From 303f7d4b28c8d054cdd0057237dd35e38127fc40 Mon Sep 17 00:00:00 2001 From: 096323827ddd6110f7ba1fd93f19e12c <096323827ddd6110f7ba1fd93f19e12c@app-learninglab.inria.fr> Date: Tue, 6 Apr 2021 20:34:11 +0000 Subject: [PATCH] Final 2 --- module3/exo3/saev_final.ipynb | 1032 +++++++++++++++++++++++++++++++++ 1 file changed, 1032 insertions(+) create mode 100644 module3/exo3/saev_final.ipynb diff --git a/module3/exo3/saev_final.ipynb b/module3/exo3/saev_final.ipynb new file mode 100644 index 0000000..5481a9e --- /dev/null +++ b/module3/exo3/saev_final.ipynb @@ -0,0 +1,1032 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Sujet 6 : Autour du Paradoxe de Simpson" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "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." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Préparation" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Chargement des données" + ] + }, + { + "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", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \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": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "%matplotlib inline\n", + "import matplotlib.pyplot as plt\n", + "import pandas as pd\n", + "\n", + "csv_url = 'https://gitlab.inria.fr/learninglab/mooc-rr/mooc-rr-ressources/-/raw/master/module3/Practical_session/Subject6_smoking.csv?inline=false'\n", + "raw_data = pd.read_csv(csv_url)\n", + "raw_data" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 1. Calcul des taux de moratlité globaux" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Trie des données brutes par catégories fumeueses / non fumeuses et vivantes / décédées." + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [], + "source": [ + "smoker = raw_data[raw_data['Smoker'] == 'Yes']\n", + "non_smoker = raw_data[raw_data['Smoker'] == 'No']\n", + "\n", + "smoker_alive = smoker[smoker['Status'] == 'Alive']\n", + "smoker_dead = smoker[smoker['Status'] == 'Dead']\n", + "non_smoker_alive = non_smoker[non_smoker['Status'] == 'Alive']\n", + "non_smoker_dead = non_smoker[non_smoker['Status'] == 'Dead']" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Calcul des effectifs :" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "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", + "
VivantDécédéTotal
0Fumeur443139945
1Non Fumeur502230369
2Total9456731314
\n", + "
" + ], + "text/plain": [ + " Vivant Décédé Total\n", + "0 Fumeur 443 139 945\n", + "1 Non Fumeur 502 230 369\n", + "2 Total 945 673 1314" + ] + }, + "execution_count": 17, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "nb_sma = len(smoker_alive)\n", + "nb_smd = len(smoker_dead)\n", + "nb_nsma = len(non_smoker_alive)\n", + "nb_nsmd = len(non_smoker_dead)\n", + " \n", + "\n", + "tab = pd.DataFrame({'': ['Fumeur', 'Non Fumeur', 'Total'],\n", + " 'Vivant': [nb_sma, nb_nsma, nb_sma + nb_nsma],\n", + " 'Décédé': [nb_smd, nb_nsmd, nb_sma + nb_nsmd],\n", + " 'Total': [nb_nsma + nb_sma, nb_nsmd + nb_smd, nb_sma + nb_smd + nb_nsma + nb_nsmd]\n", + " }, columns = ['', 'Vivant', 'Décédé', 'Total'])\n", + "tab" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Calcul des taux de mortalité :" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Taux de mortalité chez les fumeuses: 0.239\n", + "Taux de mortalité chez les non-fumeuses: 0.314\n" + ] + } + ], + "source": [ + "print(f'Taux de mortalité chez les fumeuses: {nb_smd / (nb_smd + nb_sma):.3f}')\n", + "print(f'Taux de mortalité chez les non-fumeuses: {nb_nsmd / (nb_nsmd + nb_nsma):.3f}')\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "On peut constater étonnamment que le taux de mortalité est plus élevé chez les femmes non-fumeuses que ches les femmes fumeuses." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 2. Calcul des taux de moratlité par tranche d'age" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "On décide maintenant de calculer les taux de mortalité par tranche d'age : : 18-34 ans, 35-54 ans, 55-64 ans, plus de 65 ans." + ] + }, + { + "cell_type": "code", + "execution_count": 40, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Taux de mortalité :\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", + "
AgeFumeurNon fumeur
0(18, 34)0.0279330.027397
1(35, 54)0.1703060.099476
2(55, 64)0.4434780.327731
3(65, 999)0.8571430.854922
\n", + "
" + ], + "text/plain": [ + " Age Fumeur Non fumeur\n", + "0 (18, 34) 0.027933 0.027397\n", + "1 (35, 54) 0.170306 0.099476\n", + "2 (55, 64) 0.443478 0.327731\n", + "3 (65, 999) 0.857143 0.854922" + ] + }, + "execution_count": 40, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "slices = [(int(min(raw_data['Age'])), 34), (35, 54), (55, 64), (65, int(max(raw_data['Age'])))]\n", + "\n", + "rates = {}\n", + "nbs = {}\n", + "\n", + "for sli in slices:\n", + " filter_data = raw_data[sli[0] <= raw_data['Age']]\n", + " filter_data = filter_data[filter_data['Age'] < sli[1]]\n", + " \n", + " smoker = filter_data[filter_data['Smoker'] == 'Yes']\n", + " non_smoker = filter_data[filter_data['Smoker'] == 'No']\n", + "\n", + " smoker_alive = smoker[smoker['Status'] == 'Alive']\n", + " smoker_dead = smoker[smoker['Status'] == 'Dead']\n", + " non_smoker_alive = non_smoker[non_smoker['Status'] == 'Alive']\n", + " non_smoker_dead = non_smoker[non_smoker['Status'] == 'Dead']\n", + " \n", + " nbs[sli] = (len(smoker),len(non_smoker))\n", + " rates[sli] = (len(smoker_dead) / len(smoker), len(non_smoker_dead) / len(non_smoker))\n", + " \n", + "print('Taux de mortalité :')\n", + "tab = pd.DataFrame({'Age': slices,\n", + " 'Fumeur': [rates[sli][0] for sli in slices],\n", + " 'Non fumeur': [rates[sli][1] for sli in slices],\n", + " }, columns = ['Age', 'Fumeur', 'Non fumeur'])\n", + "tab" + ] + }, + { + "cell_type": "code", + "execution_count": 39, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "x = [sli[0] for sli in slices]\n", + "pl1, = plt.plot(x, [rates[sli][0] for sli in slices], 'or')\n", + "pl2, = plt.plot(x, [rates[sli][1] for sli in slices], 'ob')\n", + "plt.legend([pl1, pl2], ['Fumeuses', 'Non Fumeuses'])\n", + "plt.title(\"Taux de mortalité par tranche d'age\")\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Ces résultats semblent contredire les résultats obtenus lors de la première partie. En effet On constate que pour chaque tranche d'age prise séparément, le taux de mortalité est plus élevé chez les fumeuses que chez les non-fumeuses. Cela semble bien illustrer le paradoxe de Simpson. Ces résultats contradictoires peuvent-être expliqués en affichant les effectifs de chaque catégorie:" + ] + }, + { + "cell_type": "code", + "execution_count": 41, + "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", + "
AgeFumeurNon fumeur
0(18, 34)179219
1(35, 54)229191
2(55, 64)115119
3(65, 999)49193
\n", + "
" + ], + "text/plain": [ + " Age Fumeur Non fumeur\n", + "0 (18, 34) 179 219\n", + "1 (35, 54) 229 191\n", + "2 (55, 64) 115 119\n", + "3 (65, 999) 49 193" + ] + }, + "execution_count": 41, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "tab = pd.DataFrame({'Age': slices,\n", + " 'Fumeur': [nbs[sli][0] for sli in slices],\n", + " 'Non fumeur': [nbs[sli][1] for sli in slices],\n", + " }, columns = ['Age', 'Fumeur', 'Non fumeur'])\n", + "tab" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "On peut constater que la réparttion des fumeuses / non fumeuses est relativement équilibrée par tranche d'age, mis à part pour la dernière tranche 65+ ans. En effet, dans cet age, le nombre de non fumeuses est 4 fois supérieur. Or c'est aussi dans cette tranche qu'on enregistre logiquement le plus haut taux de mortalité, et ce quelle que soit la catégorie. Ainsi, ce grand nombre de personnes agées ayant participé à l'étude couplé au fort taux de mortalité pour cette tranche d'age augmente fortement le taux de mortalité total pour les non fuemurs, ce qui est moins le cas chez les fumeurs. Cela peut être confirmé en calculant les taux de mortalité totaux en retirant les 65+ ans:" + ] + }, + { + "cell_type": "code", + "execution_count": 101, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Taux de mortalité chez les fumeuses: 0.182\n", + "Taux de mortalité chez les non-fumeuses: 0.121\n", + "89.9\n" + ] + } + ], + "source": [ + "filter_data = raw_data[raw_data['Age'] < 65]\n", + "\n", + "smoker = filter_data[filter_data['Smoker'] == 'Yes']\n", + "non_smoker = filter_data[filter_data['Smoker'] == 'No']\n", + "\n", + "smoker_alive = smoker[smoker['Status'] == 'Alive']\n", + "smoker_dead = smoker[smoker['Status'] == 'Dead']\n", + "non_smoker_alive = non_smoker[non_smoker['Status'] == 'Alive']\n", + "non_smoker_dead = non_smoker[non_smoker['Status'] == 'Dead']\n", + "\n", + "print(f'Taux de mortalité chez les fumeuses: {len(smoker_dead) / len(smoker):.3f}')\n", + "print(f'Taux de mortalité chez les non-fumeuses: {len(non_smoker_dead) / len(non_smoker):.3f}')\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 3. Etude probabilistique" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Afin d'éviter un biais induit par des regroupements en tranches d'âges arbitraires et non régulières, nous réalisons une régression logistique. " + ] + }, + { + "cell_type": "code", + "execution_count": 103, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "import numpy as np\n", + "from sklearn.linear_model import LogisticRegression\n", + "\n", + "def list_to_array(l):\n", + " return np.array([[i] for i in l])\n", + "\n", + "probs = []\n", + "\n", + "xs = list(range(int(min(raw_data['Age'])), int(max(raw_data['Age']))))\n", + "for fil in ['Yes', 'No']:\n", + " filter_data = raw_data[raw_data['Smoker'] == fil]\n", + " ages = filter_data['Age']\n", + " deaths = (filter_data['Status'] == 'Dead').astype(int)\n", + "\n", + " model = LogisticRegression(solver='liblinear', random_state=0).fit(list_to_array(ages), deaths)\n", + " probs.append(model.predict_proba(list_to_array(xs))[:,1].tolist())\n", + " \n", + "pl1 = plt.plot(xs, probs[0], '-r', label='Fumeuses')\n", + "pl2 = plt.plot(xs , probs[1], '-b', label='Non Fumeuses')\n", + "plt.title(\"Probabilité de décès en fonction de l'age.\")\n", + "plt.legend()\n", + "plt.show() \n", + " " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "On peut constater sur ce graphique que jusqu'à 65 ans, fumer semble augmeneter la probabilité de décès sur une période de 20 ans, ce qui prouve une diminution de l'espérance de vie. La tendance s'inverse à partir de 65 ans, explicable en partie par la différence d'effectifs dans cette tranche d'age." + ] + } + ], + "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 +} -- 2.18.1