Skip to content

Instantly share code, notes, and snippets.

@fonnesbeck

fonnesbeck/Epidemiology RW-Najwa.ipynb

Created March 30, 2017 16:30
Show Gist options
  • Save fonnesbeck/7fba27c423b62da6dd01358575662ca7 to your computer and use it in GitHub Desktop.
Save fonnesbeck/7fba27c423b62da6dd01358575662ca7 to your computer and use it in GitHub Desktop.
Download ZIP
Epidemiology RW-Najwa.ipynb
Display the source blob
Display the rendered blob
Raw
Epidemiology RW-Najwa.ipynb
{
"cells": [
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "%matplotlib inline\nimport pandas as pd\nimport numpy as np\nimport matplotlib.pyplot as plt\nfrom datetime import datetime\nimport seaborn as sb\nimport pymc3 as pm\nimport theano.tensor as tt\nsb.set_style(\"white\")",
"execution_count": 1,
"outputs": []
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "Import data"
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized = pd.read_csv('data/hospitalized.csv', index_col=0)",
"execution_count": 102,
"outputs": [
{
"output_type": "stream",
"text": "/Users/fonnescj/anaconda3/envs/dev/lib/python3.6/site-packages/IPython/core/interactiveshell.py:2717: DtypeWarning: Columns (141,143,145,147,149,182,207,213,214,263,284,285,286,300,301) have mixed types. Specify dtype option on import or set low_memory=False.\n interactivity=interactivity, compiler=compiler, result=result)\n",
"name": "stderr"
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "# Positive culture lookup\npcr_lookup = {'pcr_result___1': 'RSV',\n'pcr_result___2': 'HMPV',\n'pcr_result___3': 'flu A',\n'pcr_result___4': 'flu B',\n'pcr_result___5': 'rhino',\n'pcr_result___6': 'PIV1',\n'pcr_result___7': 'PIV2',\n'pcr_result___8': 'PIV3',\n'pcr_result___13': 'H1N1',\n'pcr_result___14': 'H3N2',\n'pcr_result___15': 'Swine',\n'pcr_result___16': 'Swine H1',\n'pcr_result___17': 'flu C',\n'pcr_result___18': 'Adeno'}",
"execution_count": 3,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized['RSV'] = hospitalized.pcr_result___1.astype(bool)\nhospitalized['HMPV'] = hospitalized.pcr_result___2.astype(bool)\nhospitalized['Rhino'] = hospitalized.pcr_result___5.astype(bool)\nhospitalized['Influenza'] = (hospitalized.pcr_result___3 | hospitalized.pcr_result___4 |\n hospitalized.pcr_result___17)\nhospitalized['Adeno'] = hospitalized.pcr_result___18.astype(bool)\nhospitalized['PIV'] = (hospitalized.pcr_result___6 | hospitalized.pcr_result___7 | \n hospitalized.pcr_result___8)\nhospitalized['No virus'] = hospitalized[list(pcr_lookup.keys())].sum(1) == 0\nhospitalized['All'] = True",
"execution_count": 4,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "viruses = ['RSV', 'HMPV', 'Rhino', 'Influenza', 'Adeno', 'PIV', 'No virus', 'All']",
"execution_count": 5,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "non_rsv_lookup = pcr_lookup.copy()\nnon_rsv_lookup.pop('pcr_result___1')",
"execution_count": 6,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "'RSV'"
},
"metadata": {},
"execution_count": 6
}
]
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "Identify individuals with coinfection"
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized['coinfection'] = hospitalized[list(pcr_lookup.keys())].sum(1) > 1",
"execution_count": 7,
"outputs": []
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "Virus frequency by coinfection status"
},
{
"metadata": {
"collapsed": false,
"scrolled": true,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized[(~hospitalized.coinfection) & (hospitalized.RSV)].shape[0]/hospitalized.shape[0]",
"execution_count": 8,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "0.2297979797979798"
},
"metadata": {},
"execution_count": 8
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized['male'] = (hospitalized.sex=='M').astype(int)\nage_groups = pd.get_dummies(pd.cut(hospitalized.age_months, [0,1,11,23]))\nage_groups.index = hospitalized.index\nage_groups.columns = 'under 2 months', '2-11 months', '12-23 months'\nhospitalized = hospitalized.join(age_groups)",
"execution_count": 9,
"outputs": []
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "Diagnosis"
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized['ros'] = hospitalized.adm_sepsis | hospitalized.adm_febrile",
"execution_count": 10,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized['pertussis-like cough'] = hospitalized.adm_pertussis | hospitalized.adm_cough",
"execution_count": 11,
"outputs": []
},
{
"metadata": {
"collapsed": true,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized['brochopneumonia'] = hospitalized.adm_bronchopneumo",
"execution_count": 12,
"outputs": []
},
{
"metadata": {
"collapsed": true,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized['bronchiolitis'] = hospitalized.adm_bronchiolitis",
"execution_count": 13,
"outputs": []
},
{
"metadata": {
"collapsed": true,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized['pneumonia'] = hospitalized.adm_pneumo",
"execution_count": 14,
"outputs": []
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "## Rate Estimation"
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized.admission_date = pd.to_datetime(hospitalized.admission_date)\nhospitalized.admission_date.describe()",
"execution_count": 15,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": "\ncount 3168\nunique 794\ntop 2011-02-07 00:00:00\nfreq 17\nfirst 2010-03-15 00:00:00\nlast 2013-03-30 00:00:00\nName: admission_date, dtype: object"
},
"metadata": {}
}
]
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "Age groups:\n\n- Under 2 months\n- 2-5 mo.\n- 6-11 mo.\n- Over 11 mo."
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "age_groups = pd.get_dummies(pd.cut(hospitalized.age_months, [0,1,5,11,24], include_lowest=True))\nage_groups.index = hospitalized.index\nage_groups.columns = 'age_under_2', 'age_2_5', 'age_6_11', 'age_over_11'\nhospitalized = hospitalized.join(age_groups)",
"execution_count": 16,
"outputs": []
},
{
"metadata": {
"collapsed": true,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "age_group_lookup = {'age_under_2': '<2', \n 'age_2_5': '2-5', \n 'age_6_11': '6-11', \n 'age_over_11': '>11'}",
"execution_count": 17,
"outputs": []
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "## Population rate estimation\n\nRecode year to virus season:\n\n- 2011: March 2010 - March 2011\n- 2012: Apr 2011 - Mar 2012\n- 2013: April 2012 - Mar 2013"
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized['virus_year'] = 2011\nhospitalized.loc[(hospitalized.admission_date >= '2011-03-31') \n & (hospitalized.admission_date <= '2012-03-31'), 'virus_year'] = 2012\nhospitalized.loc[hospitalized.admission_date > '2012-03-31', 'virus_year'] = 2013\n\nhospitalized.virus_year.value_counts()",
"execution_count": 18,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": "\n2012 1191\n2013 1179\n2011 798\nName: virus_year, dtype: int64"
},
"metadata": {}
}
]
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "Recode zones"
},
{
"metadata": {
"collapsed": true,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "zone_string = \"1, Amman Zone 1 | 2, Abdoun Zone 1 | 3, Abu Alanda Zone 3 | 4, Abu Nusair Zone 2 | 5, Airport street Zone 4 | 6, Al.Ashrafeyeh Zone 5 | 7, Al.Badya Zone 31 | 8, Al.Baqa'a Zone 22 | 9, Al.Ghour Zone 32 | 10, Al.Hashmi Zone 6 | 11, Al.Hezam Zone 29 | 12, Al.Hussein Camping Zone 1 | 13, Al.Istiklal Zone 1 | 14, Al.Jeeza Zone 8 | 15, Al.Joufeh Zone 5 | 16, Al.Karak Zone 35 | 17, Al.Lubban Zone 16 | 18, Al.Madenah Al.Reyadeyeh Zone 1 | 19, Al.Mahatta Zone 6 | 20, Al.Manarah Zone 5 | 21, Al.Mareikh Zone 5 | 22, Al.Marqab Zone 6 | 23, Al.Muhajreen Zone 5 | 24, Al.Musdar Zone 5 | 25, Al.Musherfeh Zone 15 | 26, Al.Naser Zone 6 | 27, Al.Natheef Zone 5 | 28, Al.Nuzha Zone 1 | 29, Al.Qastal Zone 10 | 30, Al.Qwesmeh Zone 5 | 31, Al.Shouneh Zone 32 | 32, Al.Taj Zone 5 | 33, Al.Taybeh Zone 7 | 34, Aqaba Zone 30 | 35, Arjan Zone 1 | 36, Bayader Wadi AL.Seer Zone 20 | 37, Bnayyat Zone 12 | 38, D.Al.Ameer Ali Zone 13 | 39, D.Al.Aqsa Zone 9 | 40, D.Haj Hasan Zone 9 | 41, Daheyet Al.Rasheed Zone 17 | 42, Daheyet al.Yasmeen Zone 1 | 43, Dead Sea Zone 32 | 44, Deer.Al.Ghbar Zone 1 | 45, Down Town (Al.Balad) Zone 1 | 46, Dra'a al.Qarbi Zone 1 | 47, Ein Al.Basha Zone 22 | 48, Ein Ghazal Zone 6 | 49, Eskan Al.Ameer Hashem Zone 29 | 50, Eskan Al.Ameer Talal Zone 29 | 51, Etha'a wal Telvesion Zone 9 | 52, Hay Al.Dabaybeh Zone 5 | 53, Hay Al.Tafayleh Zone 5 | 54, Hay Nazzal Zone 1 | 55, Huttein (shneler ) Refugee camping Zone 6 | 56, Iraq Al.Ameer Zone 23 | 57, Jabal AL.Akhdar Zone 1 | 58, Jabal Al.Ameer Faisal Zone 29 | 59, Jabal AL.Hadeed Zone 5 | 60, Jabal Al.Hussein Zone 1 | 61, Jabal Al.Qosoor Zone 1 | 62, Jabal Amman Zone 1 | 63, Jarash Zone 36 | 64, Jawa Zone 7 | 65, Juwaydeh Zone 14 | 66, Khalda Zone 18 | 67, Khreibt Al.Souk Zone 7 | 68, Ma'an Zone 31 | 69, Madaba Zone 34 | 70, Mafraq Zone 33 | 71, Marj Al.Hamam Zone 19 | 72, Marka Zone 6 | 73, Muqableen Zone 9 | 74, Muwaqqar Zone 28 | 75, Nadi Al.Sebaq Zone 6 | 76, Naur Zone 19 | 77, Petra Zone 30 | 78, Qatranah Zone 30 | 79, Raghadan Zone 1 | 80, Ras Al.Ein Zone 1 | 81, Rusayfah Zone 29 | 82, Saffout Zone 22 | 83, Sahab Zone 11 | 84, Salheyet Al.Abed Zone 6 | 85, Shafa Badran Zone 25 | 86, Sharq Al.Awsat Zone 11 | 87, Shemasani Zone 1 | 88, Street 30 Zone 5 | 89, Summaya Street Zone 5 | 90, Suweileh Zone 27 | 91, Tabarbour Zone 21 | 92, Tla'a al Ali Zone 17 | 93, Um Al.Heran Zone 11 | 94, Um Al.Summaq Zone 17 | 95, Um Nuwwara and Adan Zone 5 | 96, Um Uthayna Zone 1 | 97, Wadi Abdoun Zone 1 | 98, Wadi AL.Haddadeh Zone 1 | 99, Wadi AL.Hajar Zone 29 | 100, Wadi Al.Remam Zone 5 | 101, Wadi AL.Seer Zone 20 | 102, Wadi Saqra Zone 1 | 103, Wehdat Zone 11 | 104, Yadoudeh Zone 13 | 105, Yajouz Zone 26 | 106, Zarqa Zone 29 | 107, Zezya Zone 24\"",
"execution_count": 19,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "zones = [z.strip().split(',') for z in zone_string.split('|')]",
"execution_count": 20,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "zone_dict = {int(n):int(s.strip().split(' ')[-1]) for n,s in zones}",
"execution_count": 21,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized['zone'] = hospitalized.city_zone.replace(zone_dict)",
"execution_count": 22,
"outputs": []
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "Define Amman zone"
},
{
"metadata": {
"collapsed": true,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized['amman_zone'] = hospitalized.zone<28",
"execution_count": 23,
"outputs": []
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "Hospitalized in Amman"
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized_amman = hospitalized[hospitalized.amman_zone]",
"execution_count": 24,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized_amman.shape",
"execution_count": 25,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "(3048, 433)"
},
"metadata": {},
"execution_count": 25
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "assert hospitalized_amman.index.is_unique",
"execution_count": 26,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hosp_age_counts = hospitalized_amman.groupby('admission_date')[age_groups.columns].sum().resample('M').sum().values",
"execution_count": 27,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "pre_2013 = hospitalized_amman.admission_date < datetime(2013, 1, 1)",
"execution_count": 28,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "age_counts = hospitalized_amman[pre_2013].groupby('admission_date')[age_groups.columns].sum().resample('M').sum()\nage_counts",
"execution_count": 29,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": "\n age_under_2 age_2_5 age_6_11 age_over_11\nadmission_date \n2010-03-31 15 16 14 3\n2010-04-30 17 22 14 6\n2010-05-31 12 16 11 2\n2010-06-30 21 6 7 3\n2010-07-31 19 6 2 2\n2010-08-31 7 8 3 3\n2010-09-30 10 5 4 2\n2010-10-31 5 5 4 4\n2010-11-30 5 5 8 7\n2010-12-31 7 8 5 10\n2011-01-31 32 58 34 21\n2011-02-28 54 67 39 24\n2011-03-31 40 31 18 23\n2011-04-30 23 37 26 12\n2011-05-31 24 34 11 17\n2011-06-30 25 17 13 9\n2011-07-31 20 16 10 10\n2011-08-31 21 7 0 3\n2011-09-30 17 8 5 8\n2011-10-31 20 9 6 10\n2011-11-30 10 14 7 14\n2011-12-31 23 21 21 18\n2012-01-31 60 68 37 37\n2012-02-29 55 83 39 25\n2012-03-31 49 71 37 33\n2012-04-30 34 43 32 23\n2012-05-31 34 27 15 11\n2012-06-30 23 21 15 10\n2012-07-31 29 18 8 14\n2012-08-31 21 11 7 5\n2012-09-30 13 17 15 10\n2012-10-31 16 20 4 9\n2012-11-30 11 17 5 9\n2012-12-31 43 46 31 17",
"text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n <thead>\n <tr style=\"text-align: right;\">\n <th></th>\n <th>age_under_2</th>\n <th>age_2_5</th>\n <th>age_6_11</th>\n <th>age_over_11</th>\n </tr>\n <tr>\n <th>admission_date</th>\n <th></th>\n <th></th>\n <th></th>\n <th></th>\n </tr>\n </thead>\n <tbody>\n <tr>\n <th>2010-03-31</th>\n <td>15</td>\n <td>16</td>\n <td>14</td>\n <td>3</td>\n </tr>\n <tr>\n <th>2010-04-30</th>\n <td>17</td>\n <td>22</td>\n <td>14</td>\n <td>6</td>\n </tr>\n <tr>\n <th>2010-05-31</th>\n <td>12</td>\n <td>16</td>\n <td>11</td>\n <td>2</td>\n </tr>\n <tr>\n <th>2010-06-30</th>\n <td>21</td>\n <td>6</td>\n <td>7</td>\n <td>3</td>\n </tr>\n <tr>\n <th>2010-07-31</th>\n <td>19</td>\n <td>6</td>\n <td>2</td>\n <td>2</td>\n </tr>\n <tr>\n <th>2010-08-31</th>\n <td>7</td>\n <td>8</td>\n <td>3</td>\n <td>3</td>\n </tr>\n <tr>\n <th>2010-09-30</th>\n <td>10</td>\n <td>5</td>\n <td>4</td>\n <td>2</td>\n </tr>\n <tr>\n <th>2010-10-31</th>\n <td>5</td>\n <td>5</td>\n <td>4</td>\n <td>4</td>\n </tr>\n <tr>\n <th>2010-11-30</th>\n <td>5</td>\n <td>5</td>\n <td>8</td>\n <td>7</td>\n </tr>\n <tr>\n <th>2010-12-31</th>\n <td>7</td>\n <td>8</td>\n <td>5</td>\n <td>10</td>\n </tr>\n <tr>\n <th>2011-01-31</th>\n <td>32</td>\n <td>58</td>\n <td>34</td>\n <td>21</td>\n </tr>\n <tr>\n <th>2011-02-28</th>\n <td>54</td>\n <td>67</td>\n <td>39</td>\n <td>24</td>\n </tr>\n <tr>\n <th>2011-03-31</th>\n <td>40</td>\n <td>31</td>\n <td>18</td>\n <td>23</td>\n </tr>\n <tr>\n <th>2011-04-30</th>\n <td>23</td>\n <td>37</td>\n <td>26</td>\n <td>12</td>\n </tr>\n <tr>\n <th>2011-05-31</th>\n <td>24</td>\n <td>34</td>\n <td>11</td>\n <td>17</td>\n </tr>\n <tr>\n <th>2011-06-30</th>\n <td>25</td>\n <td>17</td>\n <td>13</td>\n <td>9</td>\n </tr>\n <tr>\n <th>2011-07-31</th>\n <td>20</td>\n <td>16</td>\n <td>10</td>\n <td>10</td>\n </tr>\n <tr>\n <th>2011-08-31</th>\n <td>21</td>\n <td>7</td>\n <td>0</td>\n <td>3</td>\n </tr>\n <tr>\n <th>2011-09-30</th>\n <td>17</td>\n <td>8</td>\n <td>5</td>\n <td>8</td>\n </tr>\n <tr>\n <th>2011-10-31</th>\n <td>20</td>\n <td>9</td>\n <td>6</td>\n <td>10</td>\n </tr>\n <tr>\n <th>2011-11-30</th>\n <td>10</td>\n <td>14</td>\n <td>7</td>\n <td>14</td>\n </tr>\n <tr>\n <th>2011-12-31</th>\n <td>23</td>\n <td>21</td>\n <td>21</td>\n <td>18</td>\n </tr>\n <tr>\n <th>2012-01-31</th>\n <td>60</td>\n <td>68</td>\n <td>37</td>\n <td>37</td>\n </tr>\n <tr>\n <th>2012-02-29</th>\n <td>55</td>\n <td>83</td>\n <td>39</td>\n <td>25</td>\n </tr>\n <tr>\n <th>2012-03-31</th>\n <td>49</td>\n <td>71</td>\n <td>37</td>\n <td>33</td>\n </tr>\n <tr>\n <th>2012-04-30</th>\n <td>34</td>\n <td>43</td>\n <td>32</td>\n <td>23</td>\n </tr>\n <tr>\n <th>2012-05-31</th>\n <td>34</td>\n <td>27</td>\n <td>15</td>\n <td>11</td>\n </tr>\n <tr>\n <th>2012-06-30</th>\n <td>23</td>\n <td>21</td>\n <td>15</td>\n <td>10</td>\n </tr>\n <tr>\n <th>2012-07-31</th>\n <td>29</td>\n <td>18</td>\n <td>8</td>\n <td>14</td>\n </tr>\n <tr>\n <th>2012-08-31</th>\n <td>21</td>\n <td>11</td>\n <td>7</td>\n <td>5</td>\n </tr>\n <tr>\n <th>2012-09-30</th>\n <td>13</td>\n <td>17</td>\n <td>15</td>\n <td>10</td>\n </tr>\n <tr>\n <th>2012-10-31</th>\n <td>16</td>\n <td>20</td>\n <td>4</td>\n <td>9</td>\n </tr>\n <tr>\n <th>2012-11-30</th>\n <td>11</td>\n <td>17</td>\n <td>5</td>\n <td>9</td>\n </tr>\n <tr>\n <th>2012-12-31</th>\n <td>43</td>\n <td>46</td>\n <td>31</td>\n <td>17</td>\n </tr>\n </tbody>\n</table>\n</div>"
},
"metadata": {}
}
]
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "Rates and demographics (via [World Bank](http://databank.worldbank.org)) and 2004 census data (via Jordan [Department of Statistics](http://www.dos.gov.jo/dos_home_e/main/population/census2004/group3/table_31.pdf))."
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "# Jordan population, 2008-2012\npopulation = 5786000, 5915000, 6046000, 6181000, 6318000\n\n# Interpolated population by gender and age, 2010-2012\nfemale_0 = 93649, 95739, 96486\nmale_0 = 98435, 100525, 101160\nfemale_1 = 87941, 93511, 93067\nmale_1 = 92548, 98306, 97763\n\nkids_0 = np.array(female_0) + np.array(male_0)\nkids_1 = np.array(female_1) + np.array(male_1)\nkids = kids_0 + kids_1\n\nkids_under6mo = kids_6to12mo = kids_0/2.\n\n# Proportion in Amman\namman_urban_2004 = 1784502.\njordan_2004 = 5103639.\namman_prop = amman_urban_2004 / jordan_2004\n\n# Birth rates (per 1000)\nbirth_rate = 29.665, 29.322, 28.869, 28.317, 27.699\n\n# Neonatal mortality (per 1000)\nneonatal_mort = 12.7, 12.4, 12.1, 11.8, 11.5\n\n# Infant mortality (per 1000)\ninfant_mort = 18.4, 17.9, 17.3, 16.8, 16.4",
"execution_count": 30,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "births = np.array(population[-3:])/1000. * birth_rate[-3:]\nbirths",
"execution_count": 31,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "array([ 174541.974, 175027.377, 175002.282])"
},
"metadata": {},
"execution_count": 31
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "births_6m = births/2.\nbirths_6m",
"execution_count": 32,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "array([ 87270.987 , 87513.6885, 87501.141 ])"
},
"metadata": {},
"execution_count": 32
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "deaths_6m = births_6m/1000. * infant_mort[-3:]\ndeaths_6m",
"execution_count": 33,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "array([ 1509.7880751, 1470.2299668, 1435.0187124])"
},
"metadata": {},
"execution_count": 33
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "amman_prop",
"execution_count": 34,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "0.3496528653378501"
},
"metadata": {},
"execution_count": 34
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "(births_6m - deaths_6m)*amman_prop",
"execution_count": 35,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "array([ 29986.6489389 , 30085.34181971, 30093.26626638])"
},
"metadata": {},
"execution_count": 35
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "n = np.array((kids_under6mo, kids_6to12mo, kids_1, kids))\nn",
"execution_count": 36,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "array([[ 96042., 98132., 98823.],\n [ 96042., 98132., 98823.],\n [ 180489., 191817., 190830.],\n [ 372573., 388081., 388476.]])"
},
"metadata": {},
"execution_count": 36
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "kids_0",
"execution_count": 37,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "array([192084, 196264, 197646])"
},
"metadata": {},
"execution_count": 37
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "kids_1/kids_0",
"execution_count": 38,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "array([ 0.93963578, 0.97734174, 0.9655141 ])"
},
"metadata": {},
"execution_count": 38
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "amman_prop",
"execution_count": 39,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "0.3496528653378501"
},
"metadata": {},
"execution_count": 39
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "n_amman = np.floor(n*amman_prop).astype(int)",
"execution_count": 40,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "n_amman",
"execution_count": 41,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "array([[ 33581, 34312, 34553],\n [ 33581, 34312, 34553],\n [ 63108, 67069, 66724],\n [130271, 135693, 135831]])"
},
"metadata": {},
"execution_count": 41
}
]
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "Monthly admissions in Ammaon"
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "admissions_by_month = hospitalized_amman.groupby('admission_date')['child_name'].count().resample('1M').sum()",
"execution_count": 42,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "admissions_by_month.head()",
"execution_count": 43,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": "\nadmission_date\n2010-03-31 48\n2010-04-30 59\n2010-05-31 41\n2010-06-30 37\n2010-07-31 29\nFreq: M, Name: child_name, dtype: int64"
},
"metadata": {}
}
]
},
{
"metadata": {
"collapsed": true,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "# Enrolling 5 days per week\np_enroll = 5./7",
"execution_count": 44,
"outputs": []
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "'age_under_2', 'age_2_5', 'age_6_11', 'age_over_11'"
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized_amman['admission_year'] = hospitalized_amman.admission_date.apply(lambda x: x.year)\n_under_6 = hospitalized_amman[hospitalized_amman.age_under_2.astype(bool) | hospitalized_amman.age_2_5].groupby('admission_year')['child_name'].count()",
"execution_count": 45,
"outputs": [
{
"output_type": "stream",
"text": "/Users/fonnescj/anaconda3/envs/dev/lib/python3.6/site-packages/ipykernel/__main__.py:1: SettingWithCopyWarning: \nA value is trying to be set on a copy of a slice from a DataFrame.\nTry using .loc[row_indexer,col_indexer] = value instead\n\nSee the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n if __name__ == '__main__':\n",
"name": "stderr"
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "_6_11 = hospitalized_amman[hospitalized_amman.age_6_11.astype(bool)].groupby('admission_year')['child_name'].count()",
"execution_count": 46,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "_over_11 = hospitalized_amman[hospitalized_amman.age_over_11.astype(bool)].groupby('admission_year')['child_name'].count()",
"execution_count": 47,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "_all = hospitalized_amman.groupby('admission_year')['child_name'].count()",
"execution_count": 48,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "hospitalized_amman.set_index(hospitalized_amman.admission_date).groupby([pd.TimeGrouper('M')]).count()['mother_name']",
"execution_count": 49,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": "\nadmission_date\n2010-03-31 48\n2010-04-30 59\n2010-05-31 41\n2010-06-30 37\n2010-07-31 29\n2010-08-31 21\n2010-09-30 21\n2010-10-31 18\n2010-11-30 25\n2010-12-31 30\n2011-01-31 145\n2011-02-28 184\n2011-03-31 112\n2011-04-30 98\n2011-05-31 84\n2011-06-30 64\n2011-07-31 56\n2011-08-31 31\n2011-09-30 38\n2011-10-31 45\n2011-11-30 45\n2011-12-31 83\n2012-01-31 202\n2012-02-29 202\n2012-03-31 190\n2012-04-30 132\n2012-05-31 87\n2012-06-30 69\n2012-07-31 68\n2012-08-31 44\n2012-09-30 55\n2012-10-31 49\n2012-11-30 42\n2012-12-31 137\n2013-01-31 179\n2013-02-28 158\n2013-03-31 117\nName: mother_name, dtype: int64"
},
"metadata": {}
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "diagnosis_df = pd.concat([_under_6, _6_11, _over_11, _all], axis=1)\ndiagnosis_df.columns = ('under 6 mo.', '6-11 mo.', '11-23 mo.', 'all under 2 yr.')\ndiagnosis_df",
"execution_count": 50,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": "\n under 6 mo. 6-11 mo. 11-23 mo. all under 2 yr.\nadmission_year \n2010 215 72 42 329\n2011 628 190 169 987\n2012 830 245 203 1278\n2013 298 98 58 454",
"text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n <thead>\n <tr style=\"text-align: right;\">\n <th></th>\n <th>under 6 mo.</th>\n <th>6-11 mo.</th>\n <th>11-23 mo.</th>\n <th>all under 2 yr.</th>\n </tr>\n <tr>\n <th>admission_year</th>\n <th></th>\n <th></th>\n <th></th>\n <th></th>\n </tr>\n </thead>\n <tbody>\n <tr>\n <th>2010</th>\n <td>215</td>\n <td>72</td>\n <td>42</td>\n <td>329</td>\n </tr>\n <tr>\n <th>2011</th>\n <td>628</td>\n <td>190</td>\n <td>169</td>\n <td>987</td>\n </tr>\n <tr>\n <th>2012</th>\n <td>830</td>\n <td>245</td>\n <td>203</td>\n <td>1278</td>\n </tr>\n <tr>\n <th>2013</th>\n <td>298</td>\n <td>98</td>\n <td>58</td>\n <td>454</td>\n </tr>\n </tbody>\n</table>\n</div>"
},
"metadata": {}
}
]
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "Use proportion in Amman to calculate proportion of kids in each age group and year in Amman"
},
{
"metadata": {
"trusted": true,
"collapsed": false
},
"cell_type": "code",
"source": "amman_prop",
"execution_count": 59,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "0.3496528653378501"
},
"metadata": {},
"execution_count": 59
}
]
},
{
"metadata": {
"trusted": true,
"collapsed": false
},
"cell_type": "code",
"source": "n.T.ravel()",
"execution_count": 65,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "array([ 96042., 96042., 180489., 372573., 98132., 98132.,\n 191817., 388081., 98823., 98823., 190830., 388476.])"
},
"metadata": {},
"execution_count": 65
}
]
},
{
"metadata": {
"trusted": true,
"collapsed": false,
"scrolled": true
},
"cell_type": "code",
"source": "n[:-1]",
"execution_count": 60,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "array([[ 96042., 98132., 98823.],\n [ 96042., 98132., 98823.],\n [ 180489., 191817., 190830.],\n [ 372573., 388081., 388476.]])"
},
"metadata": {},
"execution_count": 60
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "from theano import shared\n\ndef rate_model(diagnosis):\n \n # Extract data subset for passed virus\n diagnosis_subset = hospitalized_amman[hospitalized_amman[diagnosis]==1].copy()\n \n # Create data frame of age x year counts\n u6 = diagnosis_subset.age_under_2.astype(bool) | diagnosis_subset.age_2_5\n _under_6 = diagnosis_subset[u6].groupby('virus_year')['child_name'].count()\n _6_11 = diagnosis_subset[diagnosis_subset.age_6_11.astype(bool)].groupby('virus_year')['child_name'].count()\n _over_11 = diagnosis_subset[diagnosis_subset.age_over_11.astype(bool)].groupby('virus_year')['child_name'].count()\n _all = diagnosis_subset.groupby('virus_year')['child_name'].count()\n \n diagnosis_df = pd.concat([_under_6, _6_11, _over_11, _all], axis=1)\n diagnosis_df.columns = ('under 6 mo.', '6-11 mo.', '11-23 mo.', 'all ages')\n \n diagnosis_array = diagnosis_df.values.ravel()\n print(diagnosis_array)\n kids = n.T.ravel()\n print(kids)\n \n with pm.Model() as model:\n \n # Al Bashir hospital market share\n market_share = pm.Uniform('market_share', 0.5, 0.6)\n\n # Correct for 5 days of enrollment per week\n p_hosp = market_share * (5./7)\n\n # Number of 1 y.o. in Amman\n n_amman = pm.Binomial('n_amman', kids, amman_prop, shape=kids.shape)\n# rng = tt.shared_randomstreams.RandomStreams()\n# n_amman = rng.binomial(n=kids.astype(int), p=amman_prop, size=kids.shape)\n\n # Prior probability\n prev_diag = pm.Beta('prev_diag', 1, 5, shape=diagnosis_array.size)\n per_1000 = pm.Deterministic('per_10000', prev_diag*10000)\n\n # Infected count in Amman\n y_amman = pm.Binomial('y_amman', n_amman, prev_diag, shape=diagnosis_df.size)\n# y_amman = rng.binomial(n=n_amman, p=prev_diag, size=(diagnosis_df.size,))\n\n # Likelihood for number with virus in hospital (assumes Pr(hosp | virus) = 1)\n pm.Binomial('y_hosp', y_amman, p_hosp, observed=diagnosis_array)\n\n \n return model",
"execution_count": 72,
"outputs": []
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "### Pneumonia rates"
},
{
"metadata": {
"trusted": true,
"collapsed": true
},
"cell_type": "code",
"source": "niterations = 100000\nnkeep = 10000",
"execution_count": 73,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "with rate_model('pneumonia') as pneumo_model:\n trace_pneumo = pm.sample(niterations, step=pm.Metropolis(), njobs=2)",
"execution_count": 74,
"outputs": [
{
"output_type": "stream",
"text": "[ 56 13 11 80 75 17 32 124 117 24 30 171]\n[ 96042. 96042. 180489. 372573. 98132. 98132. 191817. 388081.\n 98823. 98823. 190830. 388476.]\n",
"name": "stdout"
},
{
"output_type": "stream",
"text": "/Users/fonnescj/Repos/pymc3/pymc3/sampling.py:166: UserWarning: Instantiated step methods cannot be automatically initialized. init argument ignored.\n warnings.warn('Instantiated step methods cannot be automatically initialized. init argument ignored.')\n100%|██████████| 100000/100000 [04:32<00:00, 366.87it/s]\n",
"name": "stderr"
}
]
},
{
"metadata": {
"collapsed": true,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "prevalence_labels = ['%s %s' % (year, age) for year in ('2011', '2012', '2013') \n for age in ('under 6 mo.', '6-11 mo.', '12-23 mo.', 'all ages')]",
"execution_count": 75,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "pm.forestplot(trace_pneumo[-nkeep:], varnames=['per_10000'], \n ylabels=prevalence_labels, main='Pneumonia (per 10000)')",
"execution_count": 76,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "<matplotlib.gridspec.GridSpec at 0x11d8d9748>"
},
"metadata": {},
"execution_count": 76
},
{
"output_type": "display_data",
"data": {
"text/plain": "<matplotlib.figure.Figure at 0x11d8d96a0>",
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAf0AAAF8CAYAAADFIGL1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n0Fp/zneEtgAAAABJRU5ErkJggg==\n"
},
"metadata": {}
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "pm.summary(trace_pneumo[-nkeep:], varnames=['per_10000'])",
"execution_count": 77,
"outputs": [
{
"output_type": "stream",
"text": "\nper_10000:\n\n Mean SD MC Error 95% HPD interval\n -------------------------------------------------------------------\n \n 45.556 6.330 0.485 [34.138, 58.135]\n 11.773 2.724 0.229 [6.757, 17.055]\n 4.956 1.582 0.140 [2.496, 8.052]\n 16.613 1.811 0.139 [13.030, 20.058]\n 58.715 7.633 0.617 [43.439, 72.344]\n 14.124 3.327 0.289 [8.041, 21.096]\n 13.498 2.351 0.187 [9.319, 18.422]\n 24.544 2.333 0.182 [20.307, 29.432]\n 91.497 9.589 0.772 [72.088, 109.119]\n 19.545 3.990 0.327 [11.782, 26.846]\n 12.403 2.219 0.178 [8.441, 17.394]\n 33.697 2.464 0.178 [29.010, 38.656]\n\n Posterior quantiles:\n 2.5 25 50 75 97.5\n |--------------|==============|==============|--------------|\n \n 34.493 41.023 45.017 49.701 58.815\n 7.082 9.725 11.507 13.680 17.400\n 2.669 3.771 4.754 5.899 8.821\n 13.148 15.350 16.531 17.825 20.279\n 44.176 53.361 58.889 63.988 73.693\n 8.300 11.828 13.839 16.185 21.471\n 9.567 11.821 13.299 14.944 18.681\n 20.071 22.945 24.485 26.069 29.309\n 72.755 84.863 91.805 98.158 110.056\n 12.562 16.631 19.308 22.077 28.472\n 8.366 10.936 12.236 13.751 17.351\n 28.888 32.045 33.633 35.383 38.582\n\n",
"name": "stdout"
}
]
},
{
"metadata": {
"trusted": true,
"collapsed": false
},
"cell_type": "code",
"source": "pneumo_df = pm.df_summary(trace_pneumo[-nkeep:], varnames=['per_10000'], )\npneumo_df.index = prevalence_labels",
"execution_count": 85,
"outputs": []
},
{
"metadata": {
"trusted": true,
"collapsed": false
},
"cell_type": "code",
"source": "pneumo_df",
"execution_count": 86,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": "\n mean sd mc_error hpd_2.5 hpd_97.5\n2011 under 6 mo. 45.556492 6.330157 0.485433 34.137976 58.134535\n2011 6-11 mo. 11.773122 2.724259 0.229005 6.757460 17.055463\n2011 12-23 mo. 4.956200 1.582006 0.139846 2.496128 8.052311\n2011 all ages 16.612553 1.810870 0.139455 13.029766 20.057730\n2012 under 6 mo. 58.714962 7.633362 0.616922 43.439193 72.343841\n2012 6-11 mo. 14.124459 3.327001 0.288640 8.040512 21.095547\n2012 12-23 mo. 13.497997 2.350882 0.187344 9.319081 18.421821\n2012 all ages 24.544155 2.332945 0.181614 20.307164 29.431807\n2013 under 6 mo. 91.496812 9.588543 0.771705 72.087852 109.118532\n2013 6-11 mo. 19.544625 3.990094 0.327031 11.781594 26.846234\n2013 12-23 mo. 12.403117 2.218972 0.178091 8.441360 17.393714\n2013 all ages 33.696513 2.463959 0.178474 29.010174 38.655915",
"text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n <thead>\n <tr style=\"text-align: right;\">\n <th></th>\n <th>mean</th>\n <th>sd</th>\n <th>mc_error</th>\n <th>hpd_2.5</th>\n <th>hpd_97.5</th>\n </tr>\n </thead>\n <tbody>\n <tr>\n <th>2011 under 6 mo.</th>\n <td>45.556492</td>\n <td>6.330157</td>\n <td>0.485433</td>\n <td>34.137976</td>\n <td>58.134535</td>\n </tr>\n <tr>\n <th>2011 6-11 mo.</th>\n <td>11.773122</td>\n <td>2.724259</td>\n <td>0.229005</td>\n <td>6.757460</td>\n <td>17.055463</td>\n </tr>\n <tr>\n <th>2011 12-23 mo.</th>\n <td>4.956200</td>\n <td>1.582006</td>\n <td>0.139846</td>\n <td>2.496128</td>\n <td>8.052311</td>\n </tr>\n <tr>\n <th>2011 all ages</th>\n <td>16.612553</td>\n <td>1.810870</td>\n <td>0.139455</td>\n <td>13.029766</td>\n <td>20.057730</td>\n </tr>\n <tr>\n <th>2012 under 6 mo.</th>\n <td>58.714962</td>\n <td>7.633362</td>\n <td>0.616922</td>\n <td>43.439193</td>\n <td>72.343841</td>\n </tr>\n <tr>\n <th>2012 6-11 mo.</th>\n <td>14.124459</td>\n <td>3.327001</td>\n <td>0.288640</td>\n <td>8.040512</td>\n <td>21.095547</td>\n </tr>\n <tr>\n <th>2012 12-23 mo.</th>\n <td>13.497997</td>\n <td>2.350882</td>\n <td>0.187344</td>\n <td>9.319081</td>\n <td>18.421821</td>\n </tr>\n <tr>\n <th>2012 all ages</th>\n <td>24.544155</td>\n <td>2.332945</td>\n <td>0.181614</td>\n <td>20.307164</td>\n <td>29.431807</td>\n </tr>\n <tr>\n <th>2013 under 6 mo.</th>\n <td>91.496812</td>\n <td>9.588543</td>\n <td>0.771705</td>\n <td>72.087852</td>\n <td>109.118532</td>\n </tr>\n <tr>\n <th>2013 6-11 mo.</th>\n <td>19.544625</td>\n <td>3.990094</td>\n <td>0.327031</td>\n <td>11.781594</td>\n <td>26.846234</td>\n </tr>\n <tr>\n <th>2013 12-23 mo.</th>\n <td>12.403117</td>\n <td>2.218972</td>\n <td>0.178091</td>\n <td>8.441360</td>\n <td>17.393714</td>\n </tr>\n <tr>\n <th>2013 all ages</th>\n <td>33.696513</td>\n <td>2.463959</td>\n <td>0.178474</td>\n <td>29.010174</td>\n <td>38.655915</td>\n </tr>\n </tbody>\n</table>\n</div>"
},
"metadata": {}
}
]
},
{
"metadata": {
"trusted": true,
"collapsed": false
},
"cell_type": "code",
"source": "pneumo_df.to_csv('pneumonia_per_10000.csv')",
"execution_count": 93,
"outputs": []
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "### Brochopneumonia rates"
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "with rate_model('brochopneumonia') as bpneumo_model:\n trace_bpneumo = pm.sample(niterations, njobs=2)",
"execution_count": 89,
"outputs": [
{
"output_type": "stream",
"text": "[ 86 106 78 270 142 124 108 374 118 128 92 338]\n[ 96042. 96042. 180489. 372573. 98132. 98132. 191817. 388081.\n 98823. 98823. 190830. 388476.]\n",
"name": "stdout"
},
{
"output_type": "stream",
"text": "Assigned NUTS to market_share_interval_\nAssigned Metropolis to n_amman\nAssigned NUTS to prev_diag_logodds_\nAssigned Metropolis to y_amman\n100%|██████████| 100000/100000 [17:14<00:00, 96.68it/s]\n",
"name": "stderr"
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "pm.forestplot(trace_bpneumo[-nkeep:], varnames=['per_10000'], \n ylabels=prevalence_labels, main='Bronchopneumonia (per 10000)')",
"execution_count": 90,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "<matplotlib.gridspec.GridSpec at 0x12ac47cc0>"
},
"metadata": {},
"execution_count": 90
},
{
"output_type": "display_data",
"data": {
"text/plain": "<matplotlib.figure.Figure at 0x12c98b0b8>",
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAf0AAAF8CAYAAADFIGL1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/wBt024z7U0wOgAAAABJRU5ErkJggg==\n"
},
"metadata": {}
}
]
},
{
"metadata": {
"collapsed": false,
"scrolled": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "pm.summary(trace_bpneumo[-nkeep:], varnames=['per_10000'])",
"execution_count": 91,
"outputs": [
{
"output_type": "stream",
"text": "\nper_10000:\n\n Mean SD MC Error 95% HPD interval\n -------------------------------------------------------------------\n \n 69.036 7.904 0.428 [53.418, 84.091]\n 83.219 9.649 0.626 [64.624, 101.797]\n 33.071 4.064 0.225 [25.284, 41.197]\n 55.128 4.178 0.325 [46.920, 63.393]\n 109.649 10.930 0.748 [89.261, 131.856]\n 95.231 9.786 0.652 [76.515, 114.594]\n 42.744 4.830 0.307 [33.168, 52.024]\n 73.140 5.424 0.458 [62.443, 83.150]\n 91.293 9.427 0.585 [72.894, 109.471]\n 97.433 10.067 0.665 [78.771, 117.983]\n 36.820 4.181 0.240 [28.748, 45.051]\n 65.432 4.763 0.390 [56.137, 74.144]\n\n Posterior quantiles:\n 2.5 25 50 75 97.5\n |--------------|==============|==============|--------------|\n \n 54.030 63.550 68.860 74.318 84.803\n 65.226 76.381 83.086 89.742 102.493\n 25.740 30.203 32.856 35.650 41.852\n 46.834 52.398 55.109 57.879 63.354\n 88.521 102.149 109.577 117.040 131.274\n 76.536 88.428 95.082 101.865 114.642\n 33.736 39.381 42.587 45.950 52.640\n 62.148 69.467 73.437 77.042 82.899\n 73.757 84.817 91.003 97.485 110.434\n 78.130 90.435 97.267 104.230 117.413\n 29.020 33.909 36.670 39.593 45.360\n 55.992 62.030 65.723 68.860 74.040\n\n",
"name": "stdout"
}
]
},
{
"metadata": {
"collapsed": true,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "bpneumo_df = pm.df_summary(trace_bpneumo[-nkeep:], varnames=['per_10000'], )\nbpneumo_df.index = prevalence_labels",
"execution_count": 92,
"outputs": []
},
{
"metadata": {
"trusted": true,
"collapsed": true
},
"cell_type": "code",
"source": "bpneumo_df.to_csv('bronchopneumonia_per_10000.csv')",
"execution_count": 97,
"outputs": []
},
{
"metadata": {
"editable": true,
"deletable": true
},
"cell_type": "markdown",
"source": "### Bronchiolitis rates"
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "with rate_model('bronchiolitis') as bronchiolitis_model:\n trace_bronchiolitis = pm.sample(niterations, njobs=2)",
"execution_count": 95,
"outputs": [
{
"output_type": "stream",
"text": "[142 27 2 171 149 36 9 194 127 29 7 163]\n[ 96042. 96042. 180489. 372573. 98132. 98132. 191817. 388081.\n 98823. 98823. 190830. 388476.]\n",
"name": "stdout"
},
{
"output_type": "stream",
"text": "Assigned NUTS to market_share_interval_\nAssigned Metropolis to n_amman\nAssigned NUTS to prev_diag_logodds_\nAssigned Metropolis to y_amman\n100%|██████████| 100000/100000 [26:50<00:00, 62.10it/s] \n",
"name": "stderr"
}
]
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"editable": true,
"deletable": true
},
"cell_type": "code",
"source": "pm.forestplot(trace_bronchiolitis[-nkeep:], varnames=['per_10000'], ylabels=prevalence_labels, \n main='Bronchiolitis (per 10000)')",
"execution_count": 98,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": "<matplotlib.gridspec.GridSpec at 0x12f188630>"
},
"metadata": {},
"execution_count": 98
},
{
"output_type": "display_data",
"data": {
"text/plain": "<matplotlib.figure.Figure at 0x12f188668>",
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAf0AAAF9CAYAAAAOfLFQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhBBC+Akp+kIIIYSfkKIvhBBC+In/A5NlQ0Au0uD/AAAAAElFTkSuQmCC\n"
},
"metadata": {}
}
]
},
{
"metadata": {
"trusted": true,
"collapsed": false
},
"cell_type": "code",
"source": "pm.summary(trace_bronchiolitis[-nkeep:], varnames=['per_10000'])",
"execution_count": 99,
"outputs": [
{
"output_type": "stream",
"text": "\nper_10000:\n\n Mean SD MC Error 95% HPD interval\n -------------------------------------------------------------------\n \n 113.751 8.321 0.526 [98.144, 130.298]\n 22.115 4.086 0.228 [14.264, 30.072]\n 1.290 0.751 0.027 [0.124, 2.778]\n 35.178 2.897 0.227 [29.865, 41.214]\n 116.663 10.889 0.884 [96.628, 138.553]\n 28.758 4.938 0.316 [19.685, 38.441]\n 4.025 1.255 0.059 [1.773, 6.544]\n 39.013 3.251 0.269 [32.710, 45.437]\n 99.030 9.092 0.693 [80.752, 116.585]\n 23.172 4.245 0.252 [15.233, 31.630]\n 3.202 1.144 0.046 [1.153, 5.484]\n 32.641 3.244 0.279 [26.752, 39.377]\n\n Posterior quantiles:\n 2.5 25 50 75 97.5\n |--------------|==============|==============|--------------|\n \n 98.199 108.028 113.540 119.318 130.392\n 14.594 19.219 21.966 24.825 30.462\n 0.268 0.735 1.140 1.702 3.107\n 29.972 33.124 35.008 37.038 41.351\n 96.767 109.021 116.022 123.982 138.777\n 20.451 25.344 28.302 31.637 39.661\n 1.945 3.137 3.886 4.785 6.860\n 32.544 36.831 38.996 41.267 45.277\n 81.507 92.812 98.908 105.014 117.594\n 15.819 20.185 22.807 25.793 32.536\n 1.354 2.371 3.062 3.890 5.808\n 27.078 30.308 32.411 34.590 39.865\n\n",
"name": "stdout"
}
]
},
{
"metadata": {
"trusted": true,
"collapsed": true
},
"cell_type": "code",
"source": "bronchiolitis_df = pm.df_summary(trace_bronchiolitis[-nkeep:], varnames=['per_10000'], )\nbronchiolitis_df.index = prevalence_labels",
"execution_count": 100,
"outputs": []
},
{
"metadata": {
"trusted": true,
"collapsed": true
},
"cell_type": "code",
"source": "bronchiolitis_df.to_csv('bronchiolitis_per_10000.csv')",
"execution_count": 101,
"outputs": []
}
],
"metadata": {
"anaconda-cloud": {},
"kernelspec": {
"name": "python3",
"display_name": "Python 3",
"language": "python"
},
"language_info": {
"name": "python",
"version": "3.6.0",
"mimetype": "text/x-python",
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"pygments_lexer": "ipython3",
"nbconvert_exporter": "python",
"file_extension": ".py"
},
"latex_envs": {
"bibliofile": "biblio.bib",
"cite_by": "apalike",
"current_citInitial": 1,
"eqLabelWithNumbers": true,
"eqNumInitial": 0
},
"nav_menu": {},
"toc": {
"navigate_menu": true,
"number_sections": false,
"sideBar": false,
"threshold": "3",
"toc_cell": true,
"toc_section_display": "block",
"toc_window_display": false
},
"gist": {
"id": "7fba27c423b62da6dd01358575662ca7",
"data": {
"description": "Epidemiology RW-Najwa.ipynb",
"public": true
}
},
"_draft": {
"nbviewer_url": "https://gist.github.com/7fba27c423b62da6dd01358575662ca7"
}
},
"nbformat": 4,
"nbformat_minor": 1
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment