{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Sujet 6 : Autour du Paradoxe de Simpson" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Contexte de l'étude" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Cette étude porte sur le [Paradoxe de Simpson](https://fr.wikipedia.org/wiki/Paradoxe_de_Simpson) (Simpson 1951, Undy 1903). Ce paradoxe est un paradoxe statistique \"dans lequel un phénomène observé de plusieurs groupes semble s'inverser lorsque les groupes sont combinés. Ce résultat qui semble impossible au premier abord est lié à des éléments qui ne sont pas pris en compte (comme la présence de variables non indépendantes ou de différences d'effectifs entre les groupes, etc.) est souvent rencontré dans la réalité, en particulier dans les sciences sociales et les statistiques médicales\" (Wikipédia). \n", "\n", "Pour représenter ce paradoxe, on utilisera les données d'un sondage des années 1970 d'une ville du nord-est de l'Angleterre sur un sixième des électeurs, complété par une seconde étude 20 ans plus tard (Vanderpump et al. 1995) sur les mêmes personnes. Le sondage initial avait été réalisé afin d'expliciter les travaux sur les maladies thyroïdiennes et cardiaques (Tunbridge et al. 1977). Le second sondage avait pour objectif de savoir si les individus étaient envore en vie, notamment au vu de leur tabagisme.\n", "\n", "Pour ce MOOC : \"Nous nous restreindrons aux femmes et parmi celles-ci aux 1314 qui ont été catégorisées comme \"fumant\n", "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\" (MOOC Recherche Reproductible)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Importation des librairies python" ] }, { "cell_type": "code", "execution_count": 146, "metadata": {}, "outputs": [], "source": [ "import os\n", "import urllib.request\n", "import pandas as pd\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "import statsmodels.api as sm\n", "from statsmodels.formula.api import logit\n", "%matplotlib inline\n", "\n", "# Supprime l'affichage des UserWarnings avec toutes les dépréciations de fonctions\n", "import warnings \n", "warnings.simplefilter('ignore')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Traitement des données\n", "\n", "Les donnés sont disponibles sur le GitLab du MOOC Reproductibilité. Par soucis d'accessibilité et pour éviter toute disparition ou de modification de lien vers les données, on enregistrera les données récupérées de manière locale. Elles seront uniquement téléchargées si la copie locale n'existe pas.\n" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "data_url = 'https://gitlab.inria.fr/learninglab/mooc-rr/mooc-rr-ressources/-/raw/master/module3/Practical_session/Subject6_smoking.csv?inline=false'\n", "data_file = 'simpson_paradox.csv'\n", "\n", "if not os.path.exists(data_file):\n", " urllib.request.urlretrieve(data_url, data_file)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Chaque ligne des données représente une personne avec comme information:\n", "- Si la personne fume (Yes/No)\n", "- Si elle est vivante ou morte au moment de la 2ème étude (Alive/Dead)\n", "- Son âge au 1er sondage (arrondi à la 1ère décimale)" ] }, { "cell_type": "code", "execution_count": 173, "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", " \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", " \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", " \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", " \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", " \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", " \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", " \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", " \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
5NoAlive36.8
6NoAlive23.8
7YesDead57.5
8YesAlive24.8
9YesAlive49.5
10YesAlive30.0
11NoDead66.0
12YesAlive49.2
13NoAlive58.4
14NoDead60.6
15NoAlive25.1
16NoAlive43.5
17NoAlive27.1
18NoAlive58.3
19YesAlive65.7
20NoDead73.2
21YesAlive38.3
22NoAlive33.4
23YesDead62.3
24NoAlive18.0
25NoAlive56.2
26YesAlive59.2
27NoAlive25.8
28NoDead36.9
29NoAlive20.2
............
1284YesDead36.0
1285YesAlive48.3
1286NoAlive63.1
1287NoAlive60.8
1288YesDead39.3
1289NoAlive36.7
1290NoAlive63.8
1291NoDead71.3
1292NoAlive57.7
1293NoAlive63.2
1294NoAlive46.6
1295YesDead82.4
1296YesAlive38.3
1297YesAlive32.7
1298NoAlive39.7
1299YesDead60.0
1300NoDead71.0
1301NoAlive20.5
1302NoAlive44.4
1303YesAlive31.2
1304YesAlive47.8
1305YesAlive60.9
1306NoDead61.4
1307YesAlive43.0
1308NoAlive42.1
1309YesAlive35.9
1310NoAlive22.3
1311YesDead62.1
1312NoDead88.6
1313NoAlive39.1
\n", "

1314 rows × 3 columns

\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\n", "5 No Alive 36.8\n", "6 No Alive 23.8\n", "7 Yes Dead 57.5\n", "8 Yes Alive 24.8\n", "9 Yes Alive 49.5\n", "10 Yes Alive 30.0\n", "11 No Dead 66.0\n", "12 Yes Alive 49.2\n", "13 No Alive 58.4\n", "14 No Dead 60.6\n", "15 No Alive 25.1\n", "16 No Alive 43.5\n", "17 No Alive 27.1\n", "18 No Alive 58.3\n", "19 Yes Alive 65.7\n", "20 No Dead 73.2\n", "21 Yes Alive 38.3\n", "22 No Alive 33.4\n", "23 Yes Dead 62.3\n", "24 No Alive 18.0\n", "25 No Alive 56.2\n", "26 Yes Alive 59.2\n", "27 No Alive 25.8\n", "28 No Dead 36.9\n", "29 No Alive 20.2\n", "... ... ... ...\n", "1284 Yes Dead 36.0\n", "1285 Yes Alive 48.3\n", "1286 No Alive 63.1\n", "1287 No Alive 60.8\n", "1288 Yes Dead 39.3\n", "1289 No Alive 36.7\n", "1290 No Alive 63.8\n", "1291 No Dead 71.3\n", "1292 No Alive 57.7\n", "1293 No Alive 63.2\n", "1294 No Alive 46.6\n", "1295 Yes Dead 82.4\n", "1296 Yes Alive 38.3\n", "1297 Yes Alive 32.7\n", "1298 No Alive 39.7\n", "1299 Yes Dead 60.0\n", "1300 No Dead 71.0\n", "1301 No Alive 20.5\n", "1302 No Alive 44.4\n", "1303 Yes Alive 31.2\n", "1304 Yes Alive 47.8\n", "1305 Yes Alive 60.9\n", "1306 No Dead 61.4\n", "1307 Yes Alive 43.0\n", "1308 No Alive 42.1\n", "1309 Yes Alive 35.9\n", "1310 No Alive 22.3\n", "1311 Yes Dead 62.1\n", "1312 No Dead 88.6\n", "1313 No Alive 39.1\n", "\n", "[1314 rows x 3 columns]" ] }, "execution_count": 173, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data = pd.read_csv(data_url)\n", "data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On vérifir que toutes nos lignes sont bien remplies et que les âges sont cohérents" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
SmokerStatusAge
\n", "
" ], "text/plain": [ "Empty DataFrame\n", "Columns: [Smoker, Status, Age]\n", "Index: []" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ " data[data.isnull().any(axis=1)]" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Ages minimaux et maximaux: [18.0, 89.9]\n" ] } ], "source": [ "print('Ages minimaux et maximaux: ' + str([data.Age.min(), data.Age.max()]))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Etudes\n", "\n", "### Décès en fonction des habitudes de tabagisme\n", "\n", "Le tableau suivant récapitule le nombre de femmes mortes ou vivantes selon sa relation au tabac." ] }, { "cell_type": "code", "execution_count": 33, "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", "
StatusAliveDeadMortality
Smoker
No5022300.314
Yes4431390.239
\n", "
" ], "text/plain": [ "Status Alive Dead Mortality\n", "Smoker \n", "No 502 230 0.314\n", "Yes 443 139 0.239" ] }, "execution_count": 33, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data_death = data.groupby(['Smoker'])['Status'].value_counts().unstack()\n", "data_death['Mortality'] = round(data_death['Dead'] / (data_death['Dead'] + data_death['Alive']), 3) \n", "data_death" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On peut afficher graphiquement les données de ce tableau:" ] }, { "cell_type": "code", "execution_count": 42, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 42, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "x = np.arange(2) # the label locations\n", "width = 0.35 # the width of the bars\n", "\n", "fig, ax = plt.subplots()\n", "ax.bar(x - width/2, data_death['Alive'], width, label='Alive')\n", "ax.bar(x + width/2, data_death['Dead'], width, label='Dead')\n", "ax2 = ax.twinx()\n", "ax2.plot(x, data_death['Mortality'], color='r', marker='o', label='Mortality')\n", "\n", "ax.set_ylabel('Number of women')\n", "ax2.set_ylabel('Mortality rate')\n", "ax2.set_ylim(0,1)\n", "ax.set_xticks(x)\n", "ax.set_xticklabels(['Non Smoker', 'Smoker'])\n", "ax.legend()\n", "ax2.legend(bbox_to_anchor=(0.8, 1))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "A partir de ces graphiques et résultats il serait logique de conclure que les non fumeuses ont une mortalité plus importante (31%) par rapport aux fumeuses (24%) et que donc fumer aide à vivre longtemps. Même en regardant les intervales de confiance sur la condition (morte **1** ou vivante **0**) de la personne suivant son statut de fumeur nous indique que les fumeurs ont plus de chance de survie." ] }, { "cell_type": "code", "execution_count": 258, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 258, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "sns.barplot(x='Smoker', y='Status', ci=95, data=data.replace('Alive', 0).replace('Dead', 1))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Mais il est de connaissance publique que \"fumer tue\". **Alors comment les données nous trompent-elles ?** Nous avons regardé les données de manière globale sans rentrer dans les détails. Si l'on regarde l'âge des femmes suivant leur statut de fumeur un paradoxe commence à apparaître:" ] }, { "cell_type": "code", "execution_count": 259, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 259, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEKCAYAAAAfGVI8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAADzVJREFUeJzt3XuQXnV9x/H3x0QHB+0IZgmpiGk7kcp4AV0viG1VhMF6CUK9TS/bykzGGS1qq2naTr116jixWnuhtmmlLtRacFqGyFgxRhFtrbAoEigi4igK2WSBosC0aMi3fzwn7RoTdlM5z9nk937NZM5zznP7wmT2nXPO85xNVSFJatdDhh5AkjQsQyBJjTMEktQ4QyBJjTMEktQ4QyBJjTMEktQ4QyBJjTMEktS45UMPsBgrVqyo1atXDz2GJB1Urr766turamKhxx0UIVi9ejUzMzNDjyFJB5Uk31rM4zw0JEmNMwSS1DhDIEmNMwSS1DhDIEmNMwSS1DhDIEmNMwSS1LiD4gtlkg5969evZ3Z2lqOPPpqNGzcOPU5TDIGkJWF2dpZbb7116DGa1GsIknwTuBu4H9hVVZNJjgQuBFYD3wReUVX/2ecckqT9G8c5gudV1QlVNdmtbwC2VtUaYGu3LkkayBAni9cC093taeCMAWaQJHX6DkEBn0xydZJ13baVVbUdoFse1fMMkqQH0PfJ4pOr6rYkRwFbknx1sU/swrEO4Nhjj+1rPklqXq8hqKrbuuXOJBcDzwB2JFlVVduTrAJ27ue5m4BNAJOTk9XnnNKQbnnnk4YeYUnYdeeRwHJ23fkt/58Ax75129jeq7dDQ0kOT/LIPbeB04DrgM3AVPewKeCSvmaQJC2szz2ClcDFSfa8zz9U1SeSXAVclORs4Bbg5T3OIElaQG8hqKpvAE/Zx/Y7gFP6el9J0oHxWkOS1DhDIEmNMwSS1DgvOidpSVhx2G5gV7fUOBkCSUvCm59819AjNMtDQ5LUOEMgSY0zBJLUOEMgSY0zBJLUOEMgSY0zBJLUOL9H0Jj169czOzvL0UcfzcaNG4ceR9ISYAgaMzs7y6233jr0GJKWEA8NSVLjDIEkNc4QSFLjDIEkNa6Zk8VPe8v5Q4+wJDzy9rtZBtxy+93+PwGufs+vDT2CNDj3CCSpcYZAkhpnCCSpcYZAkhpnCCSpcc18akgjux92+A8tJckQNObeNacNPYKkJcZDQ5LUOEMgSY0zBJLUOEMgSY0zBJLUOEMgSY3rPQRJliX5cpJLu/Ujk2xJclO3PKLvGSRJ+zeOPYI3ADfMW98AbK2qNcDWbl2SNJBeQ5DkGOBFwN/O27wWmO5uTwNn9DmDJOmB9b1H8H5gPbB73raVVbUdoFseta8nJlmXZCbJzNzcXM9jSlK7egtBkhcDO6vq6v/P86tqU1VNVtXkxMTEgzydJGmPPq81dDLw0iS/CBwG/ESSvwd2JFlVVduTrAJ29jiDJGkBve0RVNXvVtUxVbUaeBXw6ar6FWAzMNU9bAq4pK8ZJEkLG+J7BO8GTk1yE3Bqty5JGshYLkNdVZcDl3e37wBOGcf7SpIW5jeLJalxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxhkCSGmcIJKlxvYUgyWFJrkzylSTXJ3lHt/3IJFuS3NQtj+hrBknSwvrcI7gPeH5VPQU4ATg9ybOADcDWqloDbO3WJUkD6S0ENXJPt/rQ7k8Ba4Hpbvs0cEZfM0iSFtbrOYIky5JcA+wEtlTVF4GVVbUdoFse1ecMkqQH1msIqur+qjoBOAZ4RpInLva5SdYlmUkyMzc319+QktS4sXxqqKruAi4HTgd2JFkF0C137uc5m6pqsqomJyYmxjGmJDWpz08NTSR5VHf74cALgK8Cm4Gp7mFTwCV9zSBJWtjyHl97FTCdZBmj4FxUVZcm+QJwUZKzgVuAl/c4gyRpAb2FoKquBU7cx/Y7gFP6el9J0oFZ8NBQkpVJPpjkX7r147t/zUuSDgGLOUfwIeAy4Ce79a8Bb+xrIEnSeC0mBCuq6iJgN0BV7QLu73UqSdLYLCYE9yZ5NKNvBdNdJuK7vU4lSRqbxZws/i1GH/n8mST/CkwAv9TrVJKksVkwBFX1pSS/ABwHBLixqn7Q+2SSpLFYMARJztxr0+OTfBfYVlX7/FawJOngsZhDQ2cDJwGf6dafC/w7oyC8s6ou6Gk2SdIYLCYEu4EnVNUOGH2vAPgA8EzgCsAQSNJBbDGfGlq9JwKdncDjq+pOwHMFknSQW8weweeSXAp8tFs/C7giyeHAXb1NJkkai8WE4HXAmcBzuvUrgVVVdS/wvL4GkySNx4KHhqqqgJsZHQZ6GaMLxt3Q81ySpDHZ7x5BkscDrwJeDdwBXAikqtwLkKRDyAMdGvoq8DngJVX1dYAkbxrLVJKksXmgQ0NnAbPAZ5L8TZJTGH2zWJJ0CNlvCKrq4qp6JfCzjH7f8JuAlUk+kOS0Mc0nSerZYk4W31tVH66qFwPHANcAG3qfTJI0Fgf0y+ur6s6q+uuqen5fA0mSxuuAQiBJOvQYAklqnCGQpMYZAklqnCGQpMYZAklqnCGQpMYZAklqnCGQpMYZAklqnCGQpMYZAklqXG8hSPLYJJ9JckOS65O8odt+ZJItSW7qlkf0NYMkaWF97hHsAn67qp4APAt4XZLjGV3CemtVrQG24iWtJWlQvYWgqrZX1Ze623cz+oX3jwHWAtPdw6aBM/qaQZK0sLGcI0iyGjgR+CKwsqq2wygWwFHjmEGStG+9hyDJI4B/At5YVd87gOetSzKTZGZubq6/ASWpcb2GIMlDGUXgw1X1z93mHUlWdfevAnbu67lVtamqJqtqcmJios8xJalpfX5qKMAHgRuq6n3z7toMTHW3p4BL+ppBkrSw5T2+9snArwLbklzTbfs94N3ARUnOBm4BXt7jDJKkBfQWgqr6PJD93H1KX+8rSTowfrNYkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcYZAkhpnCCSpcb2FIMl5SXYmuW7etiOTbElyU7c8oq/3lyQtTp97BB8CTt9r2wZga1WtAbZ265KkAfUWgqq6Arhzr81rgenu9jRwRl/vL0lanHGfI1hZVdsBuuVR+3tgknVJZpLMzM3NjW1ASWrNkj1ZXFWbqmqyqiYnJiaGHkeSDlnjDsGOJKsAuuXOMb+/JGkv4w7BZmCquz0FXDLm95ck7aXPj49+BPgCcFyS7yQ5G3g3cGqSm4BTu3VJ0oCW9/XCVfXq/dx1Sl/vKUk6cEv2ZLEkaTwMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMMgSQ1zhBIUuMGCUGS05PcmOTrSTYMMYMkaWTsIUiyDDgXeCFwPPDqJMePew5J0sgQewTPAL5eVd+oqu8D/wisHWAOSRLDhOAxwLfnrX+n2yZJGsDyAd4z+9hWP/KgZB2wrlu9J8mNvU7VlhXA7UMPsRTkj6eGHkE/zL+be7xtXz8qD9jjFvOgIULwHeCx89aPAW7b+0FVtQnYNK6hWpJkpqomh55D2pt/N4cxxKGhq4A1SX4qycOAVwGbB5hDksQAewRVtSvJ64HLgGXAeVV1/bjnkCSNDHFoiKr6OPDxId5bgIfctHT5d3MAqfqR87SSpIZ4iQlJapwhOMRk5PNJXjhv2yuSfGLIuaT5klSS985bf3OStw84UtMMwSGmRsf6Xgu8L8lhSQ4H/gh43bCTST/kPuDMJCuGHkSG4JBUVdcBHwN+B3gbcH5V3ZxkKsmVSa5J8pdJHpJkeZILkmxLcl2Sc4adXo3YxejE8Jv2viPJ45JsTXJttzx2/OO1ZZBPDWks3gF8Cfg+MJnkicDLgGd3H+HdxOg7HDcDK6rqSQBJHjXUwGrOucC1STbutf0vGP3jZTrJa4A/A84Y+3QNMQSHqKq6N8mFwD1VdV+SFwBPB2aSADyc0TWfLgOOS/KnjD7S+8mhZlZbqup7Sc4HzgH+a95dJwFndrcvAPYOhR5khuDQtrv7A6NrPJ1XVX+w94OSPJnRZcHPAc7i/67xJPXt/Yz2XP/uAR7jZ9x75jmCdnwKeMWek3NJHp3k2CQTjL5P8lFG5xOeOuSQaktV3QlcBJw9b/O/MTpsCfDLwOfHPVdr3CNoRFVtS/IO4FNJHgL8gNGni+4HPpjR8aJidIJZGqf3Aq+ft34OcF6StwBzwG8MMlVD/GaxJDXOQ0OS1DhDIEmNMwSS1DhDIEmNMwSS1DhDoGYl+f0k13fXtLkmyTN/zNd7bpJLH6z5pHHxewRqUpKTgBcDT+0uwbECeNiA8yyvql1Dvb/a5h6BWrUKuL2q7gOoqtur6rYk30zyriRfSDKT5KlJLktyc5LXwv/+zof3dFdr3ZbklXu/eJKnJ/lykp9OcniS85Jc1W1b2z3m15N8NMnH8BpPGpB7BGrVJ4G3Jvkao8tvXFhVn+3u+3ZVnZTkT4APAScDhwHXA3/F6IJoJwBPAVYAVyW5Ys8LJ3k28OfA2qq6Jcm7gE9X1Wu6q7temeRT3cNPAp7cXWpBGoQhUJOq6p4kTwN+DngecGGSDd3dm7vlNuARVXU3cHeS/+5+kD8H+EhV3Q/sSPJZRld2/R7wBEbX2T+tqm7rXuc04KVJ3tytHwbsucb+FiOgoRkCNav7QX45cHmSbcBUd9d93XL3vNt71pczupLr/mxn9IP+RGBPCAKcVVU3zn9gd3L63h/jP0F6UHiOQE1KclySNfM2nQB8a5FPvwJ4ZZJl3dVbfx64srvvLuBFwLuSPLfbdhnwm92F/Uhy4o87v/RgMgRq1SOA6ST/keRa4Hjg7Yt87sXAtcBXgE8D66tqds+dVbUDeAlwbvev/j8EHsrot3Fd161LS4ZXH5WkxrlHIEmNMwSS1DhDIEmNMwSS1DhDIEmNMwSS1DhDIEmNMwSS1Lj/AZp5iWFf4mMkAAAAAElFTkSuQmCC\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "sns.barplot(x='Smoker', y='Age', ci=95, data=data)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "La prochaine étape est donc d'étudier les données plus précisément, notamment suivant les tranches d'âges.\n", "\n", "## Décès liés au tabagisme suivant l'âge\n", "\n", "En reprenant les données précédentes et en rajoutant une catégorie d'âge (18-34 ans, 34-54 ans, 55-64 ans, plus de 65 ans), on réalise les mêmes analyses." ] }, { "cell_type": "code", "execution_count": 260, "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", " \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", "
StatusAliveDeadMortality
AgeSmoker
(18, 34]No21260.028
Yes17250.028
(34, 54]No180190.095
Yes196410.173
(54, 64]No81400.331
Yes64510.443
(64, 100]No281650.855
Yes7420.857
\n", "
" ], "text/plain": [ "Status Alive Dead Mortality\n", "Age Smoker \n", "(18, 34] No 212 6 0.028\n", " Yes 172 5 0.028\n", "(34, 54] No 180 19 0.095\n", " Yes 196 41 0.173\n", "(54, 64] No 81 40 0.331\n", " Yes 64 51 0.443\n", "(64, 100] No 28 165 0.855\n", " Yes 7 42 0.857" ] }, "execution_count": 260, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tranche_age = [18, 34, 54, 64, 100]\n", "\n", "col_tranche = pd.cut(data['Age'], tranche_age)\n", "data_age = data.groupby(col_tranche).apply(lambda x: x.groupby(['Smoker'])['Status'].value_counts().unstack())\n", "data_age['Mortality'] = (data_age['Dead'].values / (data_age['Dead'].values + data_age['Alive'].values)).round(3) \n", "data_age" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "A partir de ce tableau on peut afficher les données:" ] }, { "cell_type": "code", "execution_count": 261, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 261, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Visualisation du taux de mortalité suivant les âges et le statut de fumeur\n", "\n", "tranche_age_label = ['[18-34]', '[35-54]', '[55-64]', '[65-100]'] # the label text\n", "x = np.arange(len(tranche_age_label)) # the label locations\n", "width = 0.35 # the width of the bars\n", "\n", "fig, ax = plt.subplots()\n", "ax.bar(x - width/2, data_age.reset_index()[data_age.reset_index().Smoker == 'Yes']['Mortality'], width, label='Smoker')\n", "ax.bar(x + width/2, data_age.reset_index()[data_age.reset_index().Smoker == 'No']['Mortality'], width, label='Non smoker')\n", "\n", "ax.set_ylabel('Mortality rate')\n", "ax.set_xlabel(\"Age group\")\n", "ax.set_xticks(x)\n", "ax.set_xticklabels(tranche_age_label)\n", "ax.legend()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On remarque sur le graphique ci-dessus que finalement pour chaque classe d'âge le taux de mortalité chez les fumeuses est supérieur ou égal à celui des non fumeuses !\n", "\n", "En s'intéressant à l'histogramme des âges chez ces deux populations ci-dessous, on s'aperçoit qu'il y a plus de non fumeuses d'âge supérieur à 65ans, qui ont donc plus de chance de décéder naturellement. Cette tranche est donc sur-représentée chez les non-fumeuses, amenant en moyenne à un taux de mortalité plus élevé.\n", "\n", "**Etudier des données dans leur ensemble peut donner des résultats très différents par rapport à des études sur des sous-groupes. Cela peut amener à des erreurs d'interprétation importantes.**" ] }, { "cell_type": "code", "execution_count": 262, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 262, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Visualisation du nombre de femmes vivantes et décédées par tranche d'âge\n", "\n", "sns.distplot(data[data.Smoker == 'Yes']['Age'], label='Smoker', kde=True)\n", "sns.distplot(data[data.Smoker == 'No']['Age'], label='Non smoker')\n", "plt.legend()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Ainsi 2 conclusions peuvent être tirées sur ce biais d'étude:\n", "- Ce biais arrive notamment à cause de la **non homogénéité de l'échantillon**. On voit bien ci-dessus que toutes les tranches d'âge ne sont pas représentées de la même manière si les femmes sont fumeuses ou non fumeuses. Il faut cependant faire attention à étudier des *tranches d'âge régulières et adaptés à l'étude*.\n", "- De plus, dans la 1ère partie l'âge des participantes avait été mis de côté au profit d'une moyenne sur l'ensemble. Cette **mise à l'écart de ce paramètre** a induit une mauvaise interprétation.\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Décès et régression logistique\n", "\n", "En dernière partie une régression logistique est réalisée afin de supprimer le biais induit par des tranches d'âges arbitraires et non régulières.\n", "\n", "Tout d'abord une nouvelle colonne est créée avec :\n", "- Si la femme est décédée: 1\n", "- Si la femme est vivante: 0\n" ] }, { "cell_type": "code", "execution_count": 263, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Exemple :\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", " \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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
SmokerStatusAge
0Yes021.0
1Yes019.3
2No157.5
3No047.1
4Yes081.4
5No036.8
6No023.8
7Yes157.5
8Yes024.8
9Yes049.5
10Yes030.0
\n", "
" ], "text/plain": [ " Smoker Status Age\n", "0 Yes 0 21.0\n", "1 Yes 0 19.3\n", "2 No 1 57.5\n", "3 No 0 47.1\n", "4 Yes 0 81.4\n", "5 No 0 36.8\n", "6 No 0 23.8\n", "7 Yes 1 57.5\n", "8 Yes 0 24.8\n", "9 Yes 0 49.5\n", "10 Yes 0 30.0" ] }, "execution_count": 263, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data_reg = data.replace('Alive', 0).replace('Dead', 1)\n", "\n", "print ('Exemple :')\n", "data_reg.loc[0:10, ]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On réalise pour chacun des groupes *'Smoker'* et *'Non smoker'* une régresion logistique pour visualiser la corrélation entre l'âge et le décès (et donc la probabilité de décès en fonction de l'âge)." ] }, { "cell_type": "code", "execution_count": 264, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Optimization terminated successfully.\n", " Current function value: 0.412727\n", " Iterations 7\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: 582
Model: Logit Df Residuals: 580
Method: MLE Df Model: 1
Date: Fri, 31 Jul 2020 Pseudo R-squ.: 0.2492
Time: 15:23:47 Log-Likelihood: -240.21
converged: True LL-Null: -319.94
LLR p-value: 1.477e-36
\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
coef std err z P>|z| [0.025 0.975]
Intercept -5.5081 0.466 -11.814 0.000 -6.422 -4.594
Age 0.0890 0.009 10.203 0.000 0.072 0.106
" ], "text/plain": [ "\n", "\"\"\"\n", " Logit Regression Results \n", "==============================================================================\n", "Dep. Variable: Status No. Observations: 582\n", "Model: Logit Df Residuals: 580\n", "Method: MLE Df Model: 1\n", "Date: Fri, 31 Jul 2020 Pseudo R-squ.: 0.2492\n", "Time: 15:23:47 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", "\"\"\"" ] }, "execution_count": 264, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Pour les Fumeuses\n", "data_reg_smoker = data_reg[data_reg.Smoker == 'Yes']\n", "model = logit('Status ~ Age', data=data_reg_smoker)\n", "result_smoker = model.fit() #algorithme de Newton-Raphson par défaut\n", "logit_smoker = result_smoker.predict(data_reg_smoker) # predictions\n", "result_smoker.summary()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Pour les fumeuses on voit que l'âge est un paramètre statistiquement important (P < 0.05), avec un coefficient de pente de 0.089 (avec une erreur de 10%), compris pour un CI de 2.5% entre 0.106 et 0.072." ] }, { "cell_type": "code", "execution_count": 265, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Optimization terminated successfully.\n", " Current function value: 0.354560\n", " Iterations 7\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: 732
Model: Logit Df Residuals: 730
Method: MLE Df Model: 1
Date: Fri, 31 Jul 2020 Pseudo R-squ.: 0.4304
Time: 15:23:49 Log-Likelihood: -259.54
converged: True LL-Null: -455.62
LLR p-value: 2.808e-87
\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
coef std err z P>|z| [0.025 0.975]
Intercept -6.7955 0.479 -14.174 0.000 -7.735 -5.856
Age 0.1073 0.008 13.742 0.000 0.092 0.123
" ], "text/plain": [ "\n", "\"\"\"\n", " Logit Regression Results \n", "==============================================================================\n", "Dep. Variable: Status No. Observations: 732\n", "Model: Logit Df Residuals: 730\n", "Method: MLE Df Model: 1\n", "Date: Fri, 31 Jul 2020 Pseudo R-squ.: 0.4304\n", "Time: 15:23:49 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", "\"\"\"" ] }, "execution_count": 265, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Pour les non Fumeuses\n", "\n", "data_reg_nosmoker = data_reg[data_reg.Smoker == 'No']\n", "model = logit('Status ~ Age', data=data_reg_nosmoker)\n", "result_nosmoker = model.fit()\n", "logit_nosmoker = result_nosmoker.predict(data_reg_nosmoker) \n", "result_nosmoker.summary()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Pour les non-fumeuses on voit que l'âge est un paramètre statistiquement important (P < 0.05), avec un coefficient de pente de 0.1073 (avec une erreur de moins de 10%, suffisamment faible pour comparer avec les résultats des fumeuses), compris pour un CI de 2.5% entre 0.123 et 0.092. Ce coefficient est plus élevé que pour les femmes fumeuses, avec cependant un coefficient d'interception plus important.\n", "\n", "Afin de mieux visualiser cette variation en fonction de l'âge, les fonctions logistiques sont tracées. Seaborn utilisant le package statsmodel pour la fonction lmplot, il est possible de l'utiliser pour visualiser de manière simple les deux courbes sur un même graphe avec les intervales de confiance pour chacune des courbes." ] }, { "cell_type": "code", "execution_count": 268, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 268, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "sns.lmplot('Age', 'Status', data=data_reg, logistic=True, ci=97.5, hue='Smoker')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "A partir des données précédentes il est possible de voir : \n", "- Pour des âges entre 35 et 60 ans, il y a plus de probabilité de décès pour les fumeuses que les non-fumeuses\n", "- Pour des âges plus élevés les courbes se rejoignent et les intervalles de confiance se recoupent, ne permettant pas de conclure sur des probabilités plus fortes de décès dans l'un ou l'autre des cas.\n", "- Le coefficient de régression des non-fumeuses est plus élevé avec une interception négative plus grande notamment parce que la probabilité de décès augmente fortement au-delà de 60 ans, comparativement à celle des non-fumeuses qui augmente de manière plus constante.\n", "\n", "**Ainsi ces régressions nous montre que l'effet du tabagisme est important pour une certaine tranche d'âge mais qu'au delà d'autres causes de décès entrent en jeu alignant le nombre de mort de manière identique entre les deux status.**\n" ] } ], "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 }