfTnbDIOPadCmPRphNMX3mQmPjETNeraZVAFg1pnuHjP5y/nY++35liW6lCvjSqFECtMgWoGVSA6kH5yevtAUCNlxdwPDTypuesVaYAb3esScOKhXG3ubL3ZBjjFu1h4YbjNyxzb3AgPw1uzpItJ+n0ye8Zrr/kXtHR0WzetIE9u3ezd+8e9u7ezZkz9seUPvd8f/q98GK657gQGsrUr79i1R+/cfbMGTw9vShXvjztH3mMxzp0xMUla41X/317ED8uXJDucZu378bNLe1ffTExMcybO4dfVyzj8KFDREZG4OXlRclSpbmn6b107tKNQoULp1l2966dTBg/jm1bt5CQkEC58hXo+UwvWrRsfcO6bFj/F72feYr7mj3A5xMmZexGRe4QCpg5iGEY4wF34D7sATMSmATUz8563Sm6NitHnxaVnO+j4xKIjk+ggJ8njfwCaFQxgH6tKtP5k9/ZeDA0RdkXHw6mVe0SzvcRMfHEJiRSOJ8XhfMV4d7gIvRrVZkOH63k0NnwTNUrLiGRkLDomx7j4+mGn7c7AFsOXUi1f1CHGnS5r1ymrntV3XIFWTSkOd4ebiQkJhGfmETdcoWY/vJ9vDZtA18t35+qjIebK2Oebkh4dDyvTduQpetK7rNr5w5eeK5Plsvv2b2Lfn16ERYWBoCPjw9RUZFs3bKZrVs2s2zpEsZ9MREPD48sX8PT05M8efxuuP9GAfb06VP0ffYZjh876tzm5+dHZGQke/fsZu+e3cz9dhafjZ9I3XopfyTv2LGdZ3t2JzY2FpvNhpubG7t37eT1VwZwcfBFOnXpmup6cXFxfPD+O/j4+PD2kKFZu1mRHEwBM2dpYppmHcMwtgKYpnnRMIys/6S9y2w5FMqQWRH8tf88B05f5nJUPAC+nm60b1CK9zvXoXA+L2a90ox6r/7Ileh4Z9k/dp1l5Y4z/PX3OQ6HhBMRkwBA/jwePNGkDO92qk1QgB8zBzajyaBFZGbo8oYDoVR6Yf5Nj5nz6v20qlOCUxcj+XXHmVT7k5KSOXw2nO1HL7LtyAVcXFx4t1PtDF1/WJc6eHu4MXfNYQZ+vZ6YuCSea1mJEd3rMfTJWsxZfdh5v1e99kg1KhTLy1szNnHyQlTGb1Zyvbx581ElOJgqVYKpHFyV0R99SGjo+XTLhYeH8+LzzxEWFkaZsmUZ/uEoqlarTnxcHPO/m8fHH33Iuj/X8PHIEQwe+m6W69eyVRuGjRiZ6XJD3nqT48eO4u7uzmtvvEX7Rx7Fx9eX+Lg41q5ZzfAP3uNcSAivvzKAxct+xcvLy1n209GjiI2NpW279gwZ+h6enp7MnjmD0aM+ZNzYT2j3yCP4+uZJcb0vJ0/k2NGjvP7mWxQtVizL9yuSUylg5izxjlnjyQCGYRQEkrK3SneOOWuOpLk9MjaBb1cf5mxYNAsGPUhAPm9a1i7OvD+POo+ZuGRfmmUvRcTxv2V/ExufyGfPNqJKCX8aVCjM+v3p/0LNqCL+3jxU0/4LZvYfh0lKI72+9NX6FNubVgnM0Lm9PWw0rFiYhMQkXpu2kahYexf/hCX7eLJpWWqWKUD9CoX5bee1UFuxWF5ebleVrYcvMHnp37dya5LL1Klbj9XrUrZoj/v0kwyVnT51CqGh5/Hy8mL8xP9RokRJANw9POjUpSuRkRGMGzuG+d+ZdO3xFEFBZSyv/42cPn2KzZs2AtCrd98ULY7uHh7c/58H8fbxoU+vnly4EMrmTRu5p+m9gH3YwLatW7DZbLw9xN4iCdD9qZ4s+mkh+/buYfu2bTS5p6nznIcPHWLqlC8JrlqVzl27/2v3KfJv0vi+nOULYD5Q2DCM94A1wEfZW6XcY9N13eLFCvj8a2XT0+W+srjZXElKSmbmHwfTPCat0JkR/r4e2FxduRAeS/h1LbYAh0KuAFDIzzPF9rG9GmJzdWHAlPVZvq7kTjabLctlF/24EICWrds4w+X1Onfpho+PD4mJiSxe9FOWr5MVoeevfWEMrlotzWOqVa/ufB0dda1V/8qVKyQlJeHvn588eVK2UpYqXRqAS5cuOrclJycz7L2hJCUlMfTdYbf0dyqSkylg5gCGYSw2DCPINM1vgCHAaOAS8IRpmnOyt3a5R+NKAc7XR85FZK5s5evKhmRuDGZ6ujUrD8Afu89y7PzNJ+tkVlhkHIlJSRT083SO8byqTIB9nFpoeKxzW48HytOkciCTlu5j+9GLiFjh6JHDzslATe+9L81jfHx9qVO3HgDr/lz7r9UNoPh1gXfP7l1pHrNrp33inaurK5WrBDu3582bF1dXV8LCLhERkfLnyskT9kl0+fMXcG77/rt5bNm8iS7delAluKpl9yCS0yhg5gzTgGWGYQwG9pum+ZlpmmNN00z7J51kmIebK6UK+dK7eUUm92sCwKGzV1iy5WS6Zb3cbZQN9OOV9lX5oEtdANbuDWHbEeuC173BgZQtYg963/yeduvlrYiOS2T9/lDcbK58/FR9vD1suLq40LdlJWqXLUh4dLyzdbZwXi/e61SbE6ERjPhuu+V1kbvXwQMHnK/Ll694w+PKl68AwOFDWf8srF+/jnZtWlK/dnWaNKhDh0fbMerD4Ry7bvLOPxUsWJAHH2oBwJQvJzNn9iyiIu1f9uLj4/l95a8MeftNAHo89TQlSl4LpN7e3tSsVZvExEQ+HP4+0dHRJCYmMmvmN+zZvRsfHx9q1KwF2GfQjx0zmqJFi/F8/5eyfI8idwKNwcwBTNM0DcP4GRgKbDIMYwbXjb00TXNMtlXuDnV2ame8PFJ3Pa37+xy9v1hDXELaQ1sD8nmxf0LHNPf9svkk/Sb/aWk9r7ZeXgiP4edNJyw991XvfLuFnwY3p9O9ZenYJIj4xCS8Pewf/eHztjm7zj/sXo/8eTzpO/FP51hNESucO3/O+Tog8Mbjh6/ui4iIICoyEh9f30xfK+TsWWw2G76+eYiMjODggf0cPLAfc+63vDHobYxOXdIs9+77HxAdHcWfa9fw4fD3+XD4+85Z5ElJSVSoWJF+z7/I4x2fSFV24Kuv8+zTPVj040J++XkRbm5uxMbaewb6vzTA2XU+auQIrly5zPCRo5xjNUVyKwXMnCMe+7JEnoAfmtxzS85djsbT3Yavlxt5vOxdw6t2n2Xot1tuOis6MSnZuaRQXh93ZxBb8NcxRny3Pc01NLMqn4877RvYW0LMNUduGHpv1caDobQZtozBT9SkfoVCuNtc2Xr4Ap//vIfv/zoGwIM1itKxSRAL/jrGsm2nAOjTohLPPFSBMgF+XAiP5ccNx/lg3rZUM85F0nO1NRBIMfv6n7y8vJ2vI6MyFzArBwdTtVp17rv/fgIDi2Cz2YiOjmbtmtWM/eRjTpw4zvBh75E/fwGat2yVqnzefPkY89l4Jowfx4zpU0lOTiY8/NpwmKioKC5dukRiYmKqcZM1a9Xm6+mz+OLzz9ixfSsJCQkEV61Kj569aN2mLQBrV69iyS8/06JlK+5rdj8As2fNYN6cbzlx4jj++fPTvEVL+r80INWMc5E7kQJmDmAYRitgDPAjUMfxFB+5BTUG/OB8XSivJ52aluXVR6qx8v3WjP5hJyPm70iz3IXw2BRLChUr4MPTD1agf5sqtK1XgtenbWT6b9Z0ZT9xTxlngL0d3ePX23L4Ah0+WpnmPm8PG5883YDLkXEMmrEJgPc71+Glh4M5dTGS+euOUr10fp5rVZlaZQvQdthyEpM0+Udylq7deqTa5u3tzUPNW1Cvfn06Gx04feoUY0aP4qEWLVOth7l37x4G9H+ec+dCMDp14YknO1GiREkuhIbyx+8rmfjF54wb+wlbNm/k8wmTcXVNOcKseo0aTPpySpp1i46OZvgH7+Hn58cbbw0GYMzoj5g+9WsCAgNp3eZh/t63l9kzZ7Bn1y6mTJ95w8XgRe4UGoOZMwzGPqFnkMKl9UKvxDJ+8V46fLSSZJJ54/EatKxdPENlT1+MYvi87fT5Yi0ebjbGPNOAaqX8LalXd0f3+MaD59l78rIl58yKNx+vQVCAH+/M2UpIWDTli+alf5sqhIRF02zwYp6fvI4H/vsLq/ecpVHFALo2y9qC73L3ur4lMiYm5obHxcRceyCBr0/mu8dvxN8/P7379APsSxLt27snxf7IyAhe6Nubs2fP0LtvP94eMpQKFSri7e1NiZIl6dr9KT4Z+zkuLi6sWb2KHxbcfF3bf5o0YTynTp7k5Vdeo3DhAI4eOcw306ZSsGAh5sxbwLARI5k19zvq1W/Atm1bWbjge8vuXSS7KGDmAKZp3mua5u7srkdut+XwBf76274cSc8HKmSq7E+bTnD8fAQ2V1e631/+lutSM6gANcvYZ5Z+89uhWz5fVlUt6c8Lravw1/5zTFtpn4jRpm4JXF1dMNceIfSKfRxZQmKyc63QtnVTLzEjcjMBha+twnAuJOSGx13dlydPniyNv7yZGrVqOV+fPJFyvPPPP/3IhQv2yW49ej6TZvmGjRo7Z4//unxZhq+7/+99zPxmGrVq1abjE08C8NvKX0lOTqZtu/YULFgQAHd3d7r1eMqxf0WGzy+SUylgyl3l9CV7A3GZIpkf43Tmkr11pUzgjR9Dl1Hd7re3AkbExPP9uqO3fL6scHGBsc82JJlkBny13rk9KMD+d/PP5ZiuPiLz6n6RjCpf4doXuoMHUz+a9No++5ecsuVu/UtcZhw6ZP+Sl79AgVRrWV7v6rqWp06lvwoFQFJSEu+/OxRw4b/vDXN2y588aQ+4JUum/LJWqlRQiv0idzIFTLmrXA1HEdGZn6hSurCjbEx8OkfenJe7jSeaBAH2yUORsdkzaebZhypSv3xhPlu0h32nUnfRXx0fepWXu31iQzIafymZE1SmLEWL2p9WtXbN6jSPiYqKYstm+xjgxk3usbwOO7Zvc74uXqJEin2urvbgF3bpElFRNx6ldCHU3srpk8Hu+7nfzmbnju08/Uwv5xJM14uJiU3xPjbWPnzAhbSfly5yJ1HAlFzB1SX9H8j3VS1C3bKFAFiz91o3nc01/bLdmpWjSH77DNc1e27cxZcR7RuUwt/X/vScbyyaMJRZRfN7M8SoxaGzVxj9w84U+445FqGvU65giu31KxRy7Ld2MXi5Ozzc/hEAlvyyOM0WwLnfziIqKgqbzUabh9tl6tzJ6Txx6nJYGFO+nAxAYJEiKRZKB6hSparzPN+Zc9M8x4ED+9m2dQsANa/rbr+RkJAQxo/7lFKlStP7uedT7Cte3B5wd+1KOdlwx3b7+rP/DMAidyIFTMkVShT0YfWINvT8TwVnS+NVxQv4MKBdVWa/0gxXVxcuhscy4Ze9zv2NKwWw+L/NebJpmVSPgSwb6Mc7T9bi02caAnD4bDizVx1Odf0JfRsTNqsbYbO6pVvXHo4xnHtPhrHxukdQ3oybzYUCeTydf/L6XHsqj7+vR4p9brb0A/Oop+qTz8eDgV9vIDY+5fJIvzgWoW9XvyTt6tm78CoXz8fA9vZfwos3q/vubnbl8mUuXbro/JOUZP//JyY6OsX265cmAnjq6V4UKlSYmOho+vfr63xiTnxcHOac2Xzx+WcAdOhopPkc8v++PYiaVStRs2qlVPsW/bSQgS/3Z8WypVy4cMG5PSYmhpW/rqBblyed4y5fee3NVDPAH2rRksKFCwPw+Wdj+HLyRMLCLgH2ltVfFv9Mv97PkJCQgLu7O526pP85HzliGBEREQx55z08PVM+jvX+Bx4E4Nfly1mxfBnJyckcPHiAKV9NTrFf5E6mdRAk16heugBje9mDYGx8IuHR8Xh52JzrYAIcPRdOj7GrOHc55UzWJpUDaVLZvshzdFwCkTEJ+Hi64eN57SOy89hFuo75g5j4rC9CXiYwD00cj52ckYnWy0YVA1g0pHma+1aNaJvi/cMfLE/RQvtPbeqWoF39Uny76hCrdp9NtX//6StMWrKP51pVZsbAZkTFJjj/HjYePM+sVdk3KUmy35MdH+P06VOptk+bOoVpU68t09P+kccYNmKk872fnx+fT5hEvz69OHzoIJ2NDvj6+hIbG0dCgn3YSeMmTXl90NuZrlNSYhIrVyxn5YrlAHh75Y2JvgAACnRJREFU++Dp6UF4eDiJifbPq4eHB6+9MYhWrdukKu/j48On4ybw4gt9uXTxIuPHjWX8uLH4+voSFRXlbCH18vbmg+EjKV066Kb1+W3lClauWE679o/SsFHjVPvLlitHl27dmT1zBq8OeBEvLy/n7PrqNWryyGOPZ/rvQCSnUcCUXOHMpWie+mwVTYMDqVeuEIH+3hT08yQxKZkToRHsOhbG4s0nmPfn0VQBcduRC/SduJamVQKpVaYAAfm8KZDHk9iERA6fDWf70Yv8uPE4C9cfJymdrrj0dGtWHldXF2LjE5mz5sgtnSsrfD3d+KhHfS6ExzB41pYbHjdoxiaOn4+g54MVCArIw9lL0fy44RgfzNtOQqLGYErWBFetxvcLf+brKV+y6o/fCTl7Bm8fb8qXr0H7Rx7j0cc7pGpdzIj6DRvy4ssD2b5tG0cOHyLschgRERH4+uahZKlSNGjYiI7Gk5QoceMVEKrXqMEPPy3GnPMtq1f9wdEjR4iMjLAvVVSiJA0aNaZzl24pHhOZlqjISEYO/wB/f39efePNGx73xqDBFCtWgu/mzeHkiZMUKlSYhxwLrbu7u9+wnMidwiW9sSty5/DvOlP/mCK30dnp6XeNisit8XLTLKfcQGMwRURERMRSCpgiIiIiYikFTBERERGxlAKmiIiIiFhKAVNERERELKWAKSIiIiKWUsAUEREREUspYIqIiIiIpRQwRURERMRSCpgiIiIiYikFTBERERGxlAKmiIiIiFhKAVNERERELKWAKSIiIiKWUsAUEREREUspYIqIiIiIpRQwRURERMRSCpgiIiIiYikFTBERERGxlAKmiIiIiFhKAVNERERELKWAKSIiIiKWUsAUEREREUspYIqIiIiIpRQwRURERMRSCpgiIiIiYikFTBERERGxlAKmiIiIiFhKAVNERERELKWAKSIiIiKWUsAUEREREUspYIqIiIiIpRQwRURERMRSCpgiIiIiYikFTBERERGxlAKmiIiIiFhKAVNERERELKWAKSIiIiKWUsAUEREREUspYIqIiIiIpRQwRURERMRSCpgiIiIiYikFTBERERGxlAKmiIiIiFhKAVNERERELKWAKSIiIiKWUsAUEREREUspYIqIiIiIpRQwRURERMRSCpgiIiIiYikFTBERERGxlAKmiIiIiFhKAVNERERELKWAKSIiIiKWUsAUEREREUspYIqIiIiIpRQwRURERMRSCpgiIiIiYikFTBERERGxlAKmiIiIiFhKAVNERERELKWAKSIiIiKWUsAUEREREUspYIqIiIiIpRQwRURERMRSCpgiIiIiYikFTBERERGxlAKmiIiIiFhKAVNERERELKWAKSIiIiKWUsAUEREREUspYIqIiIiIpRQwRURERMRSCpgiIiIiYikFTBERERGxlAKmiIiIiFhKAVNERERELKWAKSIiIiKWUsAUEREREUspYIqIiIiIpRQwRURERMRSCpgiIiIiYikFTBERERGxlAKmiIiIiFhKAVNERERELKWAKSIiIiKWUsAUEREREUu5JCcnZ3cdRO5ahmH0MU3zf9ldD5HcSp8xkeyhFkyR7NUnuysgksvpMyaSDRQwRURERMRSCpgiIiIiYikFTJHspbFhIreXPmMi2UCTfERERETEUmrBFBERERFLuWV3BURyG8MwHgO+B6qYprnPMIwgYJFpmtUMw6gH9DBN86VsraTIHcowjERgJ+AOJADTgbGmaSZZcO53gQjTNEff6rlE7nYKmCLW6wysAToB716/wzTNTcCmbKiTSG4RbZpmLQDDMAKA2UA+4J1srZWIpKCAKWIhwzDyAPcADwA/8o+AaRjG/cBrQHvgMFDLNM0wx76DjrJJwCSglKPYANM01/4L1Re5o5imec4wjD7ARkfroyswErgf8AS+ME1zsuNzuRDIj73lc4hpmgsBDMMYDPQATgDngc3/9n2I5EYagylirUeBJaZp7gcuGoZRJ62DHN15C4HHAAzDaAgcNU0zBPgM+NQ0zfpAB+Crf6XmIncg0zQPY/9dFgD0Ai47Pjv1gd6GYZQBYoDHTNOsg/3L3yeGYbgYhlEXe09DbeBxRxkRsYACpoi1OgNzHK/nON7fyFzgScfrTo73AA8B4w3D2Ia9FTSvYRh+t6GuIrmFi+O/LYAejs/OeqAgUMGxf4RhGDuAFUBxIBC4F1hgmmaUaZpXsH/eRMQC6iIXsYhhGAWB/wDVDMNIBmxAMjDhBkXWAeUNwyiMveXzA8d2V6CxaZrRt7nKInc8wzDKAonAOexB8kXTNJf+45ieQGGgrmma8YZhHAW8HLu1Vp/IbaAWTBHrdAS+MU2ztGmaQaZplgSOACXSOtg0zWRgATAG2Gua5gXHrmVA/6vHGYZR6/ZWW+TO5PhyNgkY7/g8LQX6GYbh7thf0TAMX+yTgM45wuUDQGnHKVYBjxmG4e3oJWj379+FSO6kFkwR63TGPsHgevOBt29SZi6wEeh53baXgC8c3Xlu2H8JPmddNUXuaN6OLvCryxTNwP4lDezjlYOALYZhuGCftPMoMAv4yTCMTcA2YB+AaZpbDMOY69h2DFj9L96HSK6mJ/mIiIiIiKXURS4iIiIillLAFBERERFLKWCKiIiIiKUUMEVERETEUgqYIiIiImIpBUwRERERsZTWwRSRu4JhGE2BUUBV7E9+2QsMcLx/1jTNphk8TxD2BfTdTdNMuD21FRG5sylgikiuZxhGXmAR0A8wAQ/sz6GOzc56iYjkVlpoXURyPcMw6gErTNP0/8f2KsBW7E+FiQYSTNP0NwyjLfZnw5cDLgNTTNN811HmOFASiHScpjnQEihvmmY3xzFBXNfK6XgW9lDsz8MOBYaYpjnrtt2wiEg20xhMEbkb7AcSDcOYbhhGa8Mw8gOYprkX+2M415mmmee6ABoJ9AD8gbbYn2/9qGPffY7/+jvKrLvZhR3Pwh4HtDZN0w9ogv3RhCIiuZa6yEUk1zNN84pjDOabwJdAEcMwFgO9b3D879e93WEYxrdAM+CHLFYhCahmGMZx0zTPAGeyeB4RkTuCAqaI3BUcrZU9AQzDqAzMBMYCS/95rGEYDYGRQDXs4zU9gXlZvG6kYRhPAq8BUwzDWAu8aprmvqycT0TkTqAuchG56zjC3TTsATKtgeizgR+BkqZp5gMmAS6OfWkdHwn4XPe+yD+ut9Q0zeZAUWAf9lZUEZFcSy2YIpLrOVos2wJzTdM8aRhGSaAz8BcQApQwDMPDNM04RxE/4KJpmjGGYTQAugDLHPvOY+/yLot9bCfYx1S+aRhGKeyTgt667tqBQEPgV+wTiSKwL5MkIpJrqQVTRO4G4dhD3nrDMCKxB8tdwKvASmA3cNYwjFDH8c8D7xuGEY599rd59USmaUYBw4G1hmGEGYbRyDTN5cBcYAewGfuSSFe5Oq5zGriIfSzn87frRkVEcgItUyQiIiIillILpoiIiIhYSgFTRERERCylgCkiIiIillLAFBERERFLKWCKiIiIiKUUMEVERETEUgqYIiIiImIpBUwRERERsZQCpoiIiIhY6v+ERw05UHiWIQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "count = np.array(data.groupby(['Smoker', 'Status']).count())\n",
    "count = np.reshape(count, (2, 2))\n",
    "annots = np.array([f\"{v}\\n{v/len(data):.2%}\" for v in count.flatten()]).reshape(2,2)\n",
    "\n",
    "plt.figure(figsize=(10,8))\n",
    "sns.heatmap(count, annot=annots, fmt=\"\", cmap='Blues', cbar=False, square=True,\n",
    "           xticklabels=['Alive', 'Dead'], yticklabels=['No', 'Yes'], annot_kws={\"fontsize\": 25})\n",
    "plt.title(\"Number and percentage of alive/dead women after 20 years, according to smoking habits\")\n",
    "plt.xlabel(\"Status\")\n",
    "plt.ylabel(\"Smoker\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is possible to see that the fraction of smokers and non smokers is quite balanced (in total, 582 smokers and 732 non smokers). As expected, there are less dead than alive people (369 versus 945).\n",
    "\n",
    "We can then compute the mortality rate for the two groups. For a population proportion $p$, confidence intervals are computed as $\\hat{p} \\pm z \\cdot \\sqrt{\\frac{\\hat{p}(1-\\hat{p})}{n}}$, where $\\hat{p}$ is the sample proportion, $n$ is the sample size and $z$ is the value derived from the standard normal distribution. For 95% confidence intervals, $z=1.96$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mortality rate for smokers:\t23.88% ± 3.46%\n",
      "Mortality rate for non smokers:\t31.42% ± 3.36%\n"
     ]
    }
   ],
   "source": [
    "z = 1.96\n",
    "\n",
    "num_smokers = sum(data['Smoker'] == \"Yes\")\n",
    "num_dead_smokers = sum(np.logical_and(data['Smoker'] == \"Yes\", data['Status'] == \"Dead\"))\n",
    "rate_smokers = num_dead_smokers / num_smokers\n",
    "ci_smokers = z * (rate_smokers * (1 - rate_smokers) / num_smokers) ** 0.5\n",
    "print(f\"Mortality rate for smokers:\\t{rate_smokers:.2%} \" + u\"\\u00B1\" + f\" {ci_smokers:.2%}\")\n",
    "\n",
    "num_non_smokers = len(data) - num_smokers\n",
    "num_dead_non_smokers = sum(np.logical_and(data['Smoker'] == \"No\", data['Status'] == \"Dead\"))\n",
    "rate_non_smokers = num_dead_non_smokers / num_non_smokers\n",
    "ci_non_smokers = z * (rate_non_smokers * (1 - rate_non_smokers) / num_non_smokers) ** 0.5\n",
    "print(f\"Mortality rate for non smokers:\\t{rate_non_smokers:.2%} \" + u\"\\u00B1\" + f\" {ci_non_smokers:.2%}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Surprisingly, the mortality rate is sensibly higher for women categorized as non smokers. However, we are not taking into account an important information: the age of those people at the time of the poll. This result can be expected, for example, if the average age of polled non smokers was higher than the one of smokers, leading them to earlier death independently from their smoking habits.\n",
    "\n",
    "---\n",
    "\n",
    "Let's now include the age in the analysis. The following age classes are considered: 18-34 years, 35-54 years, 55-64 years, over 65 years."
   ]
  },
  {
   "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>Smoker</th>\n",
       "      <th>Status</th>\n",
       "      <th>Age</th>\n",
       "      <th>Age group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Yes</td>\n",
       "      <td>Alive</td>\n",
       "      <td>21.0</td>\n",
       "      <td>18-34 years</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Yes</td>\n",
       "      <td>Alive</td>\n",
       "      <td>19.3</td>\n",
       "      <td>18-34 years</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>No</td>\n",
       "      <td>Dead</td>\n",
       "      <td>57.5</td>\n",
       "      <td>55-64 years</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>No</td>\n",
       "      <td>Alive</td>\n",
       "      <td>47.1</td>\n",
       "      <td>35-54 years</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Yes</td>\n",
       "      <td>Alive</td>\n",
       "      <td>81.4</td>\n",
       "      <td>Over 65 years</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Smoker Status   Age      Age group\n",
       "0    Yes  Alive  21.0    18-34 years\n",
       "1    Yes  Alive  19.3    18-34 years\n",
       "2     No   Dead  57.5    55-64 years\n",
       "3     No  Alive  47.1    35-54 years\n",
       "4    Yes  Alive  81.4  Over 65 years"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def bin_age(age):\n",
    "    if age < 18:\n",
    "        return None\n",
    "    if age < 35:\n",
    "        return \"18-34 years\"\n",
    "    elif age < 55:\n",
    "        return \"35-54 years\"\n",
    "    elif age < 65:\n",
    "        return \"55-64 years\"\n",
    "    else:\n",
    "        return \"Over 65 years\"\n",
    "\n",
    "data['Age group'] = data['Age'].apply(bin_age)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Again, let's check that no missing data are present, to ensure that no women under 18 was polled."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of rows with missing values: 0\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of rows with missing values:\", data.isnull().any(axis=1).sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's visualize the number on women alive and dead after twenty years, according to their smoking habits and age. Different colors correspond to different couples of smoking habits and status."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3EAAAJhCAYAAADv1TljAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzs3XmYFsWdwPFvAQKKQQIjAVSiEeKdGI2CYECUU8DbBiOrWVkxrhGPJJ4x6mKCxiuaaBbEeJvQbFTAEEVEza5k8YpGPKJEERUPEIUgMCxD7x/djC/vXO8wI0PL9/M880D3W11dXV3V7/t7u7rekCQJkiRJkqR8aNbUBZAkSZIklc4gTpIkSZJyxCBOkiRJknLEIE6SJEmScsQgTpIkSZJyxCBOkiRJknLEIE65EEJIQgijmrochUIILUIIvw0hfJSV75CNzOeQbPsds+Wds+WDG7XApZfn9hDCrE20rwUhhJ9sin3lXQjhzBDCOyGEdSGEy5q6PMqfEMJlIYT5NS3nXWO9T+TtulTKe8Ymvq4/HkKY1Aj51HkeGmtfm1Iey6zNk0GcapVd+JMQwvXVvLbZBVab2LHAd4HhQGdgTiPl+3aW39xGym8DIYTLQwj3fR556/MRQugC/BIYD+wAXBNC+EkIYcHnvN8jQwgzQgjvhxBWhhBeCiGcFUIIRek6hxDiEMLy7O/3IYSOn2fZ1CiuAXpuih2FECaFEB7f3PKqwQFAlfe8nDsLOL6pC/E5OAY4d/1CCGFWCOH2piuOtOkYxKkUq4AzQghfb+qCNLYQQssGbN4deDdJkjlJkryfJMmaxihTkiQVWX7/1xj5VeNo4P7PKW99Pr5Ger2eliTJe0mSrGjMzGvpB4cAfyH9oLQ3cC1wJXBewbbNgAeBXYABwEDg68ADxcHe5iCE0CyE0Lypy/F5CSFsVWq9J0myIkmSJZ93mfImSZLFSZJ8WtPrDXzfaBJJkixLkuTjpi5HY0uSZGmSJMubuhxSUzCIUynmAM8CV9eWqLo7c8XfimXDI8aFEH4TQlgWQvgwhPCDEEKrEMKvQggfhxDeDSH8oJpddAgh/CGE8GkIYVEI4dzCF0MI24YQbsi2XxlC+GsI4ZiC19cPOTkxu7vwKfDzGo4lhBB+FEJ4I4SwJoTwjxDC2QWvPw6MA76W5bmglnr5WQjhlaxMb4cQ/jOEsF0t6TcYGhNCeDKEMLGadC+HEK4sWB4ZQng+hLA6q+frQghtirbZFdiD9EM3IYQvhxAmZ3X6QQjhCqDKB8CQDuV7Ncv79RDCxSGEFgWvfzeEMDc7p0tCCH8sDvpDCN8MIczJ8ngthBDVVAcF27wdQvi3guU7srrpVrDurRDCv2f/r/W8ZWk2qg3Wo31FIYTpWZo3Qgj/UscxfjmEcHcIYWEIYVUI4e8hhB+GkH4QD+nQyf/Oki/M9vE90vb31Ww5ydKtH+Z7WQjhzayuXwohnFa0zySEMDaEcG8IYRlwT3VlS5LknCRJxmVfVLyRJMlvgUlA4bnrD+wHjEqSZG6SJP8L/AtwENC3hmPuF0KoCCHsVLT+5BDCP0MIX8qWvxLS0QCLs/VPhhD6FKQPIYRbsvO8Kqvvn4cQWhWkuSyEMD+EMCKE8CqwBtgjhLBXCOHhEMInWft/pa5zVc1xlNLuO4YQbsv61+rs/J5S8PquIYQpIYSlWZv5WwhhWMHrh4cQng0hlGdt9eZQ0K+z+pkV0j66ACgH2mTteX0b/ziE8BugVVHZqh1eGdI7sK9m9fJYSK8bhdudkNX56pD26WGhluF8WdscDfQtaK/fy17rHNI7t59k5/DxEMK3a6nzGvPKtA0h3JW1l7dDCOcVbV9K/9hgGF+2fEVW9x8BT9ZQtrbZuX4/O19vhxCuK3j98RDCrVleH2bH/LOQfrHw06yNLA4h/Kwo3y+FECZkr60OITwTQhhYUx1l25yfnfdDsuUNhlMWtJsxIb2GLg8hTA0hbF+Uz9khHca9Musv/xIKhv/XUYZLsrpYmu2vsN3uF0L4U1YPK0IIT4cQBleTzdYhvfO6PKR97KqQfnFUWKeT1h8TcBhwckHbWH/8F4X0+lCe1ePDIYStayl7rX07lHi9DyF8NYTwUNa2F4YQziyh3kKo47qWpavz3IQQ9g8hzMzqeHEI4b4QwlfrKoNyIkkS//yr8Q+4HZhFOuRmHdCv4LWE9INbtcvZulnA7QXLC4BPSIc/dAN+kuU7o2Ddhdm6PYvyXgqcSfot/1nAWuCY7PUAPAY8DhxMeudiDOkHtsOyNDtn+bwDjMrS7FLDcZ9BegdyDOkdt+8Dq4HR2evtSYcivQl0AravpQ5/Anwn2/9hwKvAHQWvH5KVa8eich6cLZ+W1Vnrgm2+naXZM1v+HvAx6YfnrwF9gL8BdxWV5cfAIwXL9wPzgUOBvYC7geXArII0lwFvkd7B2wU4HFgIjCtI86/AMGBX4FvANOB1oGX2+tbAu9l5/ibpB/yngZXAT2qpuzuB3xUsLwQ+BE7LlnfN6mH3Us7bxrZB6te+3iANcrqR3rVaC3Sv5Rg7AeeTBkK7kLbNFcC/Zq9vS3onLMnqthPQJsv77Wy5E7BtQZ/9G+kdsV2AEdnxFtZBAnxE2p92Bb5ej2vCncDDBcuXA29Uk+7tOs7tq8ClRev+G7iloM28DPyBtL13Ay4mDVL2yNI0A64AemT1fwTwHnB5UftdCTxBeh37OvClrI7uBfbMzucQYFg9r4+ltPtXgOdIg92vZedlZMG5/4D0Onlwls+RwOHZ69/I2s/1pF++DCHtA3cVlOF20j57P7AvsA/QItvmwyy/3UmvV8uB+UV1U7z8KfAQsD9pX/0r8ERBmv1J+8YVwG7AUaTXkMprVjX1tC3pFwVz+Ky9bk3ar+YCz2fHvw8wmfRaVlafvAra9QfAqVldjs3W9Suqr7r6xwIK2m62vDyrn69T8N5UVLYbgRdI22NXoBdwasHrjwPLgKuyfE7JyjcD+EW27uRs3ZCC7aZkZRiUtYMbSK896697O6+vf9I+8StgEfCNouOeVbS8DPgd6V32XqTX+cL3pmNI299ZpNfT72X5Vr5f1VAPj2d1ej1p2xucLRf2y0OyY90zO+4rsmP6ejX1/h+kbe1fSNvnuUX7mpT9fzvgz6RtaH3baJkdx3LSRx+6kvaTs8nazUb27fV1XuP1nrR9P0f6Xtcj2+8jWVkm1bLvUq5rdZ6brG5XkF6jdyftX1OA1yj4POFffv+avAD+bd5/hRf+7GL/V6BZtryxQdwDBcvNsgva9KJ1HwM/KMq7OCC5F/if7P+HkH5Y364ozW/X76/gontJCcf9NvCLonXXU/BhlaIPQPWo06NJP4g2Kyh7bUFcO9LAZERBHjcCTxfV6/eL9tMny+fLBevmAP+e/b9b9vqAgtdbkgZb68/5NqQfgAcX5X0S8Ektx9g+y7t3tvxv2ZtJYVn2ztLU9kH/e8AH2f+7Z2W5BIizdacCi+p53urdBuvZvgo/YLTIjvu0eraRG9gw2N6gjWTrfgIsKNpuF9IP2LsXrf8p8HxRf7p1I9ruIcD/AcML1k0E5lST9mngplryOpf0Q+P6frBbVq4DCs79O0CLou1mA7+sJd9zgNeL+uk6oGtRumXA9+pbB3XUT3G7H521m2o/8JLeTX0faFPD63cBTxWtOzI7nq9my7eTfkDetiBNm2y/pxZt+wx1B3FrKfhSChiZ7a91tnwP8N9F+X6fWoK4LM0k4PGidYdR8GVUtq4V6QfWn9Ynr4J2fWPRuleB8fXsHwuoGsQ9WsL5n0rB+101rz9euJ9s3UvAi0XrXgCuyf6//jp9eFGa54DfZv/fOUtzGOmH9FfXt4+C9LdTNYhbDLQqWHcB8F7B8pNUfd+9ktKCuL8VrftP4C911N8LwMVF9V7c1n4OvFO0r0kFyxt85sjWnUMauGxV1zmspWzFfXt9ndd4vSf94iZhw8B0e9L38xqDuBr2X3xdq/PcZOf490VpWpG+jx61sXXh3+bz53BK1ccFpN/mfK+B+byw/j9JkqwjfSP5W9G6D4HiiRH+UrT8JOk3TZA+iN4SeDcbNrAihLCC9K5G96LtnqqtcCGEtsCOpN/oFXoC2DmEsE1t21eT3zEhhD+HdAjoCtIPQS1JvyWsU5IknwDTSQMnQjqMcSRwR7a8PfBV4LqiY/9TlkW3LF1n4EDSDxrwWd1VTsiSpM/1PV2w+71IvzH/Q1HeE4Dt1g+9CSHsG0K4P6RDlP5JereArFzr9/VKUvBMRpIk80g/SNfmUaBjCGFv0ruF/0N6l6BfCCFk62ZnZajPeatvG6xP+3q+IJ+1pHcGvlLTAYZ0KNUFIR0KuyTL9/t8Vnf18W3Sb3+fKSrnRdWUs9Z+UE05ewIPAJclSTK9xM2SWl67nbR+B2XLpwIvJEmyvv0dQNpHPik6lu9QcCwhhFOzYU8fZK+Pp2rdfZAkycKiddcAk7LhWJeFEPYr8ZgqldDu9wdeTpLknRqy2J80AK7p+au9qL49Bz7rv5D2rcLnJHcl/bBWPNnS/9R8NJUWJUmyuGD53Wx/6/vCnsD/Fm1TfG0u1V7AR0mSvLx+RZIk5aR35/bayDyfL1p+l8/6X336R7FS+svNwHEhhHkhHXo9pHDoX+aFouX3Kbj2FKwrrG+o2g7+TNU6uo30bkvvJEneKqG8r2T1vV5hXa3f98ae69rOAyGE7UM6PPXVkA4rXUF6PMV9t7r3/R2y632pYmAr4K2QDuv8l5AN2a5JCX17vdqu93sCS5Ikea0gzWLg73UVuITrWinn5gDg6KK2/hHQmrrbu3KgRd1JpFSSJG+FdJbKK0IIcXVJqPo81VbVpCuesCOpYV1dXzIU7qsZaUBwQDXpiiccqfGB9WrKUNP+ShJC6EH6zeh40qGMH5MO6bqDNCgo1R2kE0V8hTQQawf8PnttfT2dRTrkr9j6D5BHAc8kSfLu+uKVsN/1eR9P+k1msaVZcDST9APiKaQfQCD9hnn9MQZq/0BfrSRJ3g4h/IP0G+ZepAHbs6TXrm8A/Ug/gG2wWdFydcdZ3zZYn/ZVvFxXW/4h6fDNc0m/Xf8n6beuQ2vZpibr99OL9NvW4nIUKrUfENLnSqaT3tEofo70PdJvnIt9hc/aQhVJkiwNIfwXcGpIn9U5ifRO0HrNSIciHl3N5iuzch0P3ET6BdMTpHdUjwd+VpS+yrEmSTIuhHAP6VCvQ4GLQgi/SJKkpKnlS2z3UHe739jXC9cXH1+oJk2pqmu/sGEb3ph8a1JdXht1vcjU1v/q0z+K1dlfkiR5OITQlfSLiUNIh6e/GEI4LEmSiixZY77/FZf5j6RtcTA1POdapLq6Kr5efh7nAdIvcbqSTpL0Jundqd9T9/tivd+HkyR5N4SwO+n7xaGkozmuCiH0SJLk7So7KL1vQ+3HuVHtuB7XtbrybkZ6N//Kal77qL7l0ubHO3Gqr/Gk7eb8al77EOiyfiF7CHfPatJtrOKpsA8i/ZAH6TChdqRDfuYX/RV/A1+rJJ3p6h2qTsrQB3gzSZLiN//aHEz6TdxPknTSh9dI7xbV18OkF93vkj4X8Mckm1UuSZIPSIcR7lbNsc9PkmR1lscxbDgr5UvZv73WrwjprGsHFKVZDXythrwrSJ/R2J50GMxjSZK8AnyZDd9sXwL2DCG0K9jXXqTPMNRlNmkQdwjpkKZ1pN9Cn0kaKMzO6qExz1uxRmtf1egDPJQkya1Jkvw1SZL5lPYt6RqgeJbFZ7N/u1ZTzn9sTOFCCENJn9m5rJoADtJvxncJIRTeHdsD2Im67/xMIH1O5fukQwALP3g+Q/oM2fJqjmVRlqYP8NckSa5LkuTZJEleJx3mVJIknazl5iRJjiMdUnd6qdtSWrt/Ftgr1DwJxLNA71A0AVGBl6janvuSfnh7uWrySvNJ20fvovW9qklbXy+TXnsLlfIzBdW115eAshBC5ftE9r5xIJ9dn0rNqxSN3j+KJelsib9LkuQ00i9i+tKw98H19dCnaP13qFpH95A+Z/bbEMLJDdjneht7rkvRB7g5SZJpSZK8SPpl0NeqSVfd+/6ipOYZKattG0mSlCdJ8lCSJOeR3q3chvSLzeqU0rdL8RKwfdG1sYz0GcDalHJdK+XcPEP6Zec/qmnvX7iZSrdEBnGqlyRJ/kn6LdYPq3l5FvD9EMJB2fC326nf3aa6DAvpLILdQzrD0wg++y2f2dn+7wshHB1C+FpIZ2U6M4Rw6kbsazxwZjakoXtIZy87nRpms6zF30kv4qOzMp0E/Ht9C5MN07iXdDKN4aSTSxS6GBgb0t8O2zuEsFsI4agQwgRIZ0AkDYLuL8hzPunD2jeFdLbAPUmfNflSQZoVpMf886zudwvprH4jQwhXZcneIn3G78yQzrR3GOkzXYXfEt5Leofp7pDOUtmT9HmyVSUc/mzSCR1akd6pWr/uZNLgbEFB2sY6b9WVoTHbV6G/A4dk5+DrIZ0htEcJ270JdMr6W1kIYZvsnP4WuCUbMtQtq+9TQgjVffFSq+wb4ftJhx7eE0LolP0VzmA3i/S83B1CODC7+3wX6VCfJ2rLP0mS/8mO/xrS5xwLh9fekx3jH0MIA0M6G1yPEMKFIYT1H77+DuwT0tkUdw0hnEX6ZUVdx7VtCOGmEMKhIYRdQgjfIr17UVtgVKyUdv+7LN20EEL/bF+HhRBGZK/fTPo+PDWE0Dt7fVgIYUj2+tXAfiGdaXb3kM7e9yvgntq+PMiGZ/4n6aiJI7J++wvS4fANdR1p4PkfWXs9gs/eD2q7M/AmsHt2/SjLgrXZpMMU782Of2/Sa1tr4Df1zKtOjd0/ioV0psljsvruDpxI+ozURn/RkwWXU4CbQwiDsnZwA+kzxVVmjE6S5PekX/ZNaIRr07XAyOw61y17/zpp/a4amPffgRNDCPuEEPYl7SvVBeb7hnS489dDCN8lHXFS22/4vQnsn/XJspD+5Mbo7D3hmyGdmfFE0ve5mvp7KX27FI+SDp9df23cl/S6traO7Uq5rpVybn5OGpCu3/8u2fvMDSGE6gJm5U2yGTyY59/m+0fRw9DZumakF6aEDSc26UQ65Go56Z2h06l+YpOfFOU3n/Rb/sJ1rwJXFCwnpLNJPUA6DOY94MdF22xNOmzgTdJv494nfX7q0Oz1nanj4fuCvALp8Mc3SYe6vAGcXZTmMkqY2IR08oIPSIfjzABOyMqxc/b6IdQysUlBPt/ks1kFW1azn6NIx8SvzM7B82STA5DevXupmm06kD4v8Cnpc2HjSYduFp/z0Vl+q0mHhM4FTi94/TjSmbtWk05+05f0jep7BWm+lZWvHPgH6XN9VdpDNWXsSDoZwdSCdftkdTGpKG0p521j2+BGta/q8i56fbvsHCzPzu1NWZtZUJBmgzaSrduKNDhemr12Wba+OekQpfXT6S8hDaaOL+pPo2oqU0G6x7O0xX8LitJ1Jv2g+c/sOCYDHUu8xpyV5XlQDe3zN6TP06zJ/r0f+FZBHUzI6mB5Vh8/AJLa+ilpkHBvdi5Xk44imAzsVJDm9uLjrKZ8pbT7TqSByZIs3atFr389O6ZlpH33BQomsSCdDfZZ0n6zOKuPNkXlnFVN2bbO6mZZ9jeRtH/XNbFJcV0dTMH1Klt3AmkfLift01GWZv9a6qo96fVvWZb2ewVt5/ekk7OsIm2r366j3mvKq0q7pup7UCn9YwFVJzap9TqVpbsEmEcauC3L8j244PXHqXrNqm4ijoeAuwuW22bncnFW588AAwte35miaw/pjIar+Gwiqw3aSXXthvQZ36Ro3Tmk/W4V6YiQMdm+OtRx3Sg+zg0mYiK9hs/J8l1A+uVmdZ8Xfkb6rN9y0n5+NdC8pn2R3s37c3YOEtJr5zHZvj4m7WPzKJiNdGP6dnV1nq3f4HqfpZuZ5fMO6fWuSv0U5VHnda3Uc5PV89Ts2Fdl5ZsItK+rPfu3+f+F7CRL+gILIdxPGsSV9LyPtKlkd4iGJEmyT1OXpVAI4c+kEz+cVmfiLVx2F+A20g+PnzR1efT5CSH8FDgrSZIOTV0Wbchzs+VxYhNpy/AX4L6mLoS0Xkh/8H4f0lkpz2ni4mwgpMOPd6P6SVW2eCGEH5FOorSU9Bnaq4ApBnBfLCGErUiHys4gHa3Rj3Skw01NWS55bpTyTpwkaZMLITxO+uzfZOCUJJ2wRjkQQrgTGEA6rPFt0uGglyYNmzxIm5mQ/pzNg6Q/hfEl0uHHdwJXJ+lz2moinhuBQZwkSZIk5YqzU0qSJElSjhjESZIkSVKObC4TmzimU5IkSdKWrqQflt9cgjgWLVrU1EVQLcrKyliyZElTF0PKLfuQ1DD2Ianh7Eebty5dupSc1uGUkiRJkpQjBnGSJEmSlCMGcZIkSZKUI5vNM3GSJEmSNr0kSVi9ejXr1q0jhJLm1dBGSpKEZs2a0bp16wbVtUGcJEmStAVbvXo1W221FS1aGBpsCmvXrmX16tVsvfXWG52HwyklSZKkLdi6desM4DahFi1asG7dugblYRAnSZIkbcEcQrnpNbTODeIkSZIkleyGG26gX79+9O/fnwEDBvDcc881KL85c+Zw0kknNVLpai7fLbfcwqpVq+rcvtR0Tcn7ppIkSZJK8swzzzBr1iweeughWrVqxdKlS1mzZk2TlWft2rUbDAWtrXyTJk3i2GOPrfNZtFLTNSXvxEmSJEkqyYcffkj79u1p1aoVAO3bt6dTp0706NGD8ePHM3z4cIYMGcKLL77Id7/7XXr16sWdd94JpDMzjhs3jkMPPZTDDjuMqVOnVsn/+eefZ+DAgbz11lusXLmSc889l8MPP5yBAwfy8MMPAzB58mTGjBnDySefzAknnFBS+W699VY++OADjj/+eI477jgALrjgAoYMGUK/fv245pprAKpN171798r8H3zwQc4++2wApk+fzqGHHkr//v055phjGq2OS+GdOEmSJEkl6du3L9dffz0HH3ww3/nOdzjiiCM46KCDAOjSpQvTp0/n0ksv5ZxzzuGBBx6gvLycfv36cdJJJzFjxgxeeuklHnnkEZYuXcrhhx9Oz549K/N++umnueSSS7jtttvYYYcdGD9+PL179+a6665j2bJlDB06lO985zsAPPvss8yaNYsvf/nLJZVv9OjRTJw4kSlTptC+fXsAzj//fL785S9TUVHBiBEjePnll6tNV5Nf/vKX3HPPPXTu3Jlly5Y1ZjXXyTtxkiRJkkrSpk0bHnroIX7xi1/QoUMHTj/9dCZPngzAwIEDAdhjjz341re+xbbbbkuHDh1o1aoVy5Yt46mnnuKoo46iefPmbL/99vTs2ZMXXngBgPnz53P++edz++23s8MOOwDw5z//mZtuuokBAwZw3HHHUV5ezrvvvgtAnz59qgRwdZWv2PTp0xk0aBCDBg3i73//O6+//nq96uLb3/4255xzDvfccw8VFRX12rahvBMnSZIkqWTNmzenV69e9OrVi913350pU6YAVA5hDCHQsmXLyvTNmjWjoqKCJElqzLNjx46Ul5czb948OnXqBKTDLydOnEi3bt02SPvcc8+xzTbb1Kt8I0aM2CDNwoULmTBhAn/84x9p164dZ599NqtXr642v8KZJMvLyyv/f9VVV/Hcc8/x6KOPMnDgQGbOnFnn3bvG4p04SZIkSSWZP38+b7zxRuXySy+9xI477ljStj179mTatGlUVFTw0UcfMXfuXPbdd18A2rZty5133smVV17JnDlzgHRo5G233VYZ/M2bN6/GvPv06VNn+bbddltWrFgBwD//+U+23npr2rZty+LFi3nssccqtylMB7D99tvz+uuvs27dOh566KHK9QsWLGC//fbjxz/+Me3bt2fRokUl1UNj8E6cJEmSpJKsXLmSn/zkJyxfvpwWLVqw884784tf/IJZs2bVue2QIUN49tlnGTBgACEELr74Yjp27Mj8+fOBNFi64447GDVqFNdeey1nn302l156Kf379ydJEnbcccfKSVIKLV26tDLQq6l8ACeeeCKjRo2iY8eO/Nd//Rd77703/fr1o2vXrhxwwAGV+RWnu/DCCzn55JPp0qULu+22G59++ikAV1xxBW+++SZJknDwwQez1157Nbh+SxVqu625CSWbMnJV/ZWVlbFkyZKmLoaUW/YhqWHsQ1LD1dSPVq5cWevwxM3dI488wsKFCxk9enRTF6Vk1dV5ly5dAEr6FXDvxEmSJEnKrQEDBjR1ETY5n4mTJEmSpBwxiJMkSZKkHDGIkyRJkqQcMYiTJEmSpBwxiJMkSZKkHDGIkyRJkqQc8ScGJEmSJFWqOPWIRs2v+S3T6kyzatUqRo0aRRzHLFq0iJ49ezJu3DhOOeUUAC6++GK+8Y1vMGLEiI0ux7/+67+yZMkSpk+fXrnu2muvpU2bNnz/+9/n6quvpkePHvTp06feeZ9++un8+Mc/5mtf+9pGl68+DOIkbZZuvPHGpi6C6jB27NimLoIk6Qti8uTJDBkyhObNmwPpD5PfeuutjBo1ipYtWzY4/2XLlvHiiy/Spk0bFi5cSNeuXauk+fGPf7zR+Z900kn85je/4eqrr25IMUvmcEpJkiRJTeq+++5j0KBBlcsdOnSgd+/eTJkypUraefPmMWzYMPr378/o0aP55JNP6sx/xowZDBgwgCOPPJKpU6dWm+bss8/mwQcfZPbs2Zx22mmV6+fMmcPJJ58MwBNPPMHw4cMZNGgQY8aM4dNPPwWgR48e/Pd//zdr166t13FvLIM4SZIkSU1mzZo1LFy4kJ122mmD9T/4wQ+YMGECFRUVG6w/++yzufjii5k1axa777471113XZ37eOCBBzjqqKNqDeLW69OnD8899xwrV64EYNq0aRxxxBEsXbqUG264gcmTJ/Pwww/zzW9+k4kTJwLQrFkzdt55Z15++eX6HPpGM4iTJEmS1GSWLl1K27Ztq6zv2rUr++67L/fff3/luuXLl7Ns2TIOOuggAI4//njmzp1ba/6LFy9mwYIFHHjggey66640b96cV199tcb0LVq0oF+/fjzyyCOsXbuWRx99lEGDBvHss8/y2muvceSRRzJ6UkRhAAAgAElEQVRgwACmTJnCO++8U7ldWVkZ77//fn0Pf6P4TJwkSZKkJtO6dWvKy8urfW3s2LGMGTOGHj16bHT+06ZNY9myZfTs2ROAFStWMHXqVHbfffcatxk+fDh33HEH7dq1Y99992XbbbclSRL69OnDzTffXO025eXltG7deqPLWR8GcZ8TJ2XYvDkhgyRJ0uahXbt2VFRUsHr16ipBULdu3ejevTuzZs1i3333pW3btmy33XbMnTuXHj168Ic//KEyOLvtttuAdBbKQg888AB333033/72twFYuHAhJ5xwAueff36NZerVqxc/+tGPuOeeexg+fDgA+++/PxdffDFvvvkmu+yyC6tWrWLRokXsuuuuALzxxhvstttujVMpdTCIkyRJklSplJ8EaGx9+/blqaeeqnZ6/7Fjx24w6ckvf/lLLrjgAlavXk3Xrl0rn4mbP38+BxxwwAbbvv322yxatIj999+/cl3Xrl3Zdtttee6552osT/Pmzenfvz9xHHPDDTcA6WQr119/PWeccQZr1qwB4LzzzmPXXXdl8eLFtG7dmq985SsbXwn1EJIk2SQ7qkOyaNGipi5Do/JO3ObNO3GbP/vQ5s9+pE2prKyMJUuWNHUxpFyrqR+tXLmSbbbZpglK9Jl58+YxYcIEfvWrX210HieddBKTJk1qlJ8kqK+JEyfypS99iRNOOKGk9NXVeZcuXQBCKdt7J06SJElSk9p7773p3bs3FRUVlb8VV1933nlnI5eqdNtttx3HHnvsJtufQZwkSZKkJjdy5MimLsJGGzFixCbdnz8xIEmSJEk5YhAnSZIkSTliECdJkiRJOWIQJ0mSJEk54sQmn5NdvnJSUxdBkiRJqrcj73m1UfObeuLudaZZtWoVo0aNIo5jFi1aRM+ePRk3bhynnHIKABdffDHf+MY3NmoCkcmTJ3PFFVfQuXNnPv30U7761a9yzjnnVPlNuY3Vo0cP/vSnP7HtttsycuRI4jimRYvPN8zyTpwkSZKkJjV58mSGDBlS+fMCZWVl3HrrrZU/qt1QRxxxBDNnzuTJJ5/kjDPO4NRTT+X1119vlLzXa9myJQcffDDTpn3+P5ZuECdJkiSpSd13330MGjSocrlDhw707t2bKVOmVEk7b948hg0bRv/+/Rk9ejSffPJJvfbVu3dvTjzxRO6++24AFixYwIknnsjgwYM5+uijmT9/PgAzZ85k2LBhDBw4kBEjRrB48WIAli5dygknnMDAgQM577zzSJKkMu9BgwZx//331/v468sgTpIkSVKTWbNmDQsXLmSnnXbaYP0PfvADJkyYQEVFxQbrzz77bC6++GJmzZrF7rvvznXXXVfvfe6zzz784x//AOC8885j3LhxPPTQQ1xyySVceOGFABx44IFMnz6dmTNncuSRR3LzzTcDcP3113PggQcyc+ZMBg4cyLvvvluZ7+67787zzz9f7/LUl8/ESZIkSWoyS5cupW3btlXWd+3alX333XeDO1vLly9n2bJlHHTQQQAcf/zxnHbaafXe5/q7Z59++inPPvvsBnmsH8L53nvvcfrpp/Phhx+yZs0aunbtCsD//u//MmnSJAD69+9Pu3btKrdt3rw5LVu2ZMWKFWy77bb1LlepDOIkSZIkNZnWrVtTXl5e7Wtjx45lzJgx9OjRo1H3OW/ePLp168a6deto27YtjzzySJU0l1xyCWPGjGHgwIHMmTNngzt+IYQa8y4vL6dVq1aNWt5iDqeUJEmS1GTatWtHRUUFq1evrvJat27d6N69O7NmzQKgbdu2bLfddsydOxeAP/zhD/Ts2ROA2267jdtuu63O/f3lL3/hnnvu4cQTT+RLX/oSO+20E9OnTwfSO3QvvfQSkN7169SpE8AGz+b17NmT++67D4DZs2dv8Eze0qVL6dChA1tttVW966E+Sr4TF0VRc+AZ4N04jodFUdQemAzsDCwAojiOP87SXgiMBiqAsXEcP9zI5ZYkSZL0OSjlJwEaW9++fXnqqafo06dPldfGjh27waQnv/zlL7ngggtYvXo1Xbt2rbxDNn/+/Bp/NmDatGk89dRTrFq1iq5du3LLLbfQvXt3AH79619z4YUXcsMNN7B27VqOPPJI9tprL374wx9y2mmn0alTJ/bbbz/efvttAM455xzOOOMMBg0aRM+ePdlhhx0q9zNnzhwOPfTQRquXmoTC2VRqE0XRucC3gbZZEPcLYGkcx1dGUXQB8OU4js+PomhP4HfAgUAXYBbw9TiOK2rMHJJFixY16EA2N9Mn12+WHG1aw0e0qzuRmtSNN97Y1EVQHcaOHdvURdAWpKysjCVLljR1MaRcq6kfrVy5km222aYJSvSZefPmMWHCBH71q19tdB4nnXQSkyZNomXLlo1Ysvr5t3/7Ny644AK6detWa7rq6rxLly4ANY/TLFDScMooinYEhgKTClYfCdyR/f8O4KiC9b+P47g8juM3gfmkAZ0kSZIkVbH33nvTu3fvKjNR1sedd97ZpAHcmjVrGDRoUJ0BXGMo9Zm4XwLnAesK1n0ljuP3ALJ/O2brdwDeLkj3TrZOkiRJkqo1cuTIyh/7zqOWLVty/PHHb5J91flMXBRFw4AP4zh+NoqiQ0rIs7pbgFXGbEZRNAYYAxDHMWVlZSVknScOp9ycffHam7Tp2Y+0KbVo0cI2JzVQTf3ogw8+oEULJ63flFq1atWga1opZ6s3cEQURYcDrYG2URTdDXwQRVHnOI7fi6KoM/Bhlv4doPCX+nYEqjzwFsfxRGBitpg4zl2bku1Najj7kTYln4mTGq6mflReXp7rO2B5VF5eXuVcZM/ElaTO4ZRxHF8Yx/GOcRzvDIwEZsdxPAqYBpycJTsZmJr9fxowMoqiVlEU7QJ0B54quUSSJEmSpBo15HfirgQGRFH0OjAgWyaO45eAGHgZeAg4o46ZKSVJkiRJJarX4Nc4jh8HHs/+/xFwWA3pfgb8rIFlkyRJkrSJNfZPZZXy006rVq1i1KhR3HXXXQwePJgJEyawxx57AHDzzTfz1ltvcdVVVzVquQ499FD22muvDX7W4Mwzz2To0KEMHjy48vfgNma2ySiKmDRpEm3btm3MIldqyJ04SZIkSWqwyZMnM2TIELbZZhsuv/xyLrroIpIk4b333uPuu+/mwgsvbNT9vfzyyzRv3pw5c+awatWqatNcf/31G/1zAUcffTR33XVXQ4pYK4M4SZIkSU3qvvvuY9CgQQD069ePjh07MmXKFC677DLOPfdc2rVL7+b9+te/ZujQofTv35/rr78egBUrVjBq1Cj69+/PoYceyoMPPljn/qZOncpxxx1Hr169mDVrVrVpjjrqKObNm8dvf/tbxo8fX7n+3nvv5dJLLwXSWfaHDh3KgAEDuPDCC1m3Lv1FtkGDBnH//fdvfIXUwSBOkiRJUpNZs2YNCxcuZKedPpvg/vLLL+eqq67io48+4rjjjgPg0Ucf5d133+XBBx9k5syZPPPMMzz99NM8+uij7LjjjsyaNYvZs2fTp0+fOvc5ffp0jjjiCI466iimTp1aa9phw4bxxz/+sXJ52rRpHHHEEbz66qs89NBDTJ06lUceeYSKiorKvNq3b8+nn37KsmXLNqZK6uQPQkiSJElqMkuXLq3y7FinTp3o3bs3/fv3r1z3xBNP8NhjjzFw4EAAVq5cyRtvvMF+++3H+PHj+fnPf86AAQM44IADat3fM888Q+fOnencuTNlZWWcd955LF++vMbn1zp27Ejnzp154YUX2GGHHXj77bfZb7/9mDRpEi+88AJDhgwBYPXq1XTu3Llyuw4dOvDhhx+y3XbbbVS91MYgTpIkSVKTad26NeXl5VXWN2vWjGbNNhw4eNZZZ3HCCSdUSTtjxgxmz57NuHHj6N+/P2PHjq1xf1OnTuXVV1+lR48eAPzzn//kT3/6EyNGjKhxmyOOOILp06ez4447cvjhhxNCIEkSRowYwXnnnVftNqtXr6Z169Y15tkQDqeUJEmS1GTatWtHRUUFq1evrjVd3759+d3vfsfKlSsBWLRoEUuXLuW9996jTZs2HHfccYwZM4YXX3wRgCuuuIKZM2dukEdFRQUzZszg8ccfZ+7cucydO5dJkybVOaRy6NCh/OlPf6ocSgnwne98h+nTp7N06VIgvaP47rvvArBu3To+/vjjev2Ad314J06SJElSpVJ+EqCx9e3bl6eeeqrW59kOO+ww5s+fz/DhwwFo06YNN910E6+99hrjx48nhEDLli258sorAXjllVcYNmzYBnk8+eST7LTTTmy//faV63r16sWZZ57J4sWLa9x3+/bt2XnnnVmwYAH77LMPAHvssQfnnnsuI0aMIEkSWrRowZVXXskOO+zAX//6Vw444ACaN2++0XVSm5AkyeeScT0lixYtauoyNKrG/n0NNa6muDipfm688camLoLqUNtQFamxlZWVsWTJkqYuhpRrNfWjlStXss022zRBiT4zb948JkyYsMFvtjVEkiSceOKJ3HvvvY2SX31ddNFFDB8+nIMOOqja16ur8+yuXSglf4dTSpIkSWpSe++9N71796aioqJR8gshNFkAB+nx1BTANQaHU0qSJElqciNHjmzqIjSa7373u59r/t6JkyRJkqQcMYiTJEmSpBwxiJMkSZKkHDGIkyRJkqQccWITSZIkSZUa+2d+SvlJmlWrVjFq1CjuuusuBg8ezIQJE9hjjz0AuPnmm3nrrbe46qqrGqU8V111FXEc0759e1auXMmee+7J+eefT7du3Rqc99q1a9lnn3145ZVX+PDDD/nhD3/IXXfd1Qil3pB34iRJkiQ1qcmTJzNkyBC22WYbLr/8ci666CKSJOG9997j7rvv5sILL2zU/X3/+9/nkUce4cknn2To0KEcf/zxLF26tFH30bFjR9q1a8dzzz3XqPmCQZwkSZKkJnbfffcxaNAgAPr160fHjh2ZMmUKl112Geeeey7t2rUD4Ne//jVDhw6lf//+XH/99QCsWLGCUaNG0b9/fw499FAefPDBeu37qKOOolevXkybNg2A559/nmOPPZbBgwczatQoFi9eDMCdd97J4YcfTv/+/RkzZgyrVq0CYMGCBQwbNozDDz+ca6+9doO8Bw8ezP3337/xFVMDgzhJkiRJTWbNmjUsXLiQnXbaqXLd5ZdfzlVXXcVHH33EcccdB8Cjjz7Ku+++y4MPPsjMmTN55plnePrpp3n00UfZcccdmTVrFrNnz6ZPnz71LsM+++zD/PnzKS8v56c//Sm33HILDz30EMcccwxXX301AMOGDWPGjBnMmjWLnXfemTiOAbjkkksYPXo0M2bMoEOHDhvk+81vfpO5c+dubNXUyGfiJG2WdvnKSU1dBEmStAksXbqUtm3bbrCuU6dO9O7dm/79+1eue+KJJ3jssccYOHAgACtXruSNN95gv/32Y/z48fz85z9nwIABHHDAAfUuQ5IkALz++uu89tprjBgxAoB169bRuXNnAF555RWuueYali9fzooVKyrL9uyzz3L77bcDcOyxx25wN65Dhw588MEH9S5PXQziJEmSJDWZ1q1bU15eXmV9s2bNaNZsw4GDZ511FieccEKVtDNmzGD27NmMGzeO/v37lzSZSqF58+ZVBn977LFHtUMgzzrrLO6++25233137r333spn3UIIhBCqzbe8vJzWrVvXqyylcDilJEmSpCbTrl07KioqWL16da3p+vbty+9+9ztWrlwJwKJFi1i6dCnvvfcebdq04bjjjmPMmDG8+OKLAFxxxRXMnDmzzv1PmzaNOXPmcMQRR9C9e3fef/99/vrXvwLpUM+///3vQDqDZseOHfm///u/DYK8/fbbj+nTpwNUCf7eeOMNdttttxJronTeiZMkSZJUqb53sRpD3759eeqpp2p9nu2www5j/vz5DB8+HIA2bdpw00038dprrzF+/HhCCLRs2ZIrr7wSSIc/Dhs2rNq8/vM//5M4jlm5ciV77LEHU6ZMoX379gBMnDiRSy65hBUrVlBRUcFpp53Gbrvtxo9+9COGDh3KDjvswG677VZ59/A//uM/OPPMM5k4cSKDBw/eYD9z5szhsMMOa3D9FAvrx382sWTRokVNXYZGNX3yJ01dBNVi+Ih2TV0E1cE+tPmzH2lTKisrY8mSJU1dDCnXaupHK1euZJtttmmCEn1m3rx5TJgwgV/96leNkl+SJJx44once++9jZLfxpbh6KOP5s4776zyzF91dd6lSxeA6sdlFnE4pSRJkqQmtffee9O7d28qKioaJb8QQpMGcABLlizh3//936sEcI3B4ZSSJEmSmtzIkSObugiNavvtt6+cSbOxeSdOkiRJ2oJtJo9XbVEaWucGcZIkSdIWrFmzZqxdu7api7HFWLt2bZWfTqgvh1NKkiRJW7DWrVuzevVqysvLa/y9MzWOJElo1qxZg387ziBOkiRJ2oKFENh6662buhiqB4dTSpIkSVKOGMRJkiRJUo4YxEmSJElSjhjESZIkSVKOGMRJkiRJUo4YxEmSJElSjhjESZIkSVKOGMRJkiRJUo4YxEmSJElSjhjESZIkSVKOGMRJkiRJUo4YxEmSJElSjhjESZIkSVKOGMRJkiRJUo4YxEmSJElSjhjESZIkSVKOGMRJkiRJUo4YxEmSJElSjhjESZIkSVKOGMRJkiRJUo4YxEmSJElSjhjESZIkSVKOGMRJkiRJUo60qCtBFEWtgT8DrbL0/xXH8aVRFF0GnAoszpJeFMfxjGybC4HRQAUwNo7jhz+HskuSJEnSFqfOIA4oBw6N43hFFEVbAf8TRdGfsteuj+P4msLEURTtCYwE9gK6ALOiKPp6HMcVjVlwSZIkSdoS1RnExXGcACuyxa2yv6SWTY4Efh/HcTnwZhRF84EDgb80sKySJEmStMUr6Zm4KIqaR1H0PPAh8Egcx3Ozl34QRdHfoij6bRRFX87W7QC8XbD5O9k6SZIkSVIDlTKckmwo5L5RFLUD7o+iaG/gN8A40rty44BrgVOAUE0WVe7cRVE0BhiT5U9ZWdlGHcDm65OmLoBq8cVrb19E9qHNnf1Im1KLFi1sc1ID2Y++OEoK4taL4/iTKIoeBwYXPgsXRdEtwIPZ4jvATgWb7QgsqiavicDEbDFZsmRJfYoiNYjtTWo4+5E2pbKyMtuc1ED2o81bly5dSk5b53DKKIq2z+7AEUXR1kB/4NUoijoXJDsamJf9fxowMoqiVlEU7QJ0B54quUSSJEmSpBqV8kxcZ+CxKIr+BjxN+kzcg8Avoih6MVvfDzgHII7jl4AYeBl4CDjDmSklSZIkqXGEJKltoslNJlm0qMqIy1ybPtnneTZnw0e0a+oiqA72oc2f/UibksPApIazH23esuGU1c0vUkVJs1NKkiRJkjYPBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMGcZIkSZKUIwZxkiRJkpQjBnGSJEmSlCMt6koQRVFr4M9Aqyz9f8VxfGkURe2BycDOwAIgiuP442ybC4HRQAUwNo7jhz+X0kuSJEnSFqaUO3HlwKFxHH8T2BcYHEVRT+AC4NE4jrsDj2bLRFG0JzAS2AsYDNwcRVHzz6PwkiRJkrSlqfNOXBzHCbAiW9wq+0uAI4FDsvV3AI8D52frfx/HcTnwZhRF84EDgb80ZsElSZIkaUtU0jNxURQ1j6LoeeBD4JE4jucCX4nj+D2A7N+OWfIdgLcLNn8nWydJkiRJaqA678QBxHFcAewbRVE74P4oivauJXmoZl1SvCKKojHAmCx/ysrKSilKjnzS1AVQLb547e2LyD60ubMfaVNq0aKFbU5qIPvRF0dJQdx6cRx/EkXR46TPun0QRVHnOI7fi6KoM+ldOkjvvO1UsNmOwKJq8poITMwWkyVLltS37NJGs71JDWc/0qZUVlZmm5MayH60eevSpUvJaescThlF0fbZHTiiKNoa6A+8CkwDTs6SnQxMzf4/DRgZRVGrKIp2AboDT5VcIkmSJElSjUp5Jq4z8FgURX8DniZ9Ju5B4EpgQBRFrwMDsmXiOH4JiIGXgYeAM7LhmJIkSZKkBgpJUuVxtaaQLFpUZcRlrk2f7PM8m7PhI9o1dRFUB/vQ5s9+pE3JYWBSw9mPNm/ZcMrq5hepoqTZKSVJkiRJmweDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKEYM4SZIkScoRgzhJkiRJyhGDOEmSJEnKkRZ1JYiiaCfgTqATsA6YGMfxDVEUXQacCizOkl4Ux/GMbJsLgdFABTA2juOHP4eyS5IkSdIWp84gDlgL/DCO4+eiKPoS8GwURY9kr10fx/E1hYmjKNoTGAnsBXQBZkVR9PU4jisas+CSJEmStCWqczhlHMfvxXH8XPb/fwKvADvUssmRwO/jOC6P4/hNYD5wYGMUVpIkSZK2dKXciasURdHOwLeAuUBv4AdRFJ0EPEN6t+5j0gDvfws2e4fagz5JkiRJUolKDuKiKNoW+ANwdhzHy6Mo+g0wDkiyf68FTgFCNZsn1eQ3BhgDEMcxZWVl9S/9Zu2Tpi6AavHFa29fRPahzZ39SJtSixYtbHNSA9mPvjhKCuKiKNqKNIC7J47j+wDiOP6g4PVbgAezxXeAnQo23xFYVJxnHMcTgYnZYrJkyZJ6F17aWLY3qeHsR9qUysrKbHNSA9mPNm9dunQpOW2dz8RFURSAW4FX4ji+rmB954JkRwPzsv9PA0ZGUdQqiqJdgO7AUyWXSJIkSZJUo1LuxPUG/gV4MYqi57N1FwEnRFG0L+lQyQXAaQBxHL8URVEMvEw6s+UZzkwpSZIkSY2jziAujuP/ofrn3GbUss3PgJ81oFySJEmSpGrUOZxSkiRJkrT5MIiTJEmSpBwxiJMkSZKkHDGIkyRJkqQcKfnHviVJkqQtxY033tjURVAdxo4d29RFaDLeiZMkSZKkHDGIkyRJkqQcMYiTJEmSpBwxiJMkSZKkHDGIkyRJkqQcMYiTJEmSpBwxiJMkSZKkHPF34iRJ+gLyN642f1vyb1xJahjvxEmSJElSjhjESZIkSVKOGMRJkiRJUo4YxEmSJElSjhjESZIkSVKOGMRJkiRJUo4YxEmSJElSjhjESZIkSVKOGMRJkiRJUo4YxEmSJElSjrRo6gJ8UU1a+35TF0G1GE67pi6CJEmStFG8EydJkiRJOWIQJ0mSJEk5YhAnSZIkSTliECdJkiRJOWIQJ0mSJEk5YhAnSZIkSTniTwx8Tu57/LymLoJqc+K0pi6BJEmStFG8EydJkiRJOWIQJ0mSJEk5YhAnSZIkSTliECdJkiRJOWIQJ0mSJEk5YhAnSZIkSTliECdJkiRJOWIQJ0mSJEk5YhAnSZIkSTliECdJkiRJOWIQJ0mSJEk5YhAnSZIkSTliECdJkiRJOWIQJ0mSJEk5YhAnSZIkSTliECdJkiRJOWIQJ0mSJEk5YhAnSZIkSTliECdJkiRJOWIQJ0mSJEk5YhAnSZIkSTnSoq4EURTtBNwJdALWARPjOL4hiqL2wGRgZ2ABEMVx/HG2zYXAaKACGBvH8cOfS+klSZIkaQtTyp24tcAP4zjeA+gJnBFF0Z7ABcCjcRx3Bx7NlsleGwnsBQwGbo6iqPnnUXjp/9u78yjNzrpO4N82EYSooDaEdBIIalABISgJIiqbIDpAwqg/QTTNIkEGcD0yiEDCYeLEBRkPizMxZJFJCD8kmrCMiChw1IEYFCVhUYRA2mw0i7IMwYSeP+4tLZrqruquKquers/nnD793uc+975P9Xl//db33ufeCwAAW82yIa67r+vuv55ffzrJ+5IcneTkJBfM3S5Icsr8+uQkF3f3Td394SQfTHLSWg8cAABgKzqga+Kq6rgk90nyziRHdvd1yRT0ktxx7nZ0kmsWbbZrbgMAAGCVlr0mbkFVfXWS1yb5ue7+l6raV9dtSxy1XI8AABigSURBVLTtWWJ/pyU5LUm6O9u3b1/pUIZww0YPgP061D5vh6ZPbfQAWIY6gtVRQ5vbXY88daOHwDK2cg2tKMRV1VdmCnAXdvclc/MNVXVUd19XVUcluXFu35Xk2EWbH5Pk2r332d1nJzl7Xtyze/fugxk/HBSfN1g9dQSro4ZgdQ61GtqxY8eK+67k7pTbkrwiyfu6+7cWrbosyc4kZ81/X7qo/aKq+q0kO5Icn+TyFY8IAACAfVrJmbgHJPnJJO+pqnfPbc/JFN66qp6c5KNJfjRJuvuqquok7810Z8und/ctaz5yAACALWjZENfdf56lr3NLkofuY5szk5y5inEBAACwhAO6OyUAAAAbS4gDAAAYiBAHAAAwECEOAABgICt+2DcAAGwV59x8/UYPgWU8Krff6CFsGGfiAAAABuJMHLApOQK6+W3lI6AAsJGciQMAABiIEAcAADAQIQ4AAGAgQhwAAMBAhDgAAICBCHEAAAADEeIAAAAG4jlxAHAIuuuRp270EABYJ87EAQAADESIAwAAGIgQBwAAMBAhDgAAYCBCHAAAwECEOAAAgIEIcQAAAAMR4gAAAAYixAEAAAxEiAMAABiIEAcAADAQIQ4AAGAgQhwAAMBAhDgAAICBCHEAAAADOXyjBwCwlEve+qyNHgLLefxlGz0CANiSnIkDAAAYiBAHAAAwECEOAABgIEIcAADAQIQ4AACAgQhxAAAAAxHiAAAABiLEAQAADESIAwAAGIgQBwAAMBAhDgAAYCBCHAAAwECEOAAAgIEIcQAAAAMR4gAAAAYixAEAAAxEiAMAABiIEAcAADAQIQ4AAGAgQhwAAMBAhDgAAICBCHEAAAADOXy5DlV1bpJHJrmxu+85t52R5ClJPjZ3e053v3Fe98tJnpzkliQ/091vWodxAwAAbEnLhrgk5yd5aZLf26v9xd39m4sbquruSR6b5B5JdiT5k6q6W3ffsgZjBQAA2PKWnU7Z3W9P8okV7u/kJBd3903d/eEkH0xy0irGBwAAwCIrORO3L8+oqlOTXJHkF7v7k0mOTvKORX12zW0AAACsgYMNcb+T5IVJ9sx/vyjJk5JsW6LvnqV2UFWnJTktSbo727dvP8ihbE43bPQA2K9D7fN2KFJDm5862uw+tdEDYBlqCFZnK9fQQYW47v6336+q6neTvH5e3JXk2EVdj0ly7T72cXaSs+fFPbt37z6YocBB8XmD1VNHsDpqCFbnUKuhHTt2rLjvQT1ioKqOWrT4mCRXzq8vS/LYqrp1Vd01yfFJLj+Y9wAAAODLreQRA69K8qAk26tqV5LTkzyoqk7INFXy6iRPTZLuvqqqOsl7k9yc5OnuTAkAALB2lg1x3f24JZpfsZ/+ZyY5czWDAgAAYGkHNZ0SAACAjSHEAQAADESIAwAAGIgQBwAAMBAhDgAAYCBCHAAAwECEOAAAgIEIcQAAAAMR4gAAAAYixAEAAAxEiAMAABiIEAcAADAQIQ4AAGAgQhwAAMBAhDgAAICBCHEAAAADEeIAAAAGIsQBAAAMRIgDAAAYiBAHAAAwECEOAABgIEIcAADAQIQ4AACAgQhxAAAAAxHiAAAABiLEAQAADESIAwAAGIgQBwAAMBAhDgAAYCBCHAAAwECEOAAAgIEcvtEDAADW3jk3X7/RQ2AZj8rtN3oIwKCciQMAABiIEAcAADAQIQ4AAGAgQhwAAMBAhDgAAICBCHEAAAADEeIAAAAGIsQBAAAMRIgDAAAYiBAHAAAwECEOAABgIEIcAADAQIQ4AACAgQhxAAAAAxHiAAAABiLEAQAADESIAwAAGIgQBwAAMBAhDgAAYCBCHAAAwECEOAAAgIEcvlyHqjo3ySOT3Njd95zbvj7Jq5Mcl+TqJNXdn5zX/XKSJye5JcnPdPeb1mXkAAAAW9BKzsSdn+QRe7U9O8lbuvv4JG+Zl1NVd0/y2CT3mLd5eVUdtmajBQAA2OKWDXHd/fYkn9ir+eQkF8yvL0hyyqL2i7v7pu7+cJIPJjlpjcYKAACw5R3sNXFHdvd1STL/fce5/egk1yzqt2tuAwAAYA0se03cAdq2RNuepTpW1WlJTkuS7s727dvXeCgb64aNHgD7dah93g5FamjzU0ewOmoIVmcr19DBhrgbquqo7r6uqo5KcuPcvivJsYv6HZPk2qV20N1nJzl7Xtyze/fugxwKHDifN1g9dQSro4ZgdQ61GtqxY8eK+x5siLssyc4kZ81/X7qo/aKq+q0kO5Icn+Tyg3wPAAAA9rKSRwy8KsmDkmyvql1JTs8U3rqqnpzko0l+NEm6+6qq6iTvTXJzkqd39y3rNHYAAIAtZ9kQ192P28eqh+6j/5lJzlzNoAAAAFjawd6dEgAAgA0gxAEAAAxEiAMAABiIEAcAADAQIQ4AAGAgQhwAAMBAhDgAAICBCHEAAAADEeIAAAAGIsQBAAAMRIgDAAAYiBAHAAAwECEOAABgIEIcAADAQIQ4AACAgQhxAAAAAxHiAAAABiLEAQAADOTwjR4AALD2LnnrszZ6CCzn8Zdt9AiAQTkTBwAAMBAhDgAAYCBCHAAAwECEOAAAgIEIcQAAAAMR4gAAAAYixAEAAAxEiAMAABiIEAcAADAQIQ4AAGAgQhwAAMBAhDgAAICBCHEAAAADEeIAAAAGIsQBAAAMRIgDAAAYiBAHAAAwECEOAABgIEIcAADAQIQ4AACAgQhxAAAAAxHiAAAABiLEAQAADESIAwAAGIgQBwAAMBAhDgAAYCBCHAAAwECEOAAAgIEIcQAAAAMR4gAAAAYixAEAAAxEiAMAABiIEAcAADCQw1ezcVVdneTTSW5JcnN337eqvj7Jq5Mcl+TqJNXdn1zdMAEAAEjW5kzcg7v7hO6+77z87CRv6e7jk7xlXgYAAGANrMd0ypOTXDC/viDJKevwHgAAAFvSakPcniR/XFXvqqrT5rYju/u6JJn/vuMq3wMAAIDZqq6JS/KA7r62qu6Y5M1V9f6VbjiHvtOSpLuzffv2VQ5lc7lhowfAfh1qn7dDkRra/NTR5qaGNj81BKuzlWtoVSGuu6+d/76xqv4gyUlJbqiqo7r7uqo6KsmN+9j27CRnz4t7du/evZqhwAHxeYPVU0ewOmoIVudQq6EdO3asuO9BT6esqiOq6msWXid5eJIrk1yWZOfcbWeSSw/2PQAAAPhSq7km7sgkf15Vf5vk8iRv6O4/SnJWkodV1T8kedi8DAAAwBo46OmU3f2hJPdeov3jSR66mkEBAACwtPV4xAAAAADrRIgDAAAYiBAHAAAwECEOAABgIEIcAADAQIQ4AACAgQhxAAAAAxHiAAAABiLEAQAADESIAwAAGIgQBwAAMBAhDgAAYCBCHAAAwECEOAAAgIEIcQAAAAMR4gAAAAYixAEAAAxEiAMAABiIEAcAADAQIQ4AAGAgQhwAAMBAhDgAAICBHL7RAwAAgM3mkrc+a6OHwHIef9lGj2DDOBMHAAAwECEOAABgIEIcAADAQIQ4AACAgQhxAAAAAxHiAAAABiLEAQAADESIAwAAGIgQBwAAMBAhDgAAYCBCHAAAwECEOAAAgIEIcQAAAAMR4gAAAAYixAEAAAxEiAMAABiIEAcAADAQIQ4AAGAgQhwAAMBAhDgAAICBCHEAAAADEeIAAAAGIsQBAAAMRIgDAAAYiBAHAAAwECEOAABgIEIcAADAQIQ4AACAgQhxAAAAAxHiAAAABnL4eu24qh6R5LeTHJbknO4+a73eCwAAYKtYlzNxVXVYkpcl+cEkd0/yuKq6+3q8FwAAwFayXtMpT0rywe7+UHd/IcnFSU5ep/cCAADYMtYrxB2d5JpFy7vmNgAAAFZhva6J27ZE257FC1V1WpLTkqS7s2PHjnUaygZ5wxUbPQIYmxqC1VFDsDpqiE1svULcriTHLlo+Jsm1izt099lJzl6n92eNVdUV3X3fjR4HjEoNweqoIVg9dXToWK8Q91dJjq+quyb5pySPTfLj6/ReAAAAW8a6XBPX3TcneUaSNyV539TUV63HewEAAGwl6/acuO5+Y5I3rtf++Q9n6iusjhqC1VFDsHrq6BCxbc+ePcv3AgAAYFNYr0cMAAAAsA7WbTol66eqzk3yyCQ3dvc9F7WfkOR/JvmqJDcn+S/dffkS278w08PXv5jkxiRP6O5rF62/c5L3Jjmju39zPX8W+I9WVV+V5O1Jbp3p/8Df7+7T53VnJHlKko/N3Z8zTw3fex/77aeG2Aqq6uokn05yS5KbF+54t9I6mvs+M9M19DcneUN3P2vROnXEIaOqjknysiR3z3QS5fVJfqm7v7AO73X7JOckuWemR3w9qbv/74HUJpufM3FjOj/JI5Zo//UkL+juE5I8f15eym90973mfq+f+y724iT/Z43GuiJV5YAC/1FuSvKQ7r53khOSPKKqvmvR+hd39wnzn/19ue2vnxpiq3jwXAN737J82TqqqgdnOqB4r+6+R5K9g5o64pBQVduSXJLkD7v7+CR3S/LVSc5cg30v9Zn97SR/1N3fmuTemW4yuGCl33Frpqq2VZXMscb8ZzWg7n57VR23xKo9Sb52fn277PVsvkXb/8uixSOy6EHsVXVKkg8l+exS21bVQ5M8o7sfMy8/LMnTuvs/V9XDk7wg0xmOf0zyxO7+TFU9P8mjktwmyV8meWp376mqt87LD0hyWVV9NMnpmY7q/nN3f99y/xZwoLp7T5LPzItfOf9Zs4uD1RCs2NOSnNXdNyVJd9+4sEIdcYh5SJLPd/d5SdLdt1TVzyf5cFWdnuTPMp0tuypJ5s/kLyZ5f5KXJPn2TL+zn9Hdl1bVE5L8p0wzr46Y9595269N8n1JnjC/1xeSrPhsX1W9MtMMlUvn5QuTvDrJG5KcleRBmWrrZd39v6rqq5NcmuTrMn2fPnce43GZDsL8WZL7Jzmlql6Q5L6ZvnPP7e4Xr3RcfDmp+NDyc0l+o6quyXRE85f31bGqzpz7PT7zmbiqOiLJf8305bcvf5rk26rqDvPyE5OcV1Xbkzw3yfd393ckuSLJL8x9XtrdJ85TP2+TaSrogtt39wO7+0XzOH5gPkPy6AP5weFAVNVhVfXuTNOJ39zd71y0+hlV9XdVdW5Vfd1+dvNl/dQQW8yeJH9cVe+qqtP2WreSOrpbku+tqndW1duq6sREHXFIukeSdy1umA+ofzTJNye5OEklSVUdlWRHd78rya8k+dPuPjHJgzP9jnfEvIv7J9nZ3Q/Jl/rGTNMlz6uqv6mqcxZtkyxfm+dkqqdU1e2SfHemu80/OdNBjROTnJjkKfPzoD+f5DFzvT04yYvmM49J8i1Jfq+775Nke5Kju/ue3f3tSc5b4b8d+yDEHVqeluTnu/vYJD+f5BX76tjdvzL3uzDT9QjJ9IX54u7+zH6225PklUl+Yp5zff9MR1q+K9M877+YfznemeQu82YPnr+k35PpaNE9Fu3y1Yte/0WS86vqKUkOW+HPDAesu2+ZpxMfk+Skqlq4tvR3knxTpmmW1yV50T52sa9+aoit5AHzL24/mOTpVbVwxmqldXR4pqP335Xkl5L0/MufOuJQsy1Lz/hYaO8kPzq3VZLXzK8fnuTZ82f5rZnOvN15Xvfm7v7EEvs8PMl3JPmdOTx9Nsmz53XL1mZ3vy3JN1fVHZM8Lslr5+c/PzzJqfNY3pnkG5IcP/8Mv1pVf5fkT5IcneTIeXcf6e53zK8/lOQbq+olVfWIJItnhXEQTKc8tOxM8rPz69dkOpqSqjovyX2SXNvdP7TXNhdlOkV+epL7JfmRqvr1JLdP8sWq+nx3v3Svbc5L8rpMR19e0903z1+8b+7uxy3uON9E4uVJ7tvd18wX1X7Voi7/NlWmu3+6qu6XaYrAu6vqhO7++MH8Q8BKdPen5mkrj0hyZXffsLCuqn430zWjX1ZD++oXNcQWsnBDrO6+sar+IMlJSd6+0jpKsivJJXMgu7yqvpjpaL064lBzVZIfXtwwT3s8Nsk/dvfnqurjVXWvJD+W5Klzt21Jfri7P7DXtvfLPqYaZ6qrXYtmmPx+5hC3n++uvb0y00ytxyZ50qKxPLO737TXWJ6Q5A5JvrO7/7WmGx4t1NbiuvpkVd07yQ8keXqmsPqkcNCciTu0XJvkgfPrhyT5hyTp7ifOF7D+UJJU1fGLtnl0pjnX6e7v7e7juvu4JP8jya8u8aW58MV9baYpK+fPze9I8oCq+ub5PW5bVXfLvxfy7nne9I/sa/BV9U3d/c7ufn6S3Zn+c4M1VVV3mI/cp6puk+T7M9fAPI1lwWOSXJksWUP76qeG2BKq6oiq+pqF15mO0l85L6+ojpL8YeZreebP+q2S7FZHHILekuS2VXVqMk3pz3QW7Pzu/tzc5+Ikz0pyu+5+z9z2piTPXJieWFX3We6Nuvv6JNdU1bfMTQ/NdJfXfdbmEs7PdIlOFq7Tm8fytKr6ynlfd5tr/3aZ7pb+rzXdrOguS+wv81Tnr+ju1yZ5XqazhayCM3EDqqpXZbqwdHtV7Upyene/ItNtY3+7pjsVfT7J3tcoLDhrLu4vJvlIkp8+iGFcmOQO3f3eJOnuj81HY15VVbee+zy3u/9+PtrzniRXJ/mr/ezzN+aAuS3Tf3h/exDjguUcleSC+Uv0K5J0dy8cjfz1mh7VsSfT5/WpS+9ixf32Rw0xsiOT/EFVJdPvEhd19x/N61ZaH+cmObeqrsx044Wd81m5A6GO2PR6uoHOY5K8vKqel+m7541JnrOo2+9nuqvkCxe1vTDTgYy/m4Pc1fnSazn35ZlJLqyqW2WaxvjEuX1FtdndN1TV+zIdaFlwTpLjkvz1PJaPJTklUw2+rqquSPLuzAdFl3B0puv0Fk4g7fO+DazMtj171uymbGwhVfXSJH8zh0fgAKkhWD11BGuvqm6b6YDHd3T3P2/0eFia6ZQcsKp6V5J7JfnfGz0WGJEagtVTR7D2qmrhEoOXCHCbmzNxAAAAA3EmDgAAYCBCHAAAwECEOAAAgIEIcQAAAAPxnDgANqWqemuSeye5U3fftMHDAYBNw5k4ADadqjouyfdmeijtozd2NJOqcuATgE3BFxIAm9GpSd6R5J1JdiZ5zcKKqvqGJOcneWCSDyR5U5IHdff3zOu/NclLknxnko8leV5391JvUlV3TXJBkvvM7/WBJLfr7p+Yg+SHk/xUktOTXJ3k+6rq0Un+e5Kjk7w7ydO6+33z/vYkOb67Pzgvn59kV3c/t6oelOmZZi9P8gtJPpPkV7r7wlX9SwGw5TgTB8BmdGqSC+c/P1BVRy5a97Ikn01yp0wBb+fCiqo6Ismbk1yU5I5JHpfk5VV1j328z0VJLk/yDUnOSPKTS/R5YJJvm8dxtySvSvJzSe6Q5I1JXldVt1rhz3WnJNszBcCdSc6uqm9Z4bYAkESIA2CTqarvSXKXJN3d70ryj0l+fF53WJIfTnJ6d3+uu9+b6Uzagkcmubq7z+vum7v7r5O8NsmPLPE+d05yYpLnd/cXuvvPk1y2xJDO6O7Pdvf/S/JjSd7Q3W/u7n9N8ptJbpPkuw/gR3xed9/U3W9L8oYkdQDbAoDplABsOjuT/HF3756XL5rbXpzp7NfhSa5Z1H/x67skuV9VfWpR2+FJXrnE++xI8onu/txe+zp2r36L978jyUcWFrr7i1V1TaYzayvxye7+7KLlj8z7BIAVE+IA2DSq6jaZzkwdVlXXz823TnL7qrp3kiuT3JzkmCR/P69fHLquSfK27n7YCt7uuiRfX1W3XRTk9g5wyXRzlQXXJvn2RePdNm/zT3PT55LcdlH/OyXZtWj566rqiEVB7s7zzwQAKybEAbCZnJLklkxB6QuL2jvJqd39i1V1SZIzquqnMoWgU5N8dO73+iRnVdVPJrl4bjshyWcWbj7ybzvs/khVXTHv67mZboTyqCSv28/4Osmzq+qhSd6e5GeT3JTkL+f1707y41V1VZKHZbqe7oq99vGCqnpOkvtlmv55+jL/JgDwJVwTB8BmsjPJed390e6+fuFPkpcmefx8m/9nJLldkuszTZN8VaYgle7+dJKHJ3lsprNm1yf5tUxn85by+CT3T/LxJP8tyasX9rWU7v5Akp/IdPfL3ZlC36O6eyFw/uzc9ql533+41y6uT/LJeWwXJvnp7n7/8v8sAPDvtu3Zs2f5XgCwSVXVr2V6IPjOZTsvv69XJ3l/d6/52bGFRwx09zFrvW8AthbTKQEYyvwcuFsleU+mu0s+OdOz3A5mXycm+USm58E9PMnJSc5am5ECwPoQ4gAYzddkmkK5I8mNSV6U5NKD3NedklyS6TlxuzI9uPtv1mKQALBeTKcEAAAYiBubAAAADESIAwAAGIgQBwAAMBAhDgAAYCBCHAAAwECEOAAAgIH8f68fybwukjlDAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "_, ax = plt.subplots(figsize=(15, 10))\n",
    "data.pivot_table(index=['Age group'], columns=['Smoker', 'Status'], aggfunc='size').plot(kind='bar', stacked=True, ax=ax)\n",
    "ax.set_title(\"Number of alive/dead women after 20 years, according to their smoking habits and age\")\n",
    "ax.tick_params(axis=\"x\", rotation=0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One can see that most women that were over 65 in 1972 are dead twenty years after and that at the time of the survey a great majority of polled older women were non-smokers. For the other age groups results look similar but are difficult to interpret.\n",
    "\n",
    "Let's therefore compute the mortality rates for smokers and non-smokers in different age groups."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Age group: 18-34 years\n",
      "\tMortality rate for smokers:\t3.70% ± 2.69%\n",
      "\tMortality rate for non smokers:\t2.64% ± 2.09%\n",
      "\n",
      "Age group: 35-54 years\n",
      "\tMortality rate for smokers:\t17.03% ± 4.87%\n",
      "\tMortality rate for non smokers:\t9.95% ± 4.24%\n",
      "\n",
      "Age group: 55-64 years\n",
      "\tMortality rate for smokers:\t44.35% ± 9.08%\n",
      "\tMortality rate for non smokers:\t33.06% ± 8.38%\n",
      "\n",
      "Age group: Over 65 years\n",
      "\tMortality rate for smokers:\t85.71% ± 9.80%\n",
      "\tMortality rate for non smokers:\t85.49% ± 4.97%\n",
      "\n"
     ]
    }
   ],
   "source": [
    "def ci_per_age(age_group):\n",
    "    pop = data[data['Age group'] == group]\n",
    "    print(f\"Age group: {group}\")\n",
    "    \n",
    "    num_smokers = sum(pop['Smoker'] == \"Yes\")\n",
    "    num_dead_smokers = sum(np.logical_and(pop['Smoker'] == \"Yes\", pop['Status'] == \"Dead\"))\n",
    "    rate_smokers = num_dead_smokers / num_smokers\n",
    "    ci_smokers = z * (rate_smokers * (1 - rate_smokers) / num_smokers) ** 0.5\n",
    "    print(f\"\\tMortality rate for smokers:\\t{rate_smokers:.2%} \" + u\"\\u00B1\" + f\" {ci_smokers:.2%}\")\n",
    "\n",
    "    num_non_smokers = len(pop) - num_smokers\n",
    "    num_dead_non_smokers = sum(np.logical_and(pop['Smoker'] == \"No\", pop['Status'] == \"Dead\"))\n",
    "    rate_non_smokers = num_dead_non_smokers / num_non_smokers\n",
    "    ci_non_smokers = z * (rate_non_smokers * (1 - rate_non_smokers) / num_non_smokers) ** 0.5\n",
    "    print(f\"\\tMortality rate for non smokers:\\t{rate_non_smokers:.2%} \" + u\"\\u00B1\" + f\" {ci_non_smokers:.2%}\")\n",
    "    \n",
    "for group in sorted(data['Age group'].unique()):\n",
    "    ci_per_age(group)\n",
    "    print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, the mortality rate is considerably higher for smokers, especially for people between 35 and 65 years old. This might seem like a contradiction, as before the rate was higher for non-smokers. However, from the previous bar chart it is clear that the percentage of polled smokers/non-smokers is different in different age groups, in particular for older women as mentioned. In addition, the fact that most polled women over 65 are non-smokers can be an argument in favor of the hypothesis that smoking is dangerous for health, but this can't be proven through statistics.\n",
    "\n",
    "---\n",
    "\n",
    "The age groups are fixed a-priori. In order to have more flexible results and reduce the introduced bias, it is possible to try to perform a logistic regression, studying the probability of death in the two groups according to the age. A column of 1 is also added as intercept for the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>Smoker</th>\n",
       "      <th>Status</th>\n",
       "      <th>Age</th>\n",
       "      <th>Age group</th>\n",
       "      <th>Death</th>\n",
       "      <th>Intercept</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Yes</td>\n",
       "      <td>Alive</td>\n",
       "      <td>21.0</td>\n",
       "      <td>18-34 years</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Yes</td>\n",
       "      <td>Alive</td>\n",
       "      <td>19.3</td>\n",
       "      <td>18-34 years</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>No</td>\n",
       "      <td>Dead</td>\n",
       "      <td>57.5</td>\n",
       "      <td>55-64 years</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>No</td>\n",
       "      <td>Alive</td>\n",
       "      <td>47.1</td>\n",
       "      <td>35-54 years</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Yes</td>\n",
       "      <td>Alive</td>\n",
       "      <td>81.4</td>\n",
       "      <td>Over 65 years</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Smoker Status   Age      Age group  Death  Intercept\n",
       "0    Yes  Alive  21.0    18-34 years      0          1\n",
       "1    Yes  Alive  19.3    18-34 years      0          1\n",
       "2     No   Dead  57.5    55-64 years      1          1\n",
       "3     No  Alive  47.1    35-54 years      0          1\n",
       "4    Yes  Alive  81.4  Over 65 years      0          1"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "death = []\n",
    "for status in data['Status']:\n",
    "    if status == \"Alive\":\n",
    "        death.append(0)\n",
    "    elif status == \"Dead\":\n",
    "        death.append(1)\n",
    "assert len(death) == len(data['Age'])\n",
    "data['Death'] = death\n",
    "data['Intercept'] = 1\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.412727\n",
      "         Iterations 7\n",
      "                           Logit Regression Results                           \n",
      "==============================================================================\n",
      "Dep. Variable:                  Death   No. Observations:                  582\n",
      "Model:                          Logit   Df Residuals:                      580\n",
      "Method:                           MLE   Df Model:                            1\n",
      "Date:                Thu, 02 Dec 2021   Pseudo R-squ.:                  0.2492\n",
      "Time:                        18:06:08   Log-Likelihood:                -240.21\n",
      "converged:                       True   LL-Null:                       -319.94\n",
      "                                        LLR p-value:                 1.477e-36\n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept     -5.5081      0.466    -11.814      0.000      -6.422      -4.594\n",
      "Age            0.0890      0.009     10.203      0.000       0.072       0.106\n",
      "==============================================================================\n"
     ]
    }
   ],
   "source": [
    "smokers = data[data['Smoker'] == \"Yes\"]\n",
    "log_smokers = sm.Logit(smokers[['Death']], smokers[['Intercept', 'Age']]).fit()\n",
    "print(log_smokers.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.354560\n",
      "         Iterations 7\n",
      "                           Logit Regression Results                           \n",
      "==============================================================================\n",
      "Dep. Variable:                  Death   No. Observations:                  732\n",
      "Model:                          Logit   Df Residuals:                      730\n",
      "Method:                           MLE   Df Model:                            1\n",
      "Date:                Thu, 02 Dec 2021   Pseudo R-squ.:                  0.4304\n",
      "Time:                        18:06:08   Log-Likelihood:                -259.54\n",
      "converged:                       True   LL-Null:                       -455.62\n",
      "                                        LLR p-value:                 2.808e-87\n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept     -6.7955      0.479    -14.174      0.000      -7.735      -5.856\n",
      "Age            0.1073      0.008     13.742      0.000       0.092       0.123\n",
      "==============================================================================\n"
     ]
    }
   ],
   "source": [
    "non_smokers = data[data['Smoker'] == \"No\"]\n",
    "log_non_smokers = sm.Logit(non_smokers[['Death']], non_smokers[['Intercept', 'Age']]).fit()\n",
    "print(log_non_smokers.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In both fitted models, the p value (`P>|z|` in *statsmodels*) for the hypothesis that age does not have any impact on death probability is 0. The hypothesis is therefore discarded (as one might have imagined) and age and intercept are taken into account for the analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_pred_confidence_intervals(model, X, xlabel, label=None):\n",
    "    # https://stackoverflow.com/questions/47414842/confidence-interval-of-probability-prediction-from-logistic-regression-statsmode\n",
    "    preds = model.predict(X)\n",
    "    \n",
    "    cov = model.cov_params()\n",
    "    gradient = np.array((preds * (1 - preds) * X.T).T)\n",
    "    std_errors = np.array([np.sqrt(np.dot(np.dot(g, cov), g)) for g in gradient])\n",
    "    upper = np.maximum(0, np.minimum(1, preds + std_errors * z))\n",
    "    lower = np.maximum(0, np.minimum(1, preds - std_errors * z))\n",
    "    \n",
    "    plt.plot(X[xlabel], preds, label=label)\n",
    "    plt.fill_between(X[xlabel], lower, upper, alpha=0.3)\n",
    "    \n",
    "\n",
    "plt.figure(figsize=(15, 10))\n",
    "data_pred = pd.DataFrame({'Age': np.linspace(start=15, stop=100, num=100), 'Intercept': 1})\n",
    "\n",
    "plot_pred_confidence_intervals(log_smokers, data_pred[['Intercept', 'Age']], 'Age', label=\"Smokers\")\n",
    "plot_pred_confidence_intervals(log_non_smokers, data_pred[['Intercept', 'Age']], 'Age', label=\"Non-smokers\")\n",
    "\n",
    "plt.title(\"Estimated probability of death per age, according to smoking habits\")\n",
    "plt.xlabel(\"Age\")\n",
    "plt.ylabel(\"Probability\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the previous plot, we can conclude with a good confidence that the death probability is lower for non-smokers up to the age of 45/50. From this point, the confidence interval for non-smokers become wider (due to the lower number of women smokers over 65 years of age as highlighted earlier) and predicted probabilities cross around the age of 70. Estimated death probability becomes higher for non-smokers, but results are difficult to interpret and it is not possible to draw any clear conclusion about the impact of smoking on life expectancy.\n",
    "\n",
    "We can then try to compare the obtained curve with the predictions produced by *scikit-learn*, probably the most popular Python library for machine learning. Differently from *statsmodels*, *scikit-learn* does not allow to derive prediction confidence intervals and uses regularization by default."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "sklog_smokers = LogisticRegression().fit(smokers['Age'].values.reshape([-1, 1]), smokers['Death'])\n",
    "sklog_non_smokers = LogisticRegression().fit(non_smokers['Age'].values.reshape([-1, 1]), non_smokers['Death'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 10))\n",
    "data_pred = np.linspace(start=15, stop=100, num=100)\n",
    "\n",
    "smoker_proba = sklog_smokers.predict_proba(data_pred.reshape([-1, 1]))\n",
    "plt.plot(data_pred, smoker_proba[:, 1], label=\"Smokers\")\n",
    "\n",
    "non_smoker_proba = sklog_non_smokers.predict_proba(data_pred.reshape([-1, 1]))\n",
    "plt.plot(data_pred, non_smoker_proba[:, 1], label=\"Non-smokers\")\n",
    "\n",
    "plt.title(\"Estimated probability of death per age, according to smoking habits\")\n",
    "plt.xlabel(\"Age\")\n",
    "plt.ylabel(\"Probability\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For *scikit-learn*, the plot of estimated probabilities cross earlier than for *statsmodels*. The lack of confidence intervals makes results non-interpretable."
   ]
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}