diff --git a/module3/exo3/exercice.ipynb b/module3/exo3/exercice.ipynb index 0bbbe371b01e359e381e43239412d77bf53fb1fb..4bbfd07b14090fff0edaac32333bcc9808087e0f 100644 --- a/module3/exo3/exercice.ipynb +++ b/module3/exo3/exercice.ipynb @@ -1,5 +1,755 @@ { - "cells": [], + "cells": [ + { + "cell_type": "code", + "execution_count": 85, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import statsmodels.formula.api as smf" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Autour du paradoxe de Simpson\n", + "### Contexte\n", + "En 1972-1974, à Whickham, une ville du nord-est de l'Angleterre, située à environ 6,5 kilomètres au sud-ouest de Newcastle upon Tyne, un sondage d'un sixième des électeurs a été effectué afin d'éclairer des travaux sur les maladies thyroïdiennes et cardiaques (Tunbridge et al. 1977). Une suite de cette étude a été menée vingt ans plus tard (Vanderpump et al. 1995). Certains des résultats avaient trait au tabagisme et cherchaient à savoir si les individus étaient toujours en vie lors de la seconde étude. Par simplicité, nous nous restreindrons aux femmes et parmi celles-ci aux 1314 qui ont été catégorisées comme \"fumant actuellement\" ou \"n'ayant jamais fumé\". Il y avait relativement peu de femmes dans le sondage initial ayant fumé et ayant arrêté depuis (162) et très peu pour lesquelles l'information n'était pas disponible (18). La survie à 20 ans a été déterminée pour l'ensemble des femmes du premier sondage.\n", + "
\n", + "
\n", + "Lien du jeu de données : [jeu de données](https://gitlab.inria.fr/learninglab/mooc-rr/mooc-rr-ressources/blob/master/module3/Practical_session/Subject6_smoking.csv)" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
SmokerStatusAge
0YesAlive21.0
1YesAlive19.3
2NoDead57.5
3NoAlive47.1
4YesAlive81.4
\n", + "
" + ], + "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 = pd.read_csv(\"Subject6_smoking.csv\")\n", + "data.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Inspection des données" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Smoker False\n", + "Status False\n", + "Age False\n", + "dtype: bool" + ] + }, + "execution_count": 3, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.isna().any()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Comme nous le voyons il n'y a pas de données manquantes dans ce jeu de données" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Première tache\n", + "Représentez dans un tableau le nombre total de femmes vivantes et décédées sur la période en fonction de leur habitude de tabagisme. Calculez dans chaque groupe (fumeuses / non fumeuses) le taux de mortalité (le rapport entre le nombre de femmes décédées dans un groupe et le nombre total de femmes dans ce groupe)" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
SmokerStatusNumber_of_cases
0NoAlive502
1NoDead230
2YesAlive443
3YesDead139
\n", + "
" + ], + "text/plain": [ + " Smoker Status Number_of_cases\n", + "0 No Alive 502\n", + "1 No Dead 230\n", + "2 Yes Alive 443\n", + "3 Yes Dead 139" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "smoker_agg = data.groupby(['Smoker', 'Status']).agg('count')\n", + "smoker_agg.columns = ['Number_of_cases']\n", + "smoker_agg = smoker_agg.reset_index()\n", + "smoker_agg" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Le taux de mortalité chez les fumeuses est de : 23.88 %\n" + ] + } + ], + "source": [ + "# Taux de mortalité dans le groupe de fumeuses\n", + "smoker_dead_num = float(smoker_agg[(smoker_agg.Smoker==\"Yes\") & (smoker_agg.Status==\"Dead\")]['Number_of_cases'])\n", + "mortality_smoke = 100*smoker_dead_num / smoker_agg[smoker_agg.Smoker=='Yes']['Number_of_cases'].sum()\n", + "print(\"Le taux de mortalité chez les fumeuses est de : \"+str(round(mortality_smoke,2))+' %')\n" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Le taux de mortalité chez les non fumeuses est de : 31.42 %\n" + ] + } + ], + "source": [ + "# Taux de mortalité dans le groupe des non fumeuses\n", + "nosmoker_dead_num = float(smoker_agg[(smoker_agg.Smoker==\"No\") & (smoker_agg.Status==\"Dead\")]['Number_of_cases'])\n", + "mortality_nosmoke = 100*nosmoker_dead_num / smoker_agg[smoker_agg.Smoker=='No']['Number_of_cases'].sum()\n", + "print(\"Le taux de mortalité chez les non fumeuses est de : \"+str(round(mortality_nosmoke,2))+' %')\n" + ] + }, + { + "cell_type": "raw", + "metadata": {}, + "source": [ + "Ce résultat est surprenant parce qu'on s'entendrait à avoir un plus grand taux de mortalité chez les non fumeuses mais ce n'est pas le cas" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Représentation graphique" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe4AAAHiCAYAAAAu1S8tAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABBlUlEQVR4nO3dd5gV5fnG8e9DE6UqoqEJWFlgC7A0iQgSBJEAAcSCBhVjif7EGAtJNKDRqLFgDcaIgoiioqKJsRCaKFgWQURQEQRZRJrS2y48vz9mODksWw64Z5dZ7s917bVn2jvvlHPu887MmTF3R0RERKKhXGlXQERERBKn4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwy09iZo3MzM2sQtj9ppkNSnDax8zszgOc71Iz+8WBTLsf85hmZpclcx5lgZldZWarzGyzmdUq7fqURWb2KzNbHq7jFqVdHyldCu6ICoNrp5kdnaf/nDBIGx1guReb2XsHWi93P8vdxxRVlpldDux09z8d6Lyk9JlZReAB4Ex3r+ru60q7TslmZm+b2ZklPNv7gGvCdTynhOctBxkFd7R9A5y/p8PMUoEjDrSwPa3mkuDuT7j770pqflFUktvjJzgWqAx8XtoVKQlmVgXIBKaX8KwbcoisYymagjvaxgK/juseBDwTP4KZ1TCzZ8xsjZktM7NbzKxcOOxiM3vfzEaY2TrgBeBxoH14SG59ON7ZYUt+Y3i4bnhBFdpzeNnMUgoo6zAzu8/Mvg0Prz5uZocXUt5vzGyhmW0yswVm1jJucIaZzTOzDWb2gplVjpuup5nNNbP1ZjbTzNLC/ueG9dnzt8PMphW5poNpLw3r8mPY6moY9rdwHa4O19FnZta8gDIam9m74fL8Nzxd8Gw4bM9ph8Fm9i0wxczKhdtsWVj+M2ZWIxy/k5ll5yk/dgrBzIab2YRw3Wwys0/MLD1u3Lpm9nK4b3xjZtfGDWtjZlnh8qwyswfyWZaTgS/DzvVmNsXynDoJx4udcsizz603syVmdmrYf3m4jIPipi1wf7F8juiE8z4xfN0j3Gc2mdkKM7shbrx8949w2M3h+JvM7Esz6xI3iy7A++6+I1y/L4bbZJOZfW5mmXHlpITLvj4c1itu2Ohw278RTvuhmZ2Qzzo+zMw2A+WBT81scd7ljCvvjvB1JzPLNrObwvW50sz6hOvjKzP7wcz+GDdtOTMbamaLzWxduExHxZeVp07x+1iB+4mZtQvX7Xoz+9TMOsUNuzjc9pvCfW9g3mWXQri7/iL4BywFfkHwwZlC8MbOJvhm7kCjcLxngNeAakAj4CtgcDjsYiAX+D+gAnB42O+9PPPqBKQSfNFLA1YBfcJhjcL5VQi7pwGXxZWft6wRwOvAUWGd/gXcVcAyngOsAFoDBpwINIxb/o+AumFZC4Erw2EtgNVA23C9DArHPyxP+dXD6a4oYP7xy9Ib+Dpc1xWAW4CZ4bBuwGygZljPFKBOAWXOIjjsWQn4ObAReDbPunwGqBJuj0vD+R4PVAVeAcbGbZfs/PaL8PVwIAfoD1QEbiA4SlMx3JazgT+HdTkeWAJ0i6vnReHrqkC7ApYn7/bfq7uAfSIXuCTcNncA3wKPAYcBZwKbgKpF7S/kv385cGL4eiVwWvj6SKBlUfsHcAqwHKgbtzwnxJX/OOH+Eq7f7UCPsJy7gA/CYRXD7fbHcP2eES7XKeHw0cA6oA3B/jQOGF/I+z22XAV0jwbuiNsvcsNtWxH4DbAGeC5ch82AbUDjcPwhwAdA/XAd/AN4PsF9LN/9BKgXLl8Pgn2ta9hdm2Df3hi3LuoAzUr7MzVKf6VeAf0d4Ib7X3DfEn5gdAcmhR8CHn7glAd2Ak3jprsCmBa+vhj4Nk+5F5PnwzCfeT8IjAhfNyLB4CYItS3s/UHYHvimgPm8DQwpZPkvjOv+G/B4+Hok8Jc8438JnB7XXQ74NzCykOWMX5Y3Cb/wxE2/leCL0hkEX4jaAeUKKe84gg/UI+L6Pcu+wX183PDJwG/juk8hCOMKJBbcH+Sp80rgNILQyrvt/wA8Hb5+F7gNOLqIfSHv9t+ru4B9YlHcsNRw/GPj+q0DMoraX/LuX2G/+OD+lmB/r55nnAL3D4Ivh6sJ3lsV81neb4EGcev3v3HDmgLbwtenAd/H7w/A88Dw8PVo4Mm4YT2ALwpZz/sb3NuA8mF3tXD8tnHjz+Z/X74XAl3ihtUh8X0s3/0EuJnwC2ae9/MgguBeD/QDDi9s/9Jf/n86VB59Y4ELCD7Enskz7GiCb9zL4votI/g2vMfyomZgZm3NbGp4SHUDcGVY9v6qTXAOfnZ4+Gw98FbYPz8NgMWFlPd93OutBN/4IQjT3++ZRzifBgSt8z3uJPhAu5bENAQeiivvB4JgqefuU4BHCVqNq83sCTOrnk8ZdYEf3H1rXL/81n98v7rsu/0qEJxbTkSsLHffTXBUpm64PHXzrKM/xpU7GDgZ+MLMPjazngnOLxGr4l5vC+uWt19V9n9/yasfQSAuM7PpZtY+7F/g/uHuXwPXEYTyajMbb2Z1IXYNyQZ3j98+effByhacJqgLLA/X+R5533sF7b/FYZ277wpfbwv/57eOIVgfr8ati4XALhLbxwraTxoC5+RZxz8nOBK1BTiX4HNkZXi6oMkBLeUhSsEdce6+jODwZw+Cw6jx1hJ8c24Y1+84gsPPsSLyFpnPbJ4jOFzZwN1rEBwutESql099thEcFqsZ/tVw94I+sJYD+5z3S8By4M64edR09yPc/XkAMzuP4KK+/u6esx9lXpGnzMPdfSaAuz/s7q0IWl0nAzfmU8ZK4Cgzi7+AsEE+48Wvt+/Yd/vlEnwIbyHuYkQzK8++odYgbng5gsOh34XL802e5anm7j3C5Vnk7ucDxwD3ABMsuDCrKFvC//HL+LMEpstPUftL3uXfaz7u/rG79w6XYSLwYjio0P3D3Z9z95/zv9NO94TT9QD+k2DdvwMahOt8j7zvvZ9iK8WzjiFYH2flWR+V3X0FRexjhewnywla3PFlVnH3u8Pp3nb3rgSt+y+Af/6E+h9yFNxlw2DgjPCbbEz4jftF4E4zq2bBxVTXExyeLcgqoL6ZVYrrV42gpbjdzNoQtPATsVdZYevjn8AIMzsGwMzqmVm3AqZ/ErjBzFpZ4MRwGYryT+DK8EiBmVkVCy6wq2bBb2AfIThMuCbB5YDgy8ofzKxZWO8aZnZO+Lp1OK+KBB9024HdeQsIv2RlAcPNrFLYAvxlEfN9HvidBRe1VQX+Crzg7rkEh+crh8tWkeC0yWF5pm9lZn3DVuB1wA6C85kfAZssuBDrcDMrb2bNzax1uEwXmlntcJutD8vaZ5nyWcY1BOF0YVjmpRzYl69E9pdPgWZmlmHBhYnD90wbrt+BZlYj/HK2Ma7+he0fp5jZGWZ2GMF23BY3XQ/gjQSr/yFBuN5kZhXDC7N+CYw/kHWRj7nABeE67k5wmP9APU7wGdEQwMxqm1nvcFih+1gh+8mzwC/NrFtYx8oWXOhW38yONbPeYcDvADaTwL4l/6PgLgPcfbG7ZxUw+P8IwmQJ8B5B6/mpQoqbQvCzk+/NbG3Y77fA7Wa2ieCClxcLmjiBsm4muGjnAzPbCPyX4Lxtfsv1EsEh7ecILuyZSHCRUqHCdfEbgsPXP4bzuzgc3JvgQqX37H9Xlr+ZQJmvErQoxof1ng+cFQ6uThAGPxIcDl0H3FtAUQMJztOuI7gw6wWCD6+CPEVwOuRdgiMr2wm2Ke6+gWDbPEkQllsIDoXHe43gsOSPwEVAX3fPCb/U9SQ4l/wNQev2SaBGOF134HMLrmh+CDjP3beRmN8QHHFYR3Ah1MwEp8tPgfuLu38F3B72W0Swf8e7CFgaTnclwbovav84DLibYH18T9CS/IOZ1SQ4mpLQsrj7ToKgPiss6+/Ar939i/1Z+EIMCctfT7BcE39CWQ8RHFF7J3yPf0BwDUQi+1i++0l4OqE3wemXNQQt8BsJMqccQQPiO4JTTqcDV/2E+h9yzD2/I6MiUhLM7AWCi5KGJaHs4QQXMF1Y3GUfasxsAMGplQGlXRcRtbhFSlB4WP0EC347252gVTKxlKslRVtP8NM0kVKX1OA2syFmNt+Cmw9cF/Y7yswmmdmi8P+RyayDyEHmZwQ/j9oMPAxc5bqF5UHP3d9x91mlXQ8RSOKhcgvuHDWe4AYDOwl+xnElcDnBhU53m9lQ4Eh3vzkplRARESljktniTgE+dPet4RWw04G+BIcGx4TjjAH6JLEOIiIiZUoyg3s+cJqZ1Qp/t9qD4Delx7r7ynCc70n8RhIiIiKHvKQ9fcjdF5rZPcA7BD8hmEtwN574cdzM8j1Wb8FjHy8HqFKlSqsmTXRjHREROTTMnj17rbvne5fAEvs5mJn9leD3f0OATu6+0szqENw3O9/f8e6RmZnpWVkF/UxZRESkbDGz2e6emd+wZF9VvuduR8cRnN/ec+vMQeEogwhuECEiIiIJSNqh8tDLZlaL4H7ZV7v7ejO7G3jRzAYT3GVKNzQQERFJUFKD291Py6ffOoKH0YuIiMh+SnaLW0QiLCcnh+zsbLZv317aVREpkypXrkz9+vWpWLFiwtMouEWkQNnZ2VSrVo1GjRphlsiTXEUkUe7OunXryM7OpnHjxglPp3uVi0iBtm/fTq1atRTaIklgZtSqVWu/j2gpuEWkUAptkeQ5kPeXgltE5CeYOHEiCxYsKNF5jh49mmuuuaZE53moKI3tub90jltEEtZo6BvFWt7Su88u1vISlZubS4UKxfPxN3HiRHr27EnTpk2LpbyD0a5duyhfvnxpVyPpcnNzI7E91eIWkYPW0qVLadKkCQMHDiQlJYX+/fuzdetWACZPnkyLFi1ITU3l0ksvZceOHQA0atSItWvXApCVlUWnTp0AGD58OBdddBEdOnTgoosuYtWqVfzqV78iPT2d9PR0Zs6cCcCzzz5LmzZtyMjI4IorrmDXruBOzVWrVuVPf/oT6enptGvXjlWrVjFz5kxef/11brzxRjIyMli8eDH//Oc/ad26Nenp6fTr1y9W38WLF9OuXTtSU1O55ZZbqFq1amw57733Xlq3bk1aWhrDhg3Ld108/fTTnHzyybRp04b3338/1n/NmjX069eP1q1b07p1672G7TF69Gh69+5Np06dOOmkk7jttttiwwpb3t///vekp6cza9Yshg4dStOmTUlLS+OGG26IbZ8zzjiDtLQ0unTpwrfffgvAxRdfzLXXXsupp57K8ccfz4QJEwCYNm0anTp1on///rHtuufunbNnz+b000+nVatWdOvWjZUrg0daPPzww7H5nnfeefkuW58+fejatSuNGjXi0Ucf5YEHHqBFixa0a9eOH374AYC5c+fSrl070tLS+NWvfsWPP/4IQKdOnbjuuuvIzMzknnvu2Wd7Ll68mO7du9OqVStOO+00vvjiCwBeeuklmjdvTnp6Oh07dsx3myWNux/0f61atXIRKXkLFizYq7vhzf8u1r+ifPPNNw74e++95+7ul1xyid97772+bds2r1+/vn/55Zfu7n7RRRf5iBEjgjo2bOhr1qxxd/ePP/7YTz/9dHd3HzZsmLds2dK3bt3q7u4DBgyITZObm+vr16/3BQsWeM+ePX3nzp3u7n7VVVf5mDFj3N0d8Ndff93d3W+88Ub/y1/+4u7ugwYN8pdeeilW57Vr18Ze/+lPf/KHH37Y3d3PPvtsf+6559zdfeTIkV6lShV3d3/77bf9N7/5je/evdt37drlZ599tk+fPn2v9fDdd995gwYNfPXq1b5jxw4/9dRT/eqrr3Z39/PPP99nzJjh7u7Lli3zJk2a7LMen376af/Zz37ma9eu9a1bt3qzZs38448/LnJ5X3jhhdgynXzyyb579253d//xxx/d3b1nz54+evRod3cfNWqU9+7dO7ZO+vfv77t27fLPP//cTzjhBHd3nzp1qlevXt2XL1/uu3bt8nbt2vmMGTN8586d3r59e1+9erW7u48fP94vueQSd3evU6eOb9++fa/55l22E044wTdu3OirV6/26tWr+8iRI93d/brrrott49TUVJ82bZq7u996660+ZMgQd3c//fTT/aqrroqVl3d7nnHGGf7VV1+5u/sHH3zgnTt3dnf35s2be3Z2doH12h9532fu7kCWF5CJOlQuIge1Bg0a0KFDBwAuvPBCHn74Ybp27Urjxo05+eSTARg0aBCPPfYY1113XaFl9erVi8MPPxyAKVOm8MwzzwBQvnx5atSowdixY5k9ezatW7cGYNu2bRxzzDEAVKpUiZ49ewLQqlUrJk2alO885s+fzy233ML69evZvHkz3bp1A2DWrFlMnDgRgAsuuCDWan3nnXd45513aNGiBQCbN29m0aJFe7XiPvzwQzp16kTt2sEzJ84991y++uorAP773//udU5248aNbN68ea8WPUDXrl2pVasWAH379uW9996jQoUKBS5v+fLl6devHwA1atSgcuXKDB48mJ49e8bWw6xZs3jllVcAuOiii7jpppti8+vTpw/lypWjadOmrFq1Kta/TZs21K9fH4CMjAyWLl1KzZo1mT9/Pl27dgWCQ/N16tQBIC0tjYEDB9KnTx/69OmT7zrv3Lkz1apVo1q1atSoUYNf/vKXAKSmpjJv3jw2bNjA+vXrOf3004FgfznnnHNi05977rn5lrt582Zmzpy517h7jux06NCBiy++mAEDBtC3b998p08WBbeIHNTyXnVb1FW4FSpUYPfu3QD7/MymSpUqhU7r7gwaNIi77rprn2EVK1aMzbt8+fLk5ubmW8bFF1/MxIkTSU9PZ/To0UybNq3Ief7hD3/giiuuKHS8guzevZsPPviAypUrFzpefuuxsOWtXLly7Lx2hQoV+Oijj5g8eTITJkzg0UcfZcqUKYXO77DDDou99riHWcX337Me3Z1mzZoxa9asfcp54403ePfdd/nXv/7FnXfeyWeffbbP9QnxZZYrVy7WXa5cuQK3U7yC9ovdu3dTs2ZN5s6du8+wxx9/nA8//JA33niDVq1aMXv27NgXo2TTOW4ROah9++23sQ/05557jp///OeccsopLF26lK+//hqAsWPHxlpTjRo1Yvbs2QC8/PLLBZbbpUsXRo4cCQQtvA0bNtClSxcmTJjA6tWrAfjhhx9YtmxZofWrVq0amzZtinVv2rSJOnXqkJOTw7hx42L927VrF6vP+PHjY/27devGU089xebNmwFYsWJFbP57tG3blunTp7Nu3TpycnJ46aWXYsPOPPNMHnnkkVh3fiEDMGnSJH744Qe2bdvGxIkT6dChQ8LLu3nzZjZs2ECPHj0YMWIEn376KQCnnnpqbFnGjRvHaaftc5frhJxyyimsWbMmtp1zcnL4/PPP2b17N8uXL6dz587cc889bNiwIbae9keNGjU48sgjmTFjBrD3/pJX/PasXr06jRs3jq1vd48t++LFi2nbti233347tWvXZvny5ftdrwOl4BaRg9opp5zCY489RkpKCj/++CNXXXUVlStX5umnn+acc84hNTWVcuXKceWVVwIwbNgwhgwZQmZmZqFXQj/00ENMnTqV1NRUWrVqxYIFC2jatCl33HEHZ555JmlpaXTt2jV2kVRBzjvvPO69915atGjB4sWL+ctf/kLbtm3p0KEDTZo0iY334IMP8sADD5CWlsbXX39NjRo1gCB4L7jgAtq3b09qair9+/ff64sAQJ06dRg+fDjt27enQ4cOpKSkxIY9/PDDZGVlkZaWRtOmTXn88cfzrWebNm3o168faWlp9OvXj8zMzISXd9OmTfTs2ZO0tDR+/vOf88ADDwDwyCOP8PTTT5OWlsbYsWN56KGHCl1XBalUqRITJkzg5ptvJj09nYyMDGbOnMmuXbu48MILSU1NpUWLFlx77bXUrFnzgOYxZswYbrzxRtLS0pg7dy5//vOf8x0v7/YcN24co0aNIj09nWbNmvHaa8EDLW+88UZSU1Np3rw5p556Kunp6QdUrwNRYs/j/in0PG6R0rFw4cK9QqKkLV26lJ49ezJ//vxSq0Nx2bp1K4cffjhmxvjx43n++edjIZBso0ePJisri0cffbRE5if7J7/3WWHP49Y5bhGREjB79myuueYa3J2aNWvy1FNPlXaVJKLU4haRApV2i1vkULC/LW6d4xYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuEZGfQI/1PDi9/vrr3H333cVa5tKlS3nuueeKtcwDoZ+DiUjihtco5vI2FG95CdJjPfdPFB/r2atXL3r16lWsZe4J7gsuuKBYy91fanGLyEFLj/X8Hz3WM//HerZr147PP/881t2pUyeysrJiRyU2bNhAw4YNY/ev37JlCw0aNCAnJ6fAbVVQ/YcOHcqMGTPIyMhgxIgR7Nq1ixtvvDG27f7xj38AsHLlSjp27EhGRgbNmzeP3Wq12BT02LCD6U+P9RQpHfs8bnBY9eL9K4Ie6xnQYz0LfqznAw884H/+859j6+nkk0+OLfOeddSrVy+fMmVKrOzBgwcXuq0Kq//ZZ58dm+Yf//hHbD/Yvn27t2rVypcsWeL33Xef33HHHe4e7FsbN27cp97x9FhPESlT9FhPPdazsMd6DhgwgDPPPJPbbruNF198kf79++8zzrnnnssLL7xA586dGT9+PL/97W8L3VaF1T/eO++8w7x582It8g0bNrBo0SJat27NpZdeSk5ODn369CEjIyPf6Q+UgltEDmp6rGfhDvXHetarV49atWoxb948XnjhhXwfstKrVy/++Mc/8sMPPzB79mzOOOMMoPBtVVD947k7jzzyyF6Bv8e7777LG2+8wcUXX8z111/Pr3/963zLOBA6xy0iBzU91lOP9SzqsZ7nnnsuf/vb39iwYQNpaWn7DK9atSqtW7dmyJAh9OzZM/aFpKBtVZC827pbt26MHDmSnJwcAL766iu2bNnCsmXLOPbYY/nNb37DZZddxieffHJA66UgCm4ROajpsZ56rGdRj/Xs378/48ePZ8CAAQXO49xzz+XZZ5/l3HPPjfUraFsVJC0tjfLly5Oens6IESO47LLLaNq0KS1btqR58+ZcccUV5ObmMm3aNNLT02nRogUvvPACQ4YMOaD1UhA9ZEREClTaDxnRYz2Lhx7reXDTYz1FRA5CeqynFBe1uEWkQKXd4hY5FOixniIiImWYgltEChWFo3IiUXUg7y8Ft4gUqHLlyqxbt07hLZIE7s66deuK/A1+Xro4TUQKVL9+fbKzs1mzZk1pV0WkTKpcuXLsTnKJUnCLSIEqVqxI48aNS7saIhJHh8pFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEKSGtxm9jsz+9zM5pvZ82ZW2cwam9mHZva1mb1gZpWSWQcREZGyJGnBbWb1gGuBTHdvDpQHzgPuAUa4+4nAj8DgZNVBRESkrEn2ofIKwOFmVgE4AlgJnAFMCIePAfokuQ4iIiJlRtKC291XAPcB3xIE9gZgNrDe3XPD0bKBevlNb2aXm1mWmWXprk0iIiKBZB4qPxLoDTQG6gJVgO6JTu/uT7h7prtn1q5dO0m1FBERiZZkHir/BfCNu69x9xzgFaADUDM8dA5QH1iRxDqIiIiUKckM7m+BdmZ2hJkZ0AVYAEwF+ofjDAJeS2IdREREypRknuP+kOAitE+Az8J5PQHcDFxvZl8DtYBRyaqDiIhIWZPUp4O5+zBgWJ7eS4A2yZyviIhIWaU7p8lB68svvyQjIyP2V716dR588EEAHnnkEZo0aUKzZs246aab8p3+0ksv5ZhjjqF58+Z79X/ppZdo1qwZ5cqVIysra69h8+bNo3379jRr1ozU1FS2b98OwPPPP09qaippaWl0796dtWvXFv8Ci4gkwNy9tOtQpMzMTM/7ASuHll27dlGvXj0+/PBDlixZwp133skbb7zBYYcdxurVqznmmGP2mebdd9+latWq/PrXv2b+/Pmx/gsXLqRcuXJcccUV3HfffWRmZgKQm5tLy5YtGTt2LOnp6axbt46aNWvi7tStW5cFCxZw9NFHc9NNN3HEEUcwfPjwklp8ETnEmNlsd8/Mb5ha3BIJkydP5oQTTqBhw4aMHDmSoUOHcthhhwHkG9oAHTt25Kijjtqnf0pKCqeccso+/d955x3S0tJIT08HoFatWpQvXx53x93ZsmUL7s7GjRupW7duMS6diEjiFNwSCePHj+f8888H4KuvvmLGjBm0bduW008/nY8//rhY5vHVV19hZnTr1o2WLVvyt7/9DYCKFSsycuRIUlNTYy3vwYN1p14RKR0Kbjno7dy5k9dff51zzjkHCA5p//DDD3zwwQfce++9DBgwgOI45ZObm8t7773HuHHjeO+993j11VeZPHkyOTk5jBw5kjlz5vDdd9+RlpbGXXfd9ZPnJyJyIBTcctB78803admyJcceeywA9evXp2/fvpgZbdq0oVy5csVysVj9+vXp2LEjRx99NEcccQQ9evTgk08+Ye7cuQCccMIJmBkDBgxg5syZP3l+IiIHQsEtB73nn38+dpgcoE+fPkydOhUIDm/v3LmTo48++ifPp1u3bnz22Wds3bqV3Nxcpk+fTtOmTalXrx4LFixgzz3zJ02aREpKyk+en4jIgVBwy0Fty5YtTJo0ib59+8b6XXrppSxZsoTmzZtz3nnnMWbMGMyM7777jh49esTGO//882nfvj1ffvkl9evXZ9So4F4/r776KvXr12fWrFmcffbZdOvWDYAjjzyS66+/ntatW5ORkUHLli05++yzqVu3LsOGDaNjx46kpaUxd+5c/vjHP5bsihARCennYCIiIgcZ/RxMRESkjFBwi4iIRIiCW0REJEIU3CIiIhGS1KeDySFieI3SrkHZMXxDaddARA5yanGLiIhEiIJbREQkQhTcIiIiEaLgFhERiRAFt4iISIQouEVERCJEwS0iIhIhCm4REZEIUXCLiIhEiIJbREQkQhTcIiIiEaLgFhERiRAFt4iISIQouEVERCJEwS0iIhIhCm4REZEIUXCLiIhEiIJbREQkQhTcIiIiEaLgFhERiRAFt4iISIQouEVERCJEwS0iIhIhCm4RESlW69evp3///jRp0oSUlBRmzZrF3LlzadeuHRkZGWRmZvLRRx/lO2358uXJyMggIyODXr16xfqfdtppsf5169alT58+APz444/86le/Ii0tjTZt2jB//nwAli9fTufOnWnatCnNmjXjoYceSvpyl5QKpV0BEREpW4YMGUL37t2ZMGECO3fuZOvWrQwYMIBhw4Zx1lln8Z///IebbrqJadOm7TPt4Ycfzty5c/fpP2PGjNjrfv360bt3bwD++te/kpGRwauvvsoXX3zB1VdfzeTJk6lQoQL3338/LVu2ZNOmTbRq1YquXbvStGnTZC12iUlai9vMTjGzuXF/G83sOjM7yswmmdmi8P+RyaqDiIiUrA0bNvDuu+8yePBgACpVqkTNmjUxMzZu3Bgbp27dugdU/saNG5kyZUqsxb1gwQLOOOMMAJo0acLSpUtZtWoVderUoWXLlgBUq1aNlJQUVqxY8ROX7uCQtOB29y/dPcPdM4BWwFbgVWAoMNndTwImh90iIlIGfPPNN9SuXZtLLrmEFi1acNlll7FlyxYefPBBbrzxRho0aMANN9zAXXfdle/027dvJzMzk3bt2jFx4sR9hk+cOJEuXbpQvXp1ANLT03nllVcA+Oijj1i2bBnZ2dl7TbN06VLmzJlD27Zti3dhS0lJnePuAix292VAb2BM2H8M0KeE6iAiIkmWm5vLJ598wlVXXcWcOXOoUqUKd999NyNHjmTEiBEsX76cESNGxFrkeS1btoysrCyee+45rrvuOhYvXrzX8Oeff57zzz8/1j106FDWr19PRkYGjzzyCC1atKB8+fKx4Zs3b6Zfv348+OCDsbCPOnP35M/E7CngE3d/1MzWu3vNsL8BP+7pLkhmZqZnZWUlvZ5ygIbXKO0alB3DN5R2DUR+ku+//5527dqxdOlSIDg3fffdd/Pee++xfv16zAx3p0aNGrFD5wW5+OKL6dmzJ/379wdg7dq1nHLKKaxYsYLKlSvvM76707hxY+bNm0f16tXJycmhZ8+edOvWjeuvv77YlzWZzGy2u2fmNyzpLW4zqwT0Al7KO8yDbw35fnMws8vNLMvMstasWZPkWoqISHH42c9+RoMGDfjyyy8BmDx5Mk2bNqVu3bpMnz4dgClTpnDSSSftM+2PP/7Ijh07gCCk33///b0uJpswYQI9e/bcK7TXr1/Pzp07AXjyySfp2LEj1atXx90ZPHgwKSkpkQvtopTEVeVnEbS2V4Xdq8ysjruvNLM6wOr8JnL3J4AnIGhxl0A9RUSkGDzyyCMMHDiQnTt3cvzxx/P000/Tu3dvhgwZQm5uLpUrV+aJJ54AICsri8cff5wnn3yShQsXcsUVV1CuXDl2797N0KFD9wru8ePHM3To3pdFLVy4kEGDBmFmNGvWjFGjRgHw/vvvM3bsWFJTU8nIyACCK9B79OhRMishiZJ+qNzMxgNvu/vTYfe9wDp3v9vMhgJHuftNhZWhQ+UHOR0qLz46VC4ilOKhcjOrAnQFXonrfTfQ1cwWAb8Iu0VERCQBST1U7u5bgFp5+q0juMpcRERE9pNueSoiIhIhuuWpiMihSNemFK8SvD5FLW4REZEIUXCLiIhEiIJbREQkQhTcIiIiEaLgFhERiRAFt4iISIQouEVERCJEwS0iIhIhCm4REZEIUXCLiIhEiIJbREQkQhTcIiIiEaLgFhERiRAFt4iISIQouEVERCJEwS0iIhIhCm4REZEIUXCLiIhEiIJbREQkQhTcIiIiEaLgFhERiRAFt4iISIQouEVERCJEwS0iIhIhCm4REZEIUXCLiIhEiIJbREQkQhTcIiIiEaLgFhERiRAFt4iISIQouEVERCJEwS0iIhIhCm4REZEIUXCLiIhEiIJbREQkQhTcIiIiEaLgFhERiZCkBreZ1TSzCWb2hZktNLP2ZnaUmU0ys0Xh/yOTWQcREZGyJNkt7oeAt9y9CZAOLASGApPd/SRgctgtIiIiCUhacJtZDaAjMArA3Xe6+3qgNzAmHG0M0CdZdRARESlrktnibgysAZ42szlm9qSZVQGOdfeV4TjfA8cmsQ4iIiJlSjKDuwLQEhjp7i2ALeQ5LO7uDnh+E5vZ5WaWZWZZa9asSWI1RUREoiOZwZ0NZLv7h2H3BIIgX2VmdQDC/6vzm9jdn3D3THfPrF27dhKrKSIiEh1JC253/x5YbmanhL26AAuA14FBYb9BwGvJqoOIiEhZUyHJ5f8fMM7MKgFLgEsIviy8aGaDgWXAgCTXQUREpMxIanC7+1wgM59BXZI5XxERkbJKd04TERGJEAW3iIhIhCi4RUREIkTBLSIiEiEKbhERkQhRcIuIiESIgltERCRCFNwiIiIRouAWERGJEAW3iIhIhCi4RUREIkTBLSIiEiEKbhERkQhRcIuIiESIgltERCRCFNwiIiIRouAWERGJEAW3iIhIhCi4RUREIkTBLSIiEiEKbhERkQhRcIuIiESIgltERCRCFNwiIiIRouAWERGJEAW3iIhIhCi4RUREIkTBLSIiEiEKbhERkQhRcIuIiESIgltERCRCFNwiIiIRouAWERGJEAW3iIhIhCi4RUREIkTBLSIiEiEKbhERkQhRcIuIiESIgltERCRCEgpuMzvWzEaZ2Zthd1MzG5zAdEvN7DMzm2tmWWG/o8xskpktCv8f+dMWQURE5NCRaIt7NPA2UDfs/gq4LsFpO7t7hrtnht1DgcnufhIwOewWERGRBCQa3Ee7+4vAbgB3zwV2HeA8ewNjwtdjgD4HWI6IiMghJ9Hg3mJmtQAHMLN2wIYEpnPgHTObbWaXh/2OdfeV4evvgWP3p8IiIiKHsgoJjnc98Dpwgpm9D9QG+icw3c/dfYWZHQNMMrMv4ge6u5uZ5zdhGPSXAxx33HEJVlNERKRsSyi43f0TMzsdOAUw4Et3z0lguhXh/9Vm9irQBlhlZnXcfaWZ1QFWFzDtE8ATAJmZmfmGu4iIyKEm0avK+wK9CIL7ZOCXZtYlbEkXNE0VM6u25zVwJjCfoOU+KBxtEPDagVdfRETk0JLoofLBQHtgatjdCZgNNDaz2919bD7THAu8amZ75vOcu79lZh8DL4Y/J1sGDPgJ9RcRETmkJBrcFYAUd18Fwe+6gWeAtsC7wD7B7e5LgPR8+q8DuhxohUVERA5liV5V3mBPaIdWh/1+AIo81y0iIiLFI9EW9zQz+zfwUtjdL+xXBVifjIqJiIjIvhIN7qsJwrpD2P0M8LK7O9A5GRUTERGRfSX6czAHJoR/IiIiUkoS/TlYOzP72Mw2m9lOM9tlZhuTXTkRERHZW6IXpz0KnA8sAg4HLgMeS1alREREJH8JP4/b3b8Gyrv7Lnd/GuievGqJiIhIfhK9OG2rmVUC5prZ34CV7Efoi4iISPFINHwvCse9BtgCNAD6JqtSIiIikr9Eg7uPu293943ufpu7Xw/0TGbFREREZF+JBvegfPpdXIz1EBERkQQUeo7bzM4HLiB4mMjrcYOqAT8ks2IiIiKyr6IuTptJcCHa0cD9cf03AfOSVSkRERHJX6HB7e7LCB692b5kqiMiIiKFSfTOaX3NbJGZbTCzjWa2SXdOExERKXmJ/o77b8Av3X1hMisjIiIihUv0qvJVCm0REZHSl2iLO8vMXgAmAjv29HT3V5JRKREREclfosFdHdgKnBnXzwEFt4iISAlK9HnclyS7IiIiIlK0RK8qP9nMJpvZ/LA7zcxuSW7VREREJK9EL077J/AHIAfA3ecB5yWrUiIiIpK/RIP7CHf/KE+/3OKujIiIiBQu0eBea2YnEFyQhpn1J7gVqoiIiJSgRK8qvxp4AmhiZiuAb4ALk1YrERERyVeiV5UvAX5hZlWAcu6+KbnVEhERkfwkelX5X82sprtvcfdNZnakmd2R7MqJiIjI3hI9x32Wu6/f0+HuPwI9klIjERERKVCiwV3ezA7b02FmhwOHFTK+iIiIJEGiF6eNAyab2dNh9yXAmORUSURERApSZHCbmQHPAZ8Cvwh7/8Xd305mxURERGRfRQa3u7uZ/cfdU4G3SqBOIiIiUoBEz3F/Ymatk1oTERERKVKi57jbAgPNbBmwBTCCxnha0momIiIi+0g0uLsltRYiIiKSkESD25NaCxEREUlIosH9BkF4G1AZaAx8CTRLUr1EREQkH4neqzw1vtvMWgK/TUqNREREpECJXlW+F3f/hOCCNRERESlBCbW4zez6uM5yQEvgu6TUSERERAqUaIu7WtzfYQTnvHsnMqGZlTezOWb277C7sZl9aGZfm9kLZlbpQCouIiJyKEr0HPdtAGZWNezevB/zGAIsBKqH3fcAI9x9vJk9DgwGRu5HeSIiIoesRJ/H3dzM5gCfA5+b2Wwza57AdPWBs4Enw24DzgAmhKOMAfocQL1FREQOSYkeKn8CuN7dG7p7Q+D3Yb+iPAjcBOwOu2sB6909N+zOBuolXl0REZFDW6LBXcXdp+7pcPdpQJXCJjCznsBqd599IBUzs8vNLMvMstasWXMgRYiIiJQ5id6AZYmZ3QqMDbsvBJYUMU0HoJeZ9SC4aUt14CGgpplVCFvd9YEV+U3s7k8QtuozMzN15zYRERESb3FfCtQGXgFeBo4O+xXI3f/g7vXdvRFwHjDF3QcCU4H+4WiDgNcOoN4iIiKHpEJb3GZWGbgSOBH4DPi9u+f8xHneDIw3szuAOcCon1ieiIjIIaOoQ+VjgBxgBnAWkAJct78zCc+JTwtfLwHa7G8ZIiIiUnRwN91zn3IzGwV8lPwqiYiISEGKOscdOywe9xMuERERKSVFtbjTzWxj+NqAw8NuA9zdqxc8qYiIiBS3QoPb3cuXVEVERESkaAf0WE8REREpHQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0QOedu3b6dNmzakp6fTrFkzhg0bBsCjjz7KiSeeiJmxdu3aQsvYuHEj9evX55prrgFg69atnH322TRp0oRmzZoxdOjQ2Lg7duzg3HPP5cQTT6Rt27YsXboUgJycHAYNGkRqaiopKSncddddyVlgiTQFt4gc8g477DCmTJnCp59+yty5c3nrrbf44IMP6NChA//9739p2LBhkWXceuutdOzYca9+N9xwA1988QVz5szh/fff58033wRg1KhRHHnkkXz99df87ne/4+abbwbgpZdeYseOHXz22WfMnj2bf/zjH7FQF9lDwS0ihzwzo2rVqkDQ6s3JycHMaNGiBY0aNSpy+tmzZ7Nq1SrOPPPMWL8jjjiCzp07A1CpUiVatmxJdnY2AK+99hqDBg0CoH///kyePBl3x8zYsmULubm5bNu2jUqVKlG9evViXlqJOgW3iAiwa9cuMjIyOOaYY+jatStt27ZNaLrdu3fz+9//nvvuu6/AcdavX8+//vUvunTpAsCKFSto0KABABUqVKBGjRqsW7eO/v37U6VKFerUqcNxxx3HDTfcwFFHHfXTF07KFAW3iAhQvnx55s6dS3Z2Nh999BHz589PaLq///3v9OjRg/r16+c7PDc3l/PPP59rr72W448/vtCyPvroI8qXL893333HN998w/3338+SJUv2e1mkbKtQ2hUQETmY1KxZk86dO/PWW2/RvHnzIsefNWsWM2bM4O9//zubN29m586dVK1albvvvhuAyy+/nJNOOonrrrsuNk29evVYvnw59evXJzc3lw0bNlCrVi2ee+45unfvTsWKFTnmmGPo0KEDWVlZRQa+HFrU4haRQ96aNWtYv349ANu2bWPSpEk0adIkoWnHjRvHt99+y9KlS7nvvvv49a9/HQvtW265hQ0bNvDggw/uNU2vXr0YM2YMABMmTOCMM87AzDjuuOOYMmUKAFu2bOGDDz5IuB5y6FBwi8ghb+XKlXTu3Jm0tDRat25N165d6dmzJw8//DD169cnOzubtLQ0LrvsMgCysrJirwuSnZ3NnXfeyYIFC2jZsiUZGRk8+eSTAAwePJh169Zx4okn8sADD8SC/uqrr2bz5s00a9aM1q1bc8kll5CWlpbchZfIMXcv7ToUKTMz07Oyskq7GlKQ4TVKuwZlx/ANpV0DOVTofVu8ivm9a2az3T0zv2FqcYuIiESIgltERCRCFNwiIiIRouAWERGJEP2OW0Qio9HQN0q7CmXG0sqlXQM5UGpxi4iIRIiCW0REJEKSFtxmVtnMPjKzT83sczO7Lezf2Mw+NLOvzewFM6uUrDqIiIiUNclsce8AznD3dCAD6G5m7YB7gBHufiLwIzA4iXUQEREpU5IW3B7YHHZWDP8cOAOYEPYfA/RJVh1ERETKmqSe4zaz8mY2F1gNTAIWA+vdPTccJRuol8w6iIiIlCVJDW533+XuGUB9oA2Q8GNuzOxyM8sys6w1a9Ykq4oiIiKRUiJXlbv7emAq0B6oaWZ7fj9eH1hRwDRPuHumu2fWrl27JKopIiJy0EvmVeW1zaxm+PpwoCuwkCDA+4ejDQJeS1YdREREyppk3jmtDjDGzMoTfEF40d3/bWYLgPFmdgcwBxiVxDqIiIiUKUkLbnefB7TIp/8SgvPdIiIisp905zQREZEIUXCLiIhEiIJbREQkQhTcIiIiEaLgFhERiRAFt4iISIQouEVERCJEwS0iIhIhCm4REZEIUXCLiIhEiIJbREQkQhTcIiIiEaLgLkbLly+nc+fONG3alGbNmvHQQw8BcOutt5KWlkZGRgZnnnkm3333Xb7T33TTTTRr1oyUlBSuvfZa3B2A559/ntTUVNLS0ujevTtr164ttNxx48aRlpZGamoqp556Kp9++mkJLL2IiJQEBXcxqlChAvfffz8LFizggw8+4LHHHmPBggXceOONzJs3j7lz59KzZ09uv/32faadOXMm77//PvPmzWP+/Pl8/PHHTJ8+ndzcXIYMGcLUqVOZN28eaWlpPProowAFltu4cWOmT5/OZ599xq233srll19eoutBRESSR8FdjOrUqUPLli0BqFatGikpKaxYsYLq1avHxtmyZQtmts+0Zsb27dvZuXMnO3bsICcnh2OPPRZ3x93ZsmUL7s7GjRupW7cuQIHlnnrqqRx55JEAtGvXjuzs7KQts4iIlKykPY/7ULd06VLmzJlD27ZtAfjTn/7EM888Q40aNZg6deo+47dv357OnTtTp04d3J1rrrmGlJQUAEaOHElqaipVqlThpJNO4rHHHotNV1S5o0aN4qyzzkrSUoqISElTizsJNm/eTL9+/XjwwQdjreI777yT5cuXM3DgwNih7nhff/01CxcuJDs7mxUrVjBlyhRmzJhBTk4OI0eOZM6cOXz33XekpaVx1113xaYrrNypU6cyatQo7rnnnuQusIiIlBgFdzHLycmhX79+DBw4kL59++4zfODAgbz88sv79H/11Vdp164dVatWpWrVqpx11lnMmjWLuXPnAnDCCSdgZgwYMICZM2cWWe68efO47LLLeO2116hVq1bxLaCIiJQqBXcxcncGDx5MSkoK119/faz/okWLYq9fe+01mjRpss+0xx13XOxitJycHKZPn05KSgr16tVjwYIFrFmzBoBJkybFDqEXVO63335L3759GTt2LCeffHJSllVEREqHznEXo/fff5+xY8eSmppKRkYGAH/9618ZNWoUX375JeXKlaNhw4Y8/vjjAGRlZfH444/z5JNP0r9/f6ZMmUJqaipmRvfu3fnlL38JwLBhw+jYsSMVK1akYcOGjB49GoChQ4fmW+7tt9/OunXr+O1vfwsEV7tnZWWV7MoQEZGksD2/FT6YZWZmuoLnIDa8RmnXoOwYvqG0a3BQazT0jdKuQpmxtPIFpV2FsqWY37tmNtvdM/MbpkPlIiIiEaLgFhERiRAFt4iISIQckhen6TxZ8VpaubRrICJy6FCLW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIUkLbjNrYGZTzWyBmX1uZkPC/keZ2SQzWxT+PzJZdRARESlrktnizgV+7+5NgXbA1WbWFBgKTHb3k4DJYbeIiIgkIGnB7e4r3f2T8PUmYCFQD+gNjAlHGwP0SVYdREREypoSOcdtZo2AFsCHwLHuvjIc9D1wbEnUQUREpCxIenCbWVXgZeA6d98YP8zdHfACprvczLLMLGvNmjXJrqaIiEgkJDW4zawiQWiPc/dXwt6rzKxOOLwOsDq/ad39CXfPdPfM2rVrJ7OaIiIikZHMq8oNGAUsdPcH4ga9DgwKXw8CXktWHURERMqaCkksuwNwEfCZmc0N+/0RuBt40cwGA8uAAUmsg4iISJmStOB29/cAK2Bwl2TNV0REpCzTndNEREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEJC24zewpM1ttZvPj+h1lZpPMbFH4/8hkzV9ERKQsSmaLezTQPU+/ocBkdz8JmBx2i4iISIKSFtzu/i7wQ57evYEx4esxQJ9kzV9ERKQsKulz3Me6+8rw9ffAsQWNaGaXm1mWmWWtWbOmZGonIiJykCu1i9Pc3QEvZPgT7p7p7pm1a9cuwZqJiIgcvEo6uFeZWR2A8P/qEp6/iIhIpJV0cL8ODApfDwJeK+H5i4iIRFoyfw72PDALOMXMss1sMHA30NXMFgG/CLtFREQkQRWSVbC7n1/AoC7JmqeIiEhZpzuniYiIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIQpuERGRCFFwi4iIRIiCW0REJEIU3CIiIhGi4BYREYkQBbeIiEiEKLhFREQiRMEtIiISIaUS3GbW3cy+NLOvzWxoadRBREQkiko8uM2sPPAYcBbQFDjfzJqWdD1ERESiqDRa3G2Ar919ibvvBMYDvUuhHiIiIpFTGsFdD1ge150d9hMREZEiVCjtChTEzC4HLg87N5vZl6VZHymYwdHA2tKuR5lwm5V2DeQQofdtMSv+927DggaURnCvABrEddcP++3F3Z8AniipSsmBM7Msd88s7XqISOL0vo2u0jhU/jFwkpk1NrNKwHnA66VQDxERkcgp8Ra3u+ea2TXA20B54Cl3/7yk6yEiIhJFpXKO293/A/ynNOYtSaFTGiLRo/dtRJm7l3YdREREJEG65amIiEiEKLjLODPbZWZz4/4alXadRA5lZuZmdn9c9w1mNryYyr7WzBaa2bjiKE8OTgft77il2Gxz94zSroSIxOwA+prZXe5e3L+j/i3wC3fPLuZy5SCiFvchyMyWmtnR4etMM5sWvh5uZmPMbIaZLTOzvmb2NzP7zMzeMrOK4XitzGy6mc02s7fNrE7Yf5qZZYavjzazpeHrZmb2Udjin2dmJ4X9L4zr/w8zKx/+jTaz+eF8f1fya0gkqXIJLgzbZ982s0ZmNiV8n0w2s+PC/qPN7GEzm2lmS8ysfz7TPg4cD7xpZr8L3883xA2fH5bfyMy+CMv8yszGmdkvzOx9M1tkZm3C8auY2VPhe3SOmfUO+19sZo/GlftvM+tU0HvXzE4IPz9mh58tTcL+54Tjfmpm7xbnCi7rFNxl3+Fxh8lfTWD8E4AzgF7As8BUd08FtgFnh+H9CNDf3VsBTwF3FlHmlcBDYcs/E8g2sxTgXKBD2H8XMBDIAOq5e/Nwvk/v19KKRMNjwEAzq5Gn/yPAGHdPA8YBD8cNqwP8HOgJ3J23QHe/EvgO6OzuI4qY/4nA/UCT8O+CsOwbgD+G4/wJmOLubYDOwL1mVqWQMjPI/737BPB/4efFDcDfw/5/Brq5ezrB540kSIfKy779PVT+prvnmNlnBL+zfyvs/xnQCDgFaA5MMjPCcVYWUeYs4E9mVh94xd0XmVkXoBXwcVjO4cBq4F/A8Wb2CPAG8M5+1F0kEtx9o5k9A1xL8KV4j/ZA3/D1WOBvccMmuvtuYIGZHfsTq/CNu38GYGafA5Pd3cP3faNwnDOBXnGt9srAcYWUuYQ8710zqwqcCrwUvs8BDgv/vw+MNrMXgVd+4vIcUhTch6Zc/ne0pXKeYTsA3H23meX4/34vuJtgfzHgc3dvn2i57v6cmX0InA38x8yuCMsZ4+5/yFuImaUD3Qha6gOAS/d/EUUOeg8Cn5D4UaUdca8TuTF2/PsR9n6vx5e1O657z/t8zzz6uftez4kws1b5levuP+bz3r0OWJ9f48HdrzSztgSfC7PNrJW7r0tguQ55OlR+aFpK0NoF6Lef034J1Daz9gBmVtHMmuVTbuwcnJkdDyxx94eB14A0YDLQ38yOCcc5yswahufey7n7y8AtQMv9rJ9IJLj7D8CLwOC43jMJbgMNwamjGT9hFksJ3z9m1hJovJ/Tvw38n4VNZTNrEVduhpmVM7MGBI9qJr/3rrtvBL4xs3PCcSwMd8zsBHf/0N3/DKxh72dYSCEU3Iem24CHzCyL4NxywsJnqPcH7jGzT4G5BIfCAO4DrjKzOQRPHtpjADDfzOYSHGZ/xt0XELy53zGzecAkgnN49YBp4bjPAvu0yEXKkPvZ+73yf8Al4XviImDITyj7ZeCo8FD4NcBX+zn9X4CKwLywjL+E/d8HvgEWEJyD/yTsX9B7dyAwOPy8+BzoHfa/N7yIbT7BF5ZP97N+hyzdOU1ERCRC1OIWERGJEAW3iIhIhCi4RUREIkTBLSIiEiEKbhERkQhRcIuIiESIgltERCRCFNwiIiIR8v/GMOhupPm6CwAAAABJRU5ErkJggg==\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "\n", + "labels = ['Fumeuses', 'Non fumeuses']\n", + "d1 = [mortality_smoke, mortality_nosmoke]\n", + "d2 = [100-mortality_smoke, 100-mortality_nosmoke]\n", + "x = np.arange(len(labels)) # the label locations\n", + "width = 0.35 # the width of the bars\n", + "fig, ax = plt.subplots(figsize = (8,8))\n", + "rects1 = ax.bar(x - width/2, d1, width, label='pourcentage de personnes mortes')\n", + "rects2 = ax.bar(x + width/2, d2, width, label = \"pourcentage de personnes vivantes \")\n", + "ax.set_ylabel('Pourcentage')\n", + "ax.set_title('Mortalité chez les groupes fumeuses/non fumeuses')\n", + "ax.set_xticks(x)\n", + "plt.yticks(np.arange(0, 100, 10))\n", + "ax.set_xticklabels(labels)\n", + "ax.legend()\n", + "\n", + "ax.bar_label(rects1, padding=3)\n", + "ax.bar_label(rects2, padding=3)\n", + "\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 39, + "metadata": {}, + "outputs": [], + "source": [ + "def ratio_mortality(df):\n", + " # Taux de mortalité dans le groupe des non fumeuses\n", + " nosmoker_dead_num = float(df[(df.Smoker==\"No\") & (df.Status==\"Dead\")]['Number_of_cases'])\n", + " mortality_nosmoke = 100*nosmoker_dead_num / df[df.Smoker=='No']['Number_of_cases'].sum()\n", + " mortality_nosmoke = round(mortality_nosmoke,2)\n", + " \n", + " # Taux de mortalité dans le groupe des fumeuses\n", + " smoker_dead_num = float(df[(df.Smoker==\"Yes\") & (df.Status==\"Dead\")]['Number_of_cases'])\n", + " mortality_smoke = 100*smoker_dead_num / df[df.Smoker=='Yes']['Number_of_cases'].sum()\n", + " mortality_smoke = round(mortality_smoke,2)\n", + " return [mortality_nosmoke, mortality_smoke]" + ] + }, + { + "cell_type": "code", + "execution_count": 40, + "metadata": {}, + "outputs": [], + "source": [ + "def class_separation(df):\n", + " smoker_agg = df.groupby(['Smoker', 'Status']).agg('count')\n", + " smoker_agg.columns = ['Number_of_cases']\n", + " smoker_agg = smoker_agg.reset_index()\n", + " return smoker_agg\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Seconde tache\n", + "Reprenez la question 1 (effectifs et taux de mortalité) en rajoutant une nouvelle catégorie liée à la classe d'âge. On considérera par exemple les classes suivantes : 18-34 ans, 34-54 ans, 55-64 ans, plus de 65 ans. En quoi ce résultat est-il surprenant ? Arrivez-vous à expliquer ce paradoxe ? De même, vous pourrez proposer une représentation graphique de ces données pour étayer vos explications." + ] + }, + { + "cell_type": "code", + "execution_count": 43, + "metadata": {}, + "outputs": [], + "source": [ + "g1 = class_separation(data[(data.Age>=18)&(data.Age<34)])\n", + "g2 = class_separation(data[(data.Age>=34)&(data.Age<=54)])\n", + "g3 = class_separation(data[(data.Age>54)&(data.Age<=64)])\n", + "g4 = class_separation(data[data.Age>64])\n", + "\n", + "#Mortality per category\n", + "m1 = ratio_mortality(g1)\n", + "m2 = ratio_mortality(g2)\n", + "m3 = ratio_mortality(g3)\n", + "m4 = ratio_mortality(g4)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Représentation graphique" + ] + }, + { + "cell_type": "code", + "execution_count": 45, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "\n", + "labels = ['18-34 ans', '34-54 ans', '55-64 ans', '>55 ans']\n", + "d1 = [m1[0], m2[0], m3[0], m4[0]]\n", + "d2 = [m1[1], m2[1], m3[1], m4[1]]\n", + "x = np.arange(len(labels)) # the label locations\n", + "width = 0.35 # the width of the bars\n", + "fig, ax = plt.subplots(figsize = (8,8))\n", + "rects1 = ax.bar(x - width/2, d1 , width, label='Mortalité chez les non fumeuses')\n", + "rects2 = ax.bar(x + width/2, d2 , width, label = \"Mortalité chez les fumeuses\")\n", + "ax.set_ylabel('Pourcentage')\n", + "ax.set_title(\"Mortalité chez les groupes fumeuses/non fumeuses pour différentes catégories d'age'\")\n", + "ax.set_xticks(x)\n", + "plt.yticks(np.arange(0, 100, 10))\n", + "ax.set_xticklabels(labels)\n", + "ax.legend()\n", + "\n", + "ax.bar_label(rects1, padding=3)\n", + "ax.bar_label(rects2, padding=3)\n", + "\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "raw", + "metadata": {}, + "source": [ + "On constate que :\n", + "- Le taux de mortalité des fumeuses est casiment le meme que chez les non fumeuses entre 18 et 34 ans\n", + "- Pour les femmes entre 35 et 64 ans il y'a plus de morts chez les fumeuses que chez les fumeuses\n", + "- Pour les femmes au delà de 65 ans, le taux de mortalité est plus ou moins le meme et il est très grand, ceci semble indiquer que pour les femmes de cet age le fait de fumer ou non n'influence pas forcément ce taux de mortalit\"\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Troisième tache\n", + "Afin d'éviter un biais induit par des regroupements en tranches d'âges arbitraires et non régulières, il est envisageable d'essayer de réaliser une régression logistique. Si on introduit une variable Death valant 1 ou 0 pour indiquer si l'individu est décédé durant la période de 20 ans, on peut étudier le modèle Death ~ Age pour étudier la probabilité de décès en fonction de l'âge selon que l'on considère le groupe des fumeuses ou des non fumeuses. Ces régressions vous permettent-elles de conclure sur la nocivité du tabagisme ? Vous pourrez proposer une représentation graphique de ces régressions (en n'omettant pas les régions de confiance)." + ] + }, + { + "cell_type": "code", + "execution_count": 65, + "metadata": {}, + "outputs": [], + "source": [ + "new_data = data[data.Age<=30]\n", + "new_data = new_data.replace(\"Alive\", 0)\n", + "new_data = new_data.replace(\"Dead\", 1)" + ] + }, + { + "cell_type": "code", + "execution_count": 69, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
SmokerStatusAge
0Yes021.0
1Yes019.3
6No023.8
8Yes024.8
10Yes030.0
\n", + "
" + ], + "text/plain": [ + " Smoker Status Age\n", + "0 Yes 0 21.0\n", + "1 Yes 0 19.3\n", + "6 No 0 23.8\n", + "8 Yes 0 24.8\n", + "10 Yes 0 30.0" + ] + }, + "execution_count": 69, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "new_data.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 70, + "metadata": {}, + "outputs": [], + "source": [ + "smoke_group = new_data[new_data.Smoker==\"Yes\"]\n", + "nosmoke_group = new_data[new_data.Smoker==\"No\"]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Premier cas : le groupe des fumeuses" + ] + }, + { + "cell_type": "code", + "execution_count": 86, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Optimization terminated successfully.\n", + " Current function value: 0.078375\n", + " Iterations 9\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "
Logit Regression Results
Dep. Variable: Status No. Observations: 131
Model: Logit Df Residuals: 129
Method: MLE Df Model: 1
Date: Fri, 07 Jan 2022 Pseudo R-squ.: 0.007889
Time: 01:43:22 Log-Likelihood: -10.267
converged: True LL-Null: -10.349
LLR p-value: 0.6862
\n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "
coef std err z P>|z| [0.025 0.975]
Intercept -6.1874 5.247 -1.179 0.238 -16.471 4.096
Age 0.0810 0.204 0.396 0.692 -0.320 0.482
" + ], + "text/plain": [ + "\n", + "\"\"\"\n", + " Logit Regression Results \n", + "==============================================================================\n", + "Dep. Variable: Status No. Observations: 131\n", + "Model: Logit Df Residuals: 129\n", + "Method: MLE Df Model: 1\n", + "Date: Fri, 07 Jan 2022 Pseudo R-squ.: 0.007889\n", + "Time: 01:43:22 Log-Likelihood: -10.267\n", + "converged: True LL-Null: -10.349\n", + " LLR p-value: 0.6862\n", + "==============================================================================\n", + " coef std err z P>|z| [0.025 0.975]\n", + "------------------------------------------------------------------------------\n", + "Intercept -6.1874 5.247 -1.179 0.238 -16.471 4.096\n", + "Age 0.0810 0.204 0.396 0.692 -0.320 0.482\n", + "==============================================================================\n", + "\"\"\"" + ] + }, + "execution_count": 86, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "log_reg1 = smf.logit(formula = \"Status ~ Age\", data = smoke_group).fit()\n", + "log_reg1.summary()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Second cas : le groupe des non fumeuses" + ] + }, + { + "cell_type": "code", + "execution_count": 87, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Optimization terminated successfully.\n", + " Current function value: 0.138191\n", + " Iterations 8\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "
Logit Regression Results
Dep. Variable: Status No. Observations: 170
Model: Logit Df Residuals: 168
Method: MLE Df Model: 1
Date: Fri, 07 Jan 2022 Pseudo R-squ.: 0.09495
Time: 01:44:12 Log-Likelihood: -23.492
converged: True LL-Null: -25.957
LLR p-value: 0.02640
\n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "
coef std err z P>|z| [0.025 0.975]
Intercept -10.7845 3.934 -2.741 0.006 -18.496 -3.073
Age 0.2976 0.148 2.012 0.044 0.008 0.587
" + ], + "text/plain": [ + "\n", + "\"\"\"\n", + " Logit Regression Results \n", + "==============================================================================\n", + "Dep. Variable: Status No. Observations: 170\n", + "Model: Logit Df Residuals: 168\n", + "Method: MLE Df Model: 1\n", + "Date: Fri, 07 Jan 2022 Pseudo R-squ.: 0.09495\n", + "Time: 01:44:12 Log-Likelihood: -23.492\n", + "converged: True LL-Null: -25.957\n", + " LLR p-value: 0.02640\n", + "==============================================================================\n", + " coef std err z P>|z| [0.025 0.975]\n", + "------------------------------------------------------------------------------\n", + "Intercept -10.7845 3.934 -2.741 0.006 -18.496 -3.073\n", + "Age 0.2976 0.148 2.012 0.044 0.008 0.587\n", + "==============================================================================\n", + "\"\"\"" + ] + }, + "execution_count": 87, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "log_reg2 = smf.logit(formula = \"Status ~ Age\", data = nosmoke_group).fit()\n", + "log_reg2.summary()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], "metadata": { "kernelspec": { "display_name": "Python 3", @@ -16,10 +766,9 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.6.3" + "version": "3.7.0" } }, "nbformat": 4, "nbformat_minor": 2 } -