descriptives.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[33mWARNING: You are using pip version 20.0.2; however, version 20.2.2 is available.\r\n",
      "You should consider upgrading via the '/usr/local/bin/python -m pip install --upgrade pip' command.\u001b[0m\r\n"
     ]
    }
   ],
   "source": [
    "! pip install --quiet environs cyksuid toolz psycopg2-binary typing_json backoff xxhash pyyaml geopandas dtreeviz descartes pysal facebook_business"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# KacperKubara/distython"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams['figure.figsize'] = [7, 3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from environs import Env\n",
    "env = Env()\n",
    "env.read_env('.env-bug', override=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from adopt.malaria import get_conf, get_df, lookalike\n",
    "from adopt.marketing import Marketing\n",
    "from adopt.facebook.state import CampaignState\n",
    "\n",
    "cnf = get_conf(env)\n",
    "state = CampaignState(env)\n",
    "stratum = cnf['strata'][0]\n",
    "surveys = stratum['surveys']\n",
    "m = Marketing(env, state)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import geopandas as gpd\n",
    "from adopt.malaria import load_cities\n",
    "\n",
    "def get_dist_info(path):\n",
    "    geod = gpd.read_file(path)\n",
    "    district_info = geod.groupby('disthash').apply(lambda df: df.iloc[0][['disthash', 'tot_p']]).reset_index(drop=True)\n",
    "    return district_info\n",
    "\n",
    "dist_info = get_dist_info('outs/cluster.shp')\n",
    "cities = load_cities('output/cities-22.csv')\n",
    "cities = cities.merge(dist_info, how='left', on='disthash')\n",
    "\n",
    "districts = cities.groupby('distname') \\\n",
    "                  .head(1) \\\n",
    "                  .reset_index(drop=True) \\\n",
    "                  .drop(columns=['rad', 'lng', 'lat', 'distcode', 'id'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from adopt.responses import get_forms\n",
    "from datetime import datetime\n",
    "\n",
    "survey_user = '10383123-9fb1-429b-8312-447c1b54b17a'\n",
    "shortcodes = [ 'baselineeng', 'baselinehin', 'baselineodi']\n",
    "\n",
    "timestamp = datetime(2020, 8, 5)\n",
    "\n",
    "forms = list(get_forms(survey_user, shortcodes, timestamp, cnf['chatbase']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "questions = [\n",
    "    ('gender', '20218ad0-96c8-4799-bdfe-90c689c5c206'),\n",
    "    ('dist_medical', 'd7573919-8a7e-457f-9a1d-1f8c389127a7'),\n",
    "    ('education', 'e40fa1c6-13a1-4a02-91cd-0eaade11864d'),\n",
    "    ('age', 'da3d1de1-a287-4143-8b2c-107565a4e4d4'),\n",
    "    ('religion', '8e490114-df6d-4fb3-b1be-6ce11acd9212'),\n",
    "    ('caste', '912bc5d8-db4a-402b-b506-f85ce4e89e6f'),\n",
    "    ('household_size', '8065d101-7995-4471-9ca6-54e9ce6d4310'),\n",
    "    ('owns_net', 'bd4802c6-7bdb-40f0-aac1-18cc6df7da6e'),\n",
    "    ('has_ac', 'e279587c-975f-433a-adab-1ad563876af6'),\n",
    "    ('worried_covid', '25b20f99-a30c-4549-968a-05b7276fa485'),\n",
    "    ('worried_malaria', '7c955b2b-0480-4cbc-9800-f94b77078f22'),\n",
    "    ('fever_treatment_seeking', 'f526f205-7aa9-4e3a-9acc-c39766e1bc8c'),\n",
    "    ('malaria_incidence', 'f6e69027-97cc-494e-8d52-318b75047e23'),\n",
    "    ('malaria_now', 'dad23031-8468-4900-89cc-d01841d8b660'),\n",
    "    ('if_malaria_seek_treatment', '73ef8720-5e2b-463c-aa8a-ed890a839a8b'),\n",
    "    ('if_malaria_tested', '9de3d2f5-d9c0-45aa-9065-a61c975a58dc'),\n",
    "    ('net_lastnight', 'b5b1ff58-c8df-4890-9b1c-0cd40ce6edc0'),\n",
    "    ('longsleeves', '0992f107-7274-4404-aeff-bf80a723f098'),\n",
    "    ('used_spray', '02379921-3b90-4005-b813-cf1e9aec0f40'),\n",
    "    ('home', '4d0ae478-3893-4b46-ab39-d6848c69245d'), \n",
    "    ('occupation', '4fc929c7-132d-49b1-a164-515e5cc9064f')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from adopt.responses import get_response_df\n",
    "from datetime import datetime, timezone\n",
    "from adopt.clustering import only_target_users\n",
    "import pandas as pd\n",
    "from adopt.clustering import add_res_cols\n",
    "from adopt.forms import response_translator\n",
    "from adopt.malaria import shape_df\n",
    "import seaborn as sns\n",
    "import re\n",
    "\n",
    "def in_time(df, min_date, max_date):\n",
    "    min_date = datetime(*min_date, tzinfo=timezone.utc)\n",
    "    max_date = datetime(*max_date, tzinfo=timezone.utc)\n",
    "\n",
    "    users = df[(df.timestamp > min_date) & (df.timestamp < max_date)].userid.unique()\n",
    "    return df.userid.isin(users)\n",
    "\n",
    "def filter_time(df, min_date, max_date):\n",
    "    return df[in_time(df, min_date, max_date)]\n",
    "\n",
    "def ref_translation(eng_form, other_form):\n",
    "    vals = [(a['ref'], b['ref']) for a, b \n",
    "            in zip(eng_form['fields'], other_form['fields'])]\n",
    "    lookup = dict(vals)\n",
    "    rev_lookup = {v:k for k,v in lookup.items()}\n",
    "    return lookup, rev_lookup\n",
    "\n",
    "\n",
    "def get_filtered_responses(survey_user, eng_form, other_form, shortcodes, questions, db_cnf):\n",
    "    ref_lookup, rev_ref_lookup = ref_translation(eng_form, other_form)\n",
    "    refs = [ref_lookup[r] for _, r in questions]\n",
    "    responses = get_response_df(survey_user, shortcodes, refs, db_cnf)\n",
    "\n",
    "    _, last_ref = questions[-1]\n",
    "    \n",
    "    answered = responses[responses.question_ref == last_ref].userid.unique()    \n",
    "\n",
    "    return responses[responses.userid.isin(answered)].reset_index(drop=True)\n",
    "\n",
    "def malaria_prob(groupby, key='malaria'):\n",
    "    d = groupby[key].value_counts(normalize=True)\n",
    "    d.name = 'count'\n",
    "    d = d.reset_index(level=-1)\n",
    "    d = d[d[key] == True].drop(columns=key)\n",
    "    return d\n",
    "\n",
    "\n",
    "def stats(df):\n",
    "    targets = df.kutcha.sum()\n",
    "    tot = df.shape[0]\n",
    "    non_target = tot - targets\n",
    "    return pd.Series([targets/tot, targets, non_target, tot], index=['perc', 'target', 'non_target', 'tot'])\n",
    "\n",
    "\n",
    "def col_translators(a, b, questions):\n",
    "    lookup, _ = ref_translation(a, b)\n",
    "\n",
    "    refs = [ref for _, ref in questions]\n",
    "    ts = { qt['ref']: response_translator(q, qt)\n",
    "           for qt, q in zip(a['fields'], b['fields'])\n",
    "           if qt['ref'] in refs}\n",
    "    \n",
    "    return [(n, lookup[ref], ts[ref]) for n, ref in questions]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import logging\n",
    "logging.getLogger().setLevel(logging.ERROR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "filtered_res = get_filtered_responses(survey_user, forms[0], forms[1], ['baselinehin', 'baselinehinexc'], questions, cnf['chatbase'])\n",
    "\n",
    "col_names = col_translators(forms[0], forms[1], questions)\n",
    "rr = add_res_cols(col_names, shape_df(filtered_res)) \\\n",
    "    .reset_index(drop=True) \\\n",
    "    .dropna(subset=[r for r, _ in questions if not re.match('if_', r)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# choices: \n",
    "#   ordered: true\n",
    "#   alphabetize: true\n",
    "#   booleanize: Yes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "final_clusters = pd.read_csv('outs/final-clusters.csv').disthash"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from adopt.forms import make_answers\n",
    "from pandas.api.types import CategoricalDtype\n",
    "\n",
    "get_ref = lambda r: next((f for f in forms[0]['fields'] if f['ref'] == r), None)\n",
    "\n",
    "def fix_homes(dat):\n",
    "    types = ['Kutcha (made of mud, tin, straw)', 'Semi-pucca', 'Pucca (have cement/brick wall and floor']\n",
    "    t = CategoricalDtype(categories=types, ordered=True)\n",
    "    dat['home'] = dat['home'].astype(t)\n",
    "    return dat\n",
    "\n",
    "def ordinalize_dat(dat):\n",
    "    ordinal = ['worried_covid', 'worried_malaria', 'dist_medical', 'education', 'fever_treatment_seeking', 'if_malaria_seek_treatment']\n",
    "    a = [(make_answers(get_ref(r)), q) for q, r in questions if q in ordinal]\n",
    "    a = [(q, [l for _, l in li]) for li, q in a ]\n",
    "    for col, vals in a:\n",
    "        t = CategoricalDtype(categories=vals, ordered=True)\n",
    "        dat[col] = dat[col].astype(t)\n",
    "    return dat\n",
    "\n",
    "def cast_age(x):\n",
    "    try: \n",
    "        i = int(x)\n",
    "        if i > 100 or i < 12:\n",
    "            raise ValueError('impossible age')\n",
    "        return i\n",
    "    except ValueError:\n",
    "        return None\n",
    "\n",
    "def caste_household(x):\n",
    "    try:\n",
    "        i = int(x)\n",
    "        if i < 0 or i > 30:\n",
    "            raise ValueError('impossible value')\n",
    "        return i\n",
    "    except ValueError:\n",
    "        return None\n",
    "\n",
    "def _convert_cat(col):\n",
    "    try:\n",
    "        return col.cat.codes\n",
    "    except AttributeError:\n",
    "        return None\n",
    "\n",
    "def convert_cats(df):\n",
    "    cols = [(col, _convert_cat(df[col])) for col in df.columns]\n",
    "    cols =  [(n, c) for n, c in cols if c is not None]\n",
    "    names, cols = zip(*cols)\n",
    "    return cols\n",
    "\n",
    "def make_quartile_var(df, var, name):\n",
    "    df[name] = 'Interquartile'\n",
    "    df.loc[df[var] >= df[var].quantile(.75), name] = 'Upper Quartile'\n",
    "    df.loc[df[var] <= df[var].quantile(.25), name] = 'Lower Quartile'\n",
    "    df[name] = df.dist_risk.astype(CategoricalDtype(categories=['Upper Quartile', 'Interquartile', 'Lower Quartile'], ordered=True))\n",
    "    return df\n",
    "\n",
    "def add_mean(df, col, by):\n",
    "    lookup = dd.groupby(by)[col].mean().reset_index().rename(columns={col: f'mean_{col}'})\n",
    "    return df.merge(lookup)\n",
    "\n",
    "def dist_stats(dd):\n",
    "    for col in ['under_net', 'net_ownership', 'kutcha', 'malaria_now', 'malaria']:\n",
    "        dd = add_mean(dd, col, 'disthash')\n",
    "    dd['mean_malaria_2_week'] = dd.mean_malaria_now\n",
    "    dd['mean_malaria_5_year'] = dd.mean_malaria\n",
    "    return dd\n",
    "            \n",
    "ordinalize = lambda col: col.cat.codes\n",
    "integerize = lambda col: col.astype(int)\n",
    "inverse = lambda col: ~col.astype(bool)\n",
    "\n",
    "dd = rr.merge(districts, left_on='md:clusterid', right_on='disthash')\n",
    "qrefs = [c for c in dd.columns if len(c) > 35]\n",
    "dd = dd.drop(columns=qrefs)\n",
    "\n",
    "dd['under_net'] = dd.net_lastnight == 'Yes'\n",
    "dd['net_ownership'] = dd.owns_net == 'Yes'\n",
    "dd['malaria'] = dd.malaria_incidence == 'Yes'\n",
    "dd['malaria_now'] = dd.malaria_now == 'Yes'\n",
    "dd['kutcha'] = dd.home == 'Kutcha (made of mud, tin, straw)'\n",
    "dd['pucca'] = dd.home == 'Pucca (have cement/brick wall and floor'\n",
    "dd['university'] = dd.education == 'University degree or higher'\n",
    "dd['unemployed'] = dd.occupation == 'Unemployed'\n",
    "dd['ac'] = dd.has_ac == 'Yes'\n",
    "dd['net_ownership'] = dd.owns_net == 'Yes'\n",
    "dd['disthash'] = dd['md:clusterid']\n",
    "dd['used_spray'] = dd.used_spray == 'Yes'\n",
    "dd['general_caste'] = dd.caste == 'General'\n",
    "dd['lower_caste'] = dd.caste.isin(['OBC', 'SC/Dalit', 'ST'])\n",
    "\n",
    "dd = dd.groupby('disthash').apply(dist_stats).reset_index(drop=True)\n",
    "\n",
    "dd['true_age'] = dd.age.map(cast_age)\n",
    "dd['household'] = dd.household_size.map(caste_household)\n",
    "\n",
    "dd['lives_alone'] = dd.household <= 1\n",
    "dd['large_family'] = (dd.household > 8)\n",
    "dd['student'] = dd.occupation == 'Student'\n",
    "\n",
    "dd = ordinalize_dat(dd)\n",
    "dd = fix_homes(dd)\n",
    "dd['worried_malaria_true'] = dd.worried_malaria.cat.codes < 2\n",
    "\n",
    "dd = make_quartile_var(dd, 'malaria_now', 'dist_risk')\n",
    "dd = make_quartile_var(dd, 'malaria', 'dist_risk_5_year')\n",
    "\n",
    "dd = dd.groupby('disthash').filter(lambda df: df.shape[0] > 100)\n",
    "dd['male'] = dd.gender == 'Male'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "\n",
    "def upd_legend(ax, title, labels):\n",
    "    l = ax.legend()\n",
    "    l.set_title(title)\n",
    "    for t, l in zip(l.texts, labels): \n",
    "        t.set_text(l)\n",
    "\n",
    "sns.set()\n",
    "base_color = sns.color_palette(\"Set2\")[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# Mapping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def plot_incidence_map(dat, cnf, cmap='plasma'):\n",
    "    fig, ax = plt.subplots(figsize=(10, 10))\n",
    "    dat.plot(column=cnf['var'], legend=True, ax=ax, cmap=cmap, legend_kwds={'label': cnf['title'], 'orientation': \"horizontal\", 'shrink': 0.6, 'aspect': 30, 'pad': 0.0})\n",
    "    ax.get_xaxis().set_visible(False)\n",
    "    ax.get_yaxis().set_visible(False)\n",
    "    for s in ax.spines.values():\n",
    "        s.set_color(None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 639,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "  cluster_shapes = gpd.read_file('outs/cluster.shp')\n",
    "di_shapes = gpd.read_file('outs/districts.shp')\n",
    "\n",
    "dat = dd.groupby('disthash').mean().reset_index()[['disthash', 'malaria', 'malaria_now']]\n",
    "dat = dat[dat.disthash.isin(final_clusters)]\n",
    "dat = cluster_shapes.merge(dat[dat.malaria_now > 0.0], how='right', on='disthash')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "plot_incidence_map(dat, {'var': 'malaria', 'title': '5-year malaria incidence'})\n",
    "plt.savefig('incidence-map-5-year.png', bbox_inches='tight', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "plot_incidence_map(dat, {'var': 'malaria_now', 'title': '2-week malaria incidence'})\n",
    "plt.savefig('incidence-map-2-week.png', bbox_inches='tight', dpi=300)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# Misc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 653,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "home\n",
       "Kutcha (made of mud, tin, straw)           0.709705\n",
       "Semi-pucca                                 0.714431\n",
       "Pucca (have cement/brick wall and floor    0.667627\n",
       "Name: fever_seek_treatment_rightaway, dtype: float64"
      ]
     },
     "execution_count": 653,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# treatment seeking behavior same by home\n",
    "\n",
    "dd['fever_seek_treatment_rightaway'] = dd.fever_treatment_seeking == 'I would seek treatment right away'\n",
    "dd.groupby('home').fever_seek_treatment_rightaway.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sns.barplot(x='under_net', y='home', data=dd, color=base_color)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd.plot.scatter(x='mean_under_net', y='mean_malaria_5_year')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import statsmodels.formula.api as smf\n",
    "\n",
    "smf.logit('integerize(malaria_now) ~ under_net + mean_malaria_5_year + kutcha + pucca + kutcha*under_net + ac*pucca + ac*kutcha', dd).fit().summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd[(dd.kutcha & ~dd.net_ownership)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "smf.logit('integerize(malaria_now) ~ net_ownership + kutcha:inverse(net_ownership):mean_malaria_5_year + mean_malaria_5_year + kutcha + pucca + ac*pucca + ac*kutcha', dd).fit().summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1716,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def plot_incidence_barplot(cnf):\n",
    "    with sns.color_palette(sns.cubehelix_palette(3, start=1.25, rot=.0, dark=0.8, light=.35)):\n",
    "        fig, ax = plt.subplots(figsize=(12, 4))\n",
    "        sns.barplot(x=cnf['var'], y=cnf['group'], hue='home', data=dd, ax=ax)\n",
    "        legend = plt.legend()\n",
    "        legend.set_title('Dwelling Type')\n",
    "        ax.set_xlabel(cnf['title'])\n",
    "        ax.set_ylabel('District Malaria Incidence')        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1722,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dist_risk       home                                   \n",
       "Upper Quartile  Kutcha (made of mud, tin, straw)           0.048176\n",
       "                Semi-pucca                                 0.027955\n",
       "                Pucca (have cement/brick wall and floor    0.020103\n",
       "Interquartile   Kutcha (made of mud, tin, straw)           0.023831\n",
       "                Semi-pucca                                 0.015422\n",
       "                Pucca (have cement/brick wall and floor    0.012241\n",
       "Lower Quartile  Kutcha (made of mud, tin, straw)           0.009407\n",
       "                Semi-pucca                                 0.000831\n",
       "                Pucca (have cement/brick wall and floor    0.005503\n",
       "Name: malaria_now, dtype: float64"
      ]
     },
     "execution_count": 1722,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.groupby(['dist_risk', 'home']).malaria_now.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "plot_incidence_barplot({'var': 'malaria', 'group': 'dist_risk', 'title': '5-year malaria incidence (household + neighbors)'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "plot_incidence_barplot({'var': 'malaria', 'group': 'dist_risk_5_year', 'title': '5-year malaria incidence (household + neighbors)'})\n",
    "plt.savefig('incidence-barplot-5-year.png', bbox_inches='tight', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "plot_incidence_barplot({'var': 'malaria_now', 'title': '2-week malaria incidence (household'})\n",
    "plt.savefig('incidence-barplot-2-week.png', bbox_inches='tight', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Lack of net ownership of people in Kutcha dwellings potentially increases risk\n",
    "# but those with nets are still very likely to get malaria\n",
    "\n",
    "with sns.color_palette('Set2'):\n",
    "    fig, ax = plt.subplots(figsize=(12, 5))\n",
    "    sns.barplot(x='malaria_now', y='dist_risk', hue='home', data=dd, ax=ax)\n",
    "    ax.set_xlabel('2-week malaria incidence')\n",
    "    # upd_legend(ax, 'Owns Bednet', ['No', 'Yes'])\n",
    "\n",
    "\n",
    "# leg = ax.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "logit = smf.logit('target ~ mean_malaria_5_year + mean_malaria_2_week + kutcha + pucca + ac', dd.assign(target = dd.malaria_now.astype(int)))\n",
    "\n",
    "logit = logit.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.02433389623401494, 0.03262760964147726)"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.metrics import r2_score\n",
    "from sklearn.model_selection import cross_val_predict\n",
    "\n",
    "X = risk[['mean_malaria_5_year', 'mean_malaria_2_week', 'mean_worried_malaria', 'kutcha', 'pucca', 'unemployed', 'true_age']]\n",
    "ordinal_cols = convert_cats(risk[['dist_medical', 'education']])\n",
    "\n",
    "X = pd.concat([X, *ordinal_cols], axis=1)\n",
    "X = X.dropna()\n",
    "\n",
    "y = risk['worried_malaria'].cat.codes.astype(float)\n",
    "y = y[X.index]\n",
    "\n",
    "model = RandomForestRegressor(n_estimators=200, max_depth=4)\n",
    "preds = cross_val_predict(model, X, y)\n",
    "\n",
    "model.fit(X, y)\n",
    "\n",
    "r2_score(y, preds), r2_score(y, model.predict(X))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 901,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier, DecisionTreeRegressor\n",
    "\n",
    "model = DecisionTreeRegressor(max_depth=2)\n",
    "model.fit(X, y)\n",
    "\n",
    "from dtreeviz.trees import *\n",
    "\n",
    "viz = dtreeviz(model, \n",
    "               X, \n",
    "               y,\n",
    "               target_name='worried',\n",
    "               feature_names=X.columns, \n",
    "               fancy=False\n",
    "              )  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dict(zip(X.columns, cls.feature_importances_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 714,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "       False       0.99      0.62      0.76     18712\n",
      "        True       0.03      0.64      0.05       323\n",
      "\n",
      "    accuracy                           0.62     19035\n",
      "   macro avg       0.51      0.63      0.41     19035\n",
      "weighted avg       0.97      0.62      0.75     19035\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[11581,  7131],\n",
       "       [  115,   208]])"
      ]
     },
     "execution_count": 714,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "X = dd[['mean_malaria_5_year', 'mean_malaria_2_week', 'kutcha', 'pucca', 'unemployed']]\n",
    "ordinal_cols = convert_cats(dd[['dist_medical', 'education']])\n",
    "\n",
    "X = pd.concat([X, *ordinal_cols], axis=1)\n",
    "\n",
    "X = X.sample(frac=1.0)\n",
    "y = dd['malaria_now']\n",
    "y = y[X.index]\n",
    "\n",
    "from sklearn.metrics import f1_score, classification_report, recall_score, confusion_matrix\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import cross_val_predict\n",
    "\n",
    "\n",
    "cls = RandomForestClassifier(n_estimators=200, max_depth=2, class_weight='balanced')\n",
    "\n",
    "\n",
    "preds = cross_val_predict(cls, X, y)\n",
    "\n",
    "print(classification_report(y, preds))\n",
    "confusion_matrix(y, preds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 716,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.18636787, 0.39749221, 0.18156616, 0.0706766 , 0.00224975,\n",
       "       0.08928274, 0.07236466])"
      ]
     },
     "execution_count": 716,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cls.fit(X, y)\n",
    "cls.feature_importances_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 839,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-839-6e336256903d>:3: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  return dd[idx].reset_index(drop=True).assign(risk_rf = cls.predict_proba(X[idx])[:, 1])\n"
     ]
    }
   ],
   "source": [
    "def predict_risk(dd, cls):\n",
    "    idx = ~dd.malaria_now\n",
    "    return dd[idx].reset_index(drop=True).assign(risk_rf = cls.predict_proba(X[idx])[:, 1])\n",
    "\n",
    "risk = predict_risk(dd, cls)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 718,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "risk['high_risk'] = risk.risk_rf > .5\n",
    "risk['high_risk_dist'] = risk.mean_malaria_2_week > risk.mean_malaria_2_week.quantile(0.75) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 250,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "risk = risk.groupby('disthash').apply(lambda df: df.assign(mean_worried_malaria = ordinalize(df.worried_malaria).mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "malaria_prob(dd.groupby(['education', 'home']), 'worried_malaria_true')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "smf.ols('ordinalize(worried_malaria) ~ risk_rf', risk).fit().summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "smf.ols('ordinalize(worried_malaria) ~ ordinalize(education)*kutcha + mean_malaria_2_week + mean_malaria_5_year + mean_worried_malaria + true_age', risk).fit().summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "with sns.color_palette('Set2'):\n",
    "    ax = sns.barplot(x='worried', y='worried_covid', data=risk.assign(worried=3 - ordinalize(risk.worried_malaria)))\n",
    "    ax.set_xlabel('Worried about malaria, on a scale of 0-3')\n",
    "    ax.set_ylabel('Worried about Covid-19')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "with sns.color_palette('Set2'):\n",
    "    ax = sns.barplot(x='worried', y='home', hue='education', data=risk.assign(worried=3 - ordinalize(risk.worried_malaria)))\n",
    "    ax.set_xlabel('Worried about malaria, on a scale of 0-3')\n",
    "    ax.set_ylabel('Home Type')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "with sns.color_palette('Set2'):\n",
    "    ax = sns.barplot(x='malaria_now', y='home', hue='education', data=dd)\n",
    "    ax.set_xlabel('Malaria incidence in the last 2 weeks')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# People are potentially not as worried as they should be\n",
    "\n",
    "with sns.color_palette('BuGn_r'):\n",
    "    sns.countplot(y='worried_malaria', data=risk)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "\n",
    "smf.ols('ordinalize(worried_malaria) ~ risk_rf*ordinalize(education)', risk).fit().summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "smf.logit('integerize(under_net) ~ ordinalize(worried_malaria)', risk[risk.net_ownership]).fit().summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# little correlation between worried and mean_malaria_2_week\n",
    "\n",
    "smf.ols('worried_malaria ~ kutcha*mean_malaria_2_week', risk.assign(worried_malaria = risk.worried_malaria.cat.codes)).fit().summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd.worried_malaria"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "smf.ols('mean_malaria_5_year ~ high_risk', risk.assign(worried_malaria = risk.worried_malaria.cat.codes)).fit().summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "risk.groupby('high_risk_dist').worried_malaria.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd.groupby('disthash').apply(lambda df: pd.Series({'not_worried': 1.5 - df.worried_malaria.cat.codes.mean(), 'malaria_now': df.malaria_now.mean()})).plot.scatter(x='not_worried', y='malaria_now')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 864,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f81c5f0efd0>"
      ]
     },
     "execution_count": 864,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.regplot(x='worried', y='under_net', data=dd[(dd.disthash.isin(final_clusters)) & dd.net_ownership].groupby('disthash').apply(lambda df: pd.Series({'worried': df.worried_malaria.cat.codes.mean(), 'under_net': df.under_net.mean()})))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sns.barplot(y='education', x='malaria_now', data=dd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sns.countplot(y='education', hue='worried_malaria', data=dd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 772,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Never attended school</th>\n",
       "      <td>0.039851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Primary</th>\n",
       "      <td>0.023821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Secondary</th>\n",
       "      <td>0.016226</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>University degree or higher</th>\n",
       "      <td>0.015638</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                count\n",
       "education                            \n",
       "Never attended school        0.039851\n",
       "Primary                      0.023821\n",
       "Secondary                    0.016226\n",
       "University degree or higher  0.015638"
      ]
     },
     "execution_count": 772,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "malaria_prob(dd.groupby('education'), 'malaria_now')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 636,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "worried_malaria\n",
       "Extremely worried     0.456867\n",
       "Worried               0.445308\n",
       "Somewhat worried      0.444504\n",
       "Not at all worried    0.447429\n",
       "Name: risk_rf, dtype: float64"
      ]
     },
     "execution_count": 636,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "risk.groupby('worried_malaria').risk_rf.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 637,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "worried_malaria\n",
       "Extremely worried     0.018330\n",
       "Worried               0.017148\n",
       "Somewhat worried      0.017071\n",
       "Not at all worried    0.017544\n",
       "Name: risk_log, dtype: float64"
      ]
     },
     "execution_count": 637,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "risk.groupby('worried_malaria').risk_log.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 537,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "under_net\n",
       "False    0.437084\n",
       "True     0.446863\n",
       "Name: risk_rf, dtype: float64"
      ]
     },
     "execution_count": 537,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "\n",
    "risk.groupby('under_net').risk_rf.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 544,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "under_net\n",
       "False    0.016907\n",
       "True     0.018148\n",
       "Name: risk_log, dtype: float64"
      ]
     },
     "execution_count": 544,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "risk.groupby('under_net').risk_log.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 558,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0.5, 0.8237569617822164)]"
      ]
     },
     "execution_count": 558,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.barplot(x='net_ownership', y='home', data=dd, color=base_color)\n",
    "ax.set(xlim=(.5, None))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 546,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "under_net\n",
       "False    0.016898\n",
       "True     0.018008\n",
       "Name: risk_log, dtype: float64"
      ]
     },
     "execution_count": 546,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "risk[risk.net_ownership].groupby('under_net').risk_log.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 547,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "longsleeves\n",
       "No     0.017544\n",
       "Yes    0.017625\n",
       "Name: risk_log, dtype: float64"
      ]
     },
     "execution_count": 547,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "risk.groupby('longsleeves').risk_log.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 540,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "net_ownership\n",
       "False    0.017320\n",
       "True     0.017682\n",
       "Name: risk_log, dtype: float64"
      ]
     },
     "execution_count": 540,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "risk.groupby('net_ownership').risk_log.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 448,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dist_dd = dd.groupby('disthash').head(1).reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "smf.ols('mean_malaria_2_week ~ mean_malaria_5_year - 1', dist_dd).fit().summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 433,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "y = smf.ols('mean_malaria_2_week ~ mean_malaria_5_year - 1', dist_dd).fit().resid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dist_dd.mean_net_ownership"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 492,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worried_malaria</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Extremely worried</th>\n",
       "      <td>0.203999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Worried</th>\n",
       "      <td>0.222588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Somewhat worried</th>\n",
       "      <td>0.251194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Not at all worried</th>\n",
       "      <td>0.162203</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       count\n",
       "worried_malaria             \n",
       "Extremely worried   0.203999\n",
       "Worried             0.222588\n",
       "Somewhat worried    0.251194\n",
       "Not at all worried  0.162203"
      ]
     },
     "execution_count": 492,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "malaria_prob(dd.groupby('worried_malaria'), 'malaria')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 496,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>worried_malaria</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Extremely worried</th>\n",
       "      <td>0.024888</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Worried</th>\n",
       "      <td>0.022661</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Somewhat worried</th>\n",
       "      <td>0.019387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Not at all worried</th>\n",
       "      <td>0.011648</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       count\n",
       "worried_malaria             \n",
       "Extremely worried   0.024888\n",
       "Worried             0.022661\n",
       "Somewhat worried    0.019387\n",
       "Not at all worried  0.011648"
      ]
     },
     "execution_count": 496,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "malaria_prob(dd.groupby('worried_malaria'), 'malaria_now')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 488,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f81d09d9580>"
      ]
     },
     "execution_count": 488,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot(y=dist_dd.mean_net_ownership, x=dist_dd.mean_malaria_5_year, color=base_color)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 721,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "under_net\n",
       "False    0.463426\n",
       "True     0.463493\n",
       "Name: risk_rf, dtype: float64"
      ]
     },
     "execution_count": 721,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "risk.groupby(['under_net']).risk_rf.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 723,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "home                                     dist_risk     \n",
       "Kutcha (made of mud, tin, straw)         Upper Quartile    0.462676\n",
       "                                         Interquartile     0.468106\n",
       "                                         Lower Quartile    0.462302\n",
       "Semi-pucca                               Upper Quartile    0.464711\n",
       "                                         Interquartile     0.462761\n",
       "                                         Lower Quartile    0.460111\n",
       "Pucca (have cement/brick wall and floor  Upper Quartile    0.463952\n",
       "                                         Interquartile     0.461663\n",
       "                                         Lower Quartile    0.464765\n",
       "Name: risk_rf, dtype: float64"
      ]
     },
     "execution_count": 723,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "risk.groupby(['home', 'dist_risk']).risk_rf.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1378,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.4691648752336979\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='malaria_risk', ylabel='under_net'>"
      ]
     },
     "execution_count": 1378,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 700x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "var = 'under_net'\n",
    "\n",
    "dat = dd[dd.net_ownership].groupby(['home', 'dist_risk', 'dist_medical']).apply(lambda df: pd.Series({'malaria_risk': df.malaria_now.mean(), \n",
    "                                                                                                      var: df[var].mean()})).reset_index()\n",
    "\n",
    "\n",
    "print(dat.malaria_risk.corr(dat[var]))\n",
    "sns.scatterplot(x='malaria_risk', y=var, data=dat, color=base_color)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "with sns.color_palette('Set2'):\n",
    "    fig, ax = plt.subplots(figsize=(10, 4))\n",
    "    sns.barplot(x='net_ownership', y='dist_risk', hue='occupation', data=risk, ax=ax)\n",
    "    ax.set_xlabel('Net Ownership')\n",
    "    ax.set_ylabel('District Malaria Incidence')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1052,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.cluster import DBSCAN\n",
    "\n",
    "dat = dd[['household', 'dist_risk', 'net_ownership', 'ac', 'dist_medical']].copy()\n",
    "dat['caste']  = 1\n",
    "dat.loc[dd.lower_caste == True, 'caste'] = 0\n",
    "\n",
    "ordenals = ['dist_risk', 'dist_medical']\n",
    "\n",
    "for col in ordenals:\n",
    "    dat[col] = dat[col].cat.codes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier, DecisionTreeRegressor\n",
    "from dtreeviz.trees import *\n",
    "\n",
    "cls = DecisionTreeClassifier(max_depth=5, min_samples_leaf=100, min_impurity_decrease=.00005)\n",
    "X = dd.copy()\n",
    "\n",
    "y = X['net_ownership']\n",
    "X = X[['household', 'home', 'ac', 'lower_caste', 'dist_risk', 'student', 'male']]\n",
    "X = X.dropna()\n",
    "y = y[X.index]\n",
    "\n",
    "ordenals = ['home', 'dist_risk']\n",
    "for col in ordenals:\n",
    "    X[col] = X[col].cat.codes\n",
    "\n",
    "cls.fit(X, y)\n",
    "\n",
    "\n",
    "viz = dtreeviz(cls,\n",
    "               X, \n",
    "               y,\n",
    "               target_name='owns net',\n",
    "               feature_names=X.columns, \n",
    "               class_names=['no', 'yes']\n",
    "              )  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "31       False\n",
       "32       False\n",
       "33        True\n",
       "34       False\n",
       "35        True\n",
       "         ...  \n",
       "21505    False\n",
       "21506    False\n",
       "21507    False\n",
       "21508     True\n",
       "21509     True\n",
       "Name: lives_alone, Length: 20288, dtype: bool"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.lives_alone"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd.groupby(['lives_alone']).male.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd.groupby(['lives_alone', 'student']).male.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd.groupby(['lives_alone', 'male']).net_ownership.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9225157728706624"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.male.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "print(f'N = {X.shape[0]}')\n",
    "\n",
    "viz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from dtreeviz.trees import *\n",
    "\n",
    "\n",
    "cls = DecisionTreeClassifier(max_depth=5, min_samples_leaf=100, min_impurity_decrease=.0001)\n",
    "X = dd[(dd.net_ownership == True) & (dd.mean_malaria_now > dd.mean_malaria_now.quantile(0.5))].copy()\n",
    "\n",
    "y = X['under_net']\n",
    "X = X[['household', 'home', 'ac', 'lower_caste', 'true_age', 'student', 'male']]\n",
    "X = X.dropna()\n",
    "y = y[X.index]\n",
    "\n",
    "ordenals = ['home']\n",
    "for col in ordenals:\n",
    "    X[col] = X[col].cat.codes\n",
    "\n",
    "cls.fit(X, y)\n",
    "\n",
    "\n",
    "viz = dtreeviz(cls,\n",
    "               X, \n",
    "               y,\n",
    "               target_name='under_net',\n",
    "               feature_names=X.columns, \n",
    "               class_names=['no net', 'slept with net']\n",
    "              )  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "print(f'N = {X.shape[0]}')\n",
    "viz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1535,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2079"
      ]
     },
     "execution_count": 1535,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat = dd[(dd.net_ownership) & (dd.under_net == False) & (dd.mean_malaria_now > dd.mean_malaria_now.quantile(0.5))].reset_index(drop=True)\n",
    "\n",
    "N = dat.shape[0]\n",
    "N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1536,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.30822510822510824"
      ]
     },
     "execution_count": 1536,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    ".9 * dat[(dat.lives_alone) & (dat.ac == False) & (~dat.kutcha)].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1537,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17.0"
      ]
     },
     "execution_count": 1537,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(dat.lives_alone) & (dat.ac == False) & (~dat.kutcha)].true_age.quantile(.05)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1538,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "37.0"
      ]
     },
     "execution_count": 1538,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(dat.lives_alone) & (dat.ac == False) & (~dat.kutcha)].true_age.quantile(.9)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pucca (have cement/brick wall and floor    0.492514\n",
       "Semi-pucca                                 0.258471\n",
       "Kutcha (made of mud, tin, straw)           0.249015\n",
       "Name: home, dtype: float64"
      ]
     },
     "execution_count": 1425,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.home.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1424,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pucca (have cement/brick wall and floor    0.545151\n",
       "Semi-pucca                                 0.248606\n",
       "Kutcha (made of mud, tin, straw)           0.206243\n",
       "Name: home, dtype: float64"
      ]
     },
     "execution_count": 1424,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(dat.lives_alone) & (dat.ac == False)].home.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1418,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "556"
      ]
     },
     "execution_count": 1418,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(~dat.lives_alone) & (dat.ac == False) & (dat.lower_caste)].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1436,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2674362674362674"
      ]
     },
     "execution_count": 1436,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(~dat.lives_alone) & (dat.ac == False) & (dat.lower_caste)].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1442,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.37373737373737376"
      ]
     },
     "execution_count": 1442,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(~dat.lives_alone) & (dat.ac == False) & (~dat.pucca | dat.lower_caste)].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1476,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "36.0"
      ]
     },
     "execution_count": 1476,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.true_age.quantile(.9)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1539,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.28744588744588745"
      ]
     },
     "execution_count": 1539,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    ".9 * dat[(~dat.lives_alone) & (dat.ac == False) & (dat.kutcha | dat.lower_caste)].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1552,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.11951788491446345"
      ]
     },
     "execution_count": 1552,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd[~dd.lives_alone].large_family.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1550,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.05555555555555555"
      ]
     },
     "execution_count": 1550,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(~dat.lives_alone)].large_family.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1546,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.058734939759036146"
      ]
     },
     "execution_count": 1546,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(~dat.lives_alone) & (dat.ac == False) & (dat.kutcha | dat.lower_caste)].large_family.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1553,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.27056277056277056"
      ]
     },
     "execution_count": 1553,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    ".9 * dat[(~dat.lives_alone) & (dat.ac == False) & (dat.kutcha | dat.lower_caste) & (~dat.large_family)].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1463,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "18.0"
      ]
     },
     "execution_count": 1463,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(~dat.lives_alone) & (dat.ac == False) & (dat.kutcha | dat.lower_caste)].true_age.quantile(.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1477,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "29.0"
      ]
     },
     "execution_count": 1477,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(~dat.lives_alone) & (dat.ac == False) & (dat.kutcha)].true_age.quantile(.9)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1444,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.16354016354016354"
      ]
     },
     "execution_count": 1444,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(~dat.lives_alone) & (dat.ac == False) & (dat.lower_caste) & (~dat.pucca)].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1422,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pucca (have cement/brick wall and floor    461\n",
       "Semi-pucca                                 312\n",
       "Kutcha (made of mud, tin, straw)           289\n",
       "Name: home, dtype: int64"
      ]
     },
     "execution_count": 1422,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(~dat.lives_alone)].home.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1390,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1017"
      ]
     },
     "execution_count": 1390,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dat[(dat.lives_alone)].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dat[(dat.lives_alone)].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1394,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "lives_alone  large_family  home                                     lower_caste\n",
       "True         False         Pucca (have cement/brick wall and floor  True           255.0\n",
       "                                                                    False          234.0\n",
       "False        False         Pucca (have cement/brick wall and floor  True           202.0\n",
       "                                                                    False          181.0\n",
       "                           Semi-pucca                               True           165.0\n",
       "                           Kutcha (made of mud, tin, straw)         True           156.0\n",
       "True         False         Semi-pucca                               True           136.0\n",
       "False        False         Semi-pucca                               False          109.0\n",
       "True         False         Kutcha (made of mud, tin, straw)         True           107.0\n",
       "False        False         Kutcha (made of mud, tin, straw)         False          102.0\n",
       "True         False         Semi-pucca                               False           87.0\n",
       "                           Kutcha (made of mud, tin, straw)         False           78.0\n",
       "False        True          Pucca (have cement/brick wall and floor  True            14.0\n",
       "                           Semi-pucca                               True            12.0\n",
       "                           Pucca (have cement/brick wall and floor  False           12.0\n",
       "                           Kutcha (made of mud, tin, straw)         True             7.0\n",
       "                                                                    False            6.0\n",
       "                           Semi-pucca                               False            4.0\n",
       "True         True          Kutcha (made of mud, tin, straw)         False            NaN\n",
       "                                                                    True             NaN\n",
       "                           Semi-pucca                               False            NaN\n",
       "                                                                    True             NaN\n",
       "                           Pucca (have cement/brick wall and floor  False            NaN\n",
       "                                                                    True             NaN\n",
       "Name: userid, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with pd.option_context('display.max_rows', None, 'display.expand_frame_repr', False):\n",
    "    display(dd[(dd.ac == False) & (dd.net_ownership) & (dd.under_net == False) & (dd.mean_malaria_now > dd.mean_malaria_now.quantile(0.5))].groupby(['lives_alone', 'large_family', 'home', 'lower_caste']).count().userid.sort_values(ascending=False))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "with sns.color_palette(sns.cubehelix_palette(3, start=1.25, rot=.0, dark=0.8, light=.35)):\n",
    "    sns.barplot(y='dist_risk', hue='home', x='under_net', data= dd[dd.net_ownership])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1520,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "N = dd[~dd.net_ownership].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1521,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "48.0"
      ]
     },
     "execution_count": 1521,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd[~dd.net_ownership & (dd.lives_alone == True) & (dd.student == False) & (dd.ac == False)].true_age.quantile(.95)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1492,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Male      0.923614\n",
       "Female    0.072971\n",
       "Other     0.003415\n",
       "Name: gender, dtype: float64"
      ]
     },
     "execution_count": 1492,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.gender.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1523,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4472804190169219"
      ]
     },
     "execution_count": 1523,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    ".9 * dd[~dd.net_ownership & (dd.lives_alone == True) & (dd.student == False) & (dd.ac == False) & (dd.gender)].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1524,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2127316680096696"
      ]
     },
     "execution_count": 1524,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd[~dd.net_ownership & (dd.lives_alone == False) & (dd.large_family == False) & (dd.ac == False)].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1525,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.13215149073327961"
      ]
     },
     "execution_count": 1525,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd[~dd.net_ownership &  (dd.lives_alone == False) & (dd.ac == False) & (dd.large_family == False) & ((~dd.pucca))].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1526,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.11059629331184528"
      ]
     },
     "execution_count": 1526,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd[~dd.net_ownership &  (dd.lives_alone == False) & (dd.ac == False) & (dd.large_family == False) & ((dd.lower_caste))].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd.home.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd[~dd.net_ownership &  (dd.lives_alone == False) & (dd.ac == False) & (dd.large_family == False) & (~(dd.lower_caste))].home.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1533,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16.0"
      ]
     },
     "execution_count": 1533,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd[~dd.net_ownership &  (dd.lives_alone == False) & (dd.ac == False) & (dd.large_family == False) & ((dd.kutcha) | (dd.lower_caste))].true_age.quantile(.05)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1534,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.14645447219983884"
      ]
     },
     "execution_count": 1534,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd[~dd.net_ownership &  (dd.lives_alone == False) & (dd.ac == False) & (dd.large_family == False) & ((dd.kutcha) | (dd.lower_caste))].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2127316680096696"
      ]
     },
     "execution_count": 1500,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd[~dd.net_ownership &  (dd.lives_alone == False) & (dd.ac == False) & (dd.large_family == False)].shape[0] / N"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd[~dd.net_ownership & (dd.lives_alone == False) & (dd.ac == False) & (dd.household < 10)].household.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "with pd.option_context('display.max_rows', None, 'display.expand_frame_repr', False):\n",
    "    display(dd[~dd.net_ownership].groupby(['lives_alone', 'large_family', 'home', 'ac', 'lower_caste', 'student']).count().userid.sort_values(ascending=False))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd[(dd.under_net == False) & (dd.dist_risk == 'Upper Quartile')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# 33% report living alone\n",
    "sns.distplot(dat, kde=False, bins=20, color=base_color, hist_kws={'alpha': 1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Interquartile range (middle 50%) between 21-29 years old\n",
    "sns.distplot(dd.true_age, color=base_color, kde=False, hist_kws={'alpha': 1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# 59% have a university degree or higher\n",
    "sns.countplot(y='education', data=dd, color=base_color)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# 45% report being unemployed, 25% report being students\n",
    "sns.countplot(y='occupation', data=dd, color=base_color)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "with sns.color_palette(sns.cubehelix_palette(6, start=1.5, rot=-1, dark=0.3, light=.8)):\n",
    "    fig, ax = plt.subplots(figsize=(10, 4))\n",
    "    sns.barplot(x='net_ownership', y='dist_risk_5_year', hue='education', data=risk, ax=ax)\n",
    "    ax.set_xlabel('Net Ownership')\n",
    "    ax.set_ylabel('District Malaria Incidence')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1740,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>net_ownership</th>\n",
       "      <th>malaria_now</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>net_ownership</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.047535</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>malaria_now</th>\n",
       "      <td>0.047535</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               net_ownership  malaria_now\n",
       "net_ownership       1.000000     0.047535\n",
       "malaria_now         0.047535     1.000000"
      ]
     },
     "execution_count": 1740,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.groupby('disthash').agg({'net_ownership': np.mean, 'malaria_now': np.mean}).corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd.groupby('disthash').agg({'net_ownership': np.mean, 'malaria_now': np.mean, 'malaria': np.mean}).plot.scatter(x='net_ownership', y='malaria_now')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "ax = sns.distplot(dd.groupby('disthash').net_ownership.mean(), color=base_color, bins=20, hist_kws={'alpha': 1}, kde=False)\n",
    "ax.set_xlabel('Mean net ownership by district')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1353,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "lives_alone  student\n",
       "False        False      0.892333\n",
       "             True       0.902987\n",
       "True         False      0.391729\n",
       "             True       0.489041\n",
       "Name: net_ownership, dtype: float64"
      ]
     },
     "execution_count": 1353,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.groupby(['lives_alone', 'student']).net_ownership.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1352,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "large_family  lives_alone\n",
       "False         False          0.895522\n",
       "              True           0.414737\n",
       "True          False          0.891347\n",
       "Name: net_ownership, dtype: float64"
      ]
     },
     "execution_count": 1352,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.groupby(['large_family', 'lives_alone']).net_ownership.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1349,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8116090065284007"
      ]
     },
     "execution_count": 1349,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.groupby('disthash').net_ownership.mean().quantile(.75)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(figsize=(10, 3))\n",
    "sns.barplot(x='net_ownership', y='dist_risk', data=risk, ax=ax, color=base_color)\n",
    "ax.set_xlabel('Net Ownership')\n",
    "ax.set_ylabel('District Malaria Incidence')\n",
    "ax.set_xlim(0.5, None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "with sns.color_palette(sns.cubehelix_palette(6, start=1.5, rot=-1, dark=0.3, light=.8)):\n",
    "    fig, ax = plt.subplots(figsize=(10, 4))\n",
    "    sns.barplot(x='net_ownership', y='dist_risk', hue='education', data=risk, ax=ax)\n",
    "    ax.set_xlabel('Net Ownership')\n",
    "    ax.set_ylabel('District Malaria Incidence')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# likelihood of sleeping under a net, given that they have one, rather consistent\n",
    "with sns.color_palette(sns.cubehelix_palette(6, start=1.5, rot=-1, dark=0.3, light=.8)):\n",
    "    sns.barplot(x='under_net', y='dist_risk', hue='education', data=risk[risk.net_ownership])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dat = dd.groupby(['home', 'dist_risk', 'dist_medical']).apply(lambda df: pd.Series({'malaria_risk': df.malaria_now.mean(), \n",
    "                                                                                    var: df[var].mean()})).reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 727,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "home                                     dist_risk     \n",
       "Kutcha (made of mud, tin, straw)         Upper Quartile    0.748107\n",
       "                                         Interquartile     0.703237\n",
       "                                         Lower Quartile    0.704610\n",
       "Semi-pucca                               Upper Quartile    0.777157\n",
       "                                         Interquartile     0.763393\n",
       "                                         Lower Quartile    0.773256\n",
       "Pucca (have cement/brick wall and floor  Upper Quartile    0.752780\n",
       "                                         Interquartile     0.735589\n",
       "                                         Lower Quartile    0.715802\n",
       "Name: net_ownership, dtype: float64"
      ]
     },
     "execution_count": 727,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.groupby(['home', 'dist_risk']).net_ownership.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd[dd.net_ownership].groupby(['home', 'dist_risk']).longsleeves.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd[dd.net_ownership].groupby(['home', 'dist_risk']).used_spray.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd[dd.net_ownership].groupby(['home', 'dist_risk']).under_net"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sns.barplot('under_net', 'dist_risk', hue='home', data=dd[dd.net_ownership])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "risk.groupby(['under_net']).risk_rf.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 483,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "home                                     net_ownership\n",
       "Kutcha (made of mud, tin, straw)         False            1405\n",
       "                                         True             3763\n",
       "Semi-pucca                               False            1178\n",
       "                                         True             4029\n",
       "Pucca (have cement/brick wall and floor  False            2549\n",
       "                                         True             7258\n",
       "Name: surveyid, dtype: int64"
      ]
     },
     "execution_count": 483,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.groupby(['home', 'net_ownership']).count().surveyid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 481,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>home</th>\n",
       "      <th>net_ownership</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">Kutcha (made of mud, tin, straw)</th>\n",
       "      <th>False</th>\n",
       "      <td>0.155160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <td>0.748339</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">Semi-pucca</th>\n",
       "      <th>False</th>\n",
       "      <td>0.078947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <td>0.709109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">Pucca (have cement/brick wall and floor</th>\n",
       "      <th>False</th>\n",
       "      <td>0.067085</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <td>0.682557</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                          count\n",
       "home                                    net_ownership          \n",
       "Kutcha (made of mud, tin, straw)        False          0.155160\n",
       "                                        True           0.748339\n",
       "Semi-pucca                              False          0.078947\n",
       "                                        True           0.709109\n",
       "Pucca (have cement/brick wall and floor False          0.067085\n",
       "                                        True           0.682557"
      ]
     },
     "execution_count": 481,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "malaria_prob(dd.groupby(['home', 'net_ownership']), 'under_net')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 436,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f81c6dd7190>"
      ]
     },
     "execution_count": 436,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.regplot(y=dist_dd.mean_malaria_5_year, x=dist_dd.mean_under_net)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 434,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f81c683fd00>"
      ]
     },
     "execution_count": 434,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEHCAYAAACwUAEWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nO3dfZxcdX3o8c93HvYxm+wm2eUhCYSFQEzwAVwDKsaIiqiVWKUKWosWS+yV6r22vVJrow16K33d6sWCbaJyi2gB5fXyun2pTQXcRpRIFihqMJKwBDYhYfOw2exmd3aevvePc2b3zGT27DzuzOx8369XsjNnzpz5zsP5fc/v4fyOqCrGGGPMTAKVDsAYY0x1s0RhjDHGlyUKY4wxvixRGGOM8WWJwhhjjK9QpQMoxNKlS3XlypWVDsMYY2rK448/flRVO/N9Xk0mipUrV9Lf31/pMIwxpqaIyPOFPM+anowxxviyRGGMMcaXJQpjjDG+LFEYY4zxZYnCGGOMr5oc9WRMPvr2DLF1xwCDw+Os6Ghh0/puNqzuqnRYxtQMq1GYea1vzxCbe3czNBqhvTnM0GiEzb276dszVOnQjKkZlijMvLZ1xwDhoNDSEELE+RsOClt3DFQ6NGNqhiUKM68NDo/THA6mLWsOBzkwPF6hiIypPZYozLy2oqOFiVgibdlELMHyjpYKRWRM7bFEYea1Teu7iSWU8WgcVedvLKFsWt9d6dCMqRmWKMy8tmF1F1uuWUtXWxMjEzG62prYcs1aG/VkTB5seKyZ9zas7rLEYEwRrEZhjDHGlyUKY4wxvixRGGOM8WWJwhhjjC9LFMYYY3xZojDGGOPLEoUxxhhfJUkUInK1iPxORPaJyC1ZHm8Ukfvdx38pIis9j71CRB4Vkd0i8msRaSpFTMYYY0qj6EQhIkHgTuDtwBrgehFZk7HajcCwql4AfAW4zX1uCPg28DFVXQtsAGLFxmSMMaZ0SlGjWAfsU9UBVY0C9wEbM9bZCNzt3n4AeLOICHAV8CtVfQpAVY+pagJjjDFVoxSJYhkw6Ll/wF2WdR1VjQMjwBLgQkBFZLuIPCEi/3OmFxGRm0SkX0T6jxw5UoKwjTHG5KLSndkh4Argg+7f3xeRN2dbUVW3qWqPqvZ0dnbOZYzGGFPXSpEoDgIrPPeXu8uyruP2SywCjuHUPnao6lFVHQd+BFxagpiMMcaUSCkSxS5glYicJyINwHVAb8Y6vcAN7u1rgYdVVYHtwMtFpMVNIG8Eni5BTMYYY0qk6GnGVTUuIjfjFPpB4C5V3S0iW4B+Ve0FvgncIyL7gOM4yQRVHRaRL+MkGwV+pKo/LDYmY4wxpSPOgX1t6enp0f7+/kqHYYwxNUVEHlfVnnyfV+nObGOMMVXOEoUxxhhfliiMMcb4skRhjDHGlyUKY4wxvixRGGOM8WWJwhhjjC9LFMYYY3xZojDGGOPLEoUxxhhfliiMMcb4skRhjDHGlyUKY4wxvixRGGOM8WWJwhhjjC9LFMYYY3wVfYU7U3l9e4bYumOAweFxVnS0sGl9NxtWd1U6LGPMPGE1ihrXt2eIzb27GRqN0N4cZmg0wube3fTtGap0aMaYecISRY3bumOAcFBoaQgh4vwNB4WtOwYqHZoxZp6wRFHjBofHaQ4H05Y1h4McGB6vUETGmPnGEkWNW9HRwkQskbZsIpZgeUdLhSIyxsw3lihq3Kb13cQSyng0jqrzN5ZQNq3vrnRoxph5whJFjduwuost16ylq62JkYkYXW1NbLlmrY16MsaUjA2PnQc2rO6yxGCMKRurURhjjPFlNQpjTE7sxM76ZTUKY8ys7MTO+laSRCEiV4vI70Rkn4jckuXxRhG53338lyKyMuPxc0RkTET+ohTxGGNKy07srG9FJwoRCQJ3Am8H1gDXi8iajNVuBIZV9QLgK8BtGY9/GfhxsbEYY8rDTuysb6WoUawD9qnqgKpGgfuAjRnrbATudm8/ALxZRARARN4NPAfsLkEsxpgysBM761spEsUyYNBz/4C7LOs6qhoHRoAlIrIA+DTwt7O9iIjcJCL9ItJ/5MiREoRtjMmVndhZ3yrdmf154CuqOjbbiqq6TVV7VLWns7Oz/JEZY6bYiZ31rRTDYw8CKzz3l7vLsq1zQERCwCLgGHAZcK2I/D3QDiRFJKKqd5QgLmNMCdmJnfWrFIliF7BKRM7DSQjXAR/IWKcXuAF4FLgWeFhVFXhDagUR+TwwZknCGGOqS9GJQlXjInIzsB0IAnep6m4R2QL0q2ov8E3gHhHZBxzHSSbGGGNqgDgH9rWlp6dH+/v7Kx2GMcbUFBF5XFV78n1epTuzjTHGVDlLFMYYY3xZojDGGOPLEoUxxhhfliiMMcb4skRhjDHGlyUKY4wxvixRGGOM8WWJwhhjjC9LFMYYY3xZojDGGOPLEoUxxhhfliiMMcb4skRhjDHGlyUKY4wxvixRGGOM8WWJwhhjjC9LFMYYY3xZojDGGOPLEoUxxhhfoUoHYIypbn17hti6Y4DB4XFWdLSwaX03G1Z3VTosM4esRmGMmVHfniE29+5maDRCe3OYodEIm3t307dnqNKhmTlkicIYM6OtOwYIB4WWhhAizt9wUNi6Y6DSoZk5ZInCGDOjweFxmsPBtGXN4SAHhscrFJGpBEsUxpgZrehoYSKWSFs2EUuwvKOlQhGZSrBEYYyZ0ab13cQSyng0jqrzN5ZQNq3vrnRoZg5ZojDGzGjD6i62XLOWrrYmRiZidLU1seWatTbqqc7Y8FhjjK8Nq7ssMdS5ktQoRORqEfmdiOwTkVuyPN4oIve7j/9SRFa6y98qIo+LyK/dv1eWIp5q0rdniOu37eSK2x7m+m07bVihMabmFJ0oRCQI3Am8HVgDXC8iazJWuxEYVtULgK8At7nLjwLvUtWXAzcA9xQbTzWxMejGmPmgFDWKdcA+VR1Q1ShwH7AxY52NwN3u7QeAN4uIqOqTqvqiu3w30CwijSWIqSrYGHRjzHxQij6KZcCg5/4B4LKZ1lHVuIiMAEtwahQp7wWeUNXJbC8iIjcBNwGcc845JQi7/AaHx2lvDqctszHoptrZlB0mU1WMehKRtTjNUZtmWkdVt6lqj6r2dHZ2zl1wRbAx6KbWWHOpyaYUieIgsMJzf7m7LOs6IhICFgHH3PvLge8Df6Sqz5YgnqphY9BNrbHmUpNNKZqedgGrROQ8nIRwHfCBjHV6cTqrHwWuBR5WVRWRduCHwC2q+vMSxFJVNqzuYgvOzndgeJzlJa7GWxOBKbVyNpfa77V2FZ0o3D6Hm4HtQBC4S1V3i8gWoF9Ve4FvAveIyD7gOE4yAbgZuADYLCKb3WVXqeq8qeeWawx6qokgHJS0JoJrD5zg0YHjtjOagqzoaGFoNEJLw3TRUIrm0pl+r1vAfp81QFS10jHkbcHyi/Rdm++u60Lw+m07T9uhj4xGGB6PsbyjmeZwkIlYglhC7UxakzNvgV7K31C23+t4NE5XWxP33nR5KUI3ORCRx1W1J9/nVUVndr5CAan7TrZss3qORuLEk0lrXzYFK9eUHTYLbW2r2Sk8WhpCjEfjbN0xUPNHy4W03WZrIpiMJ2kMpud+2xlNNn6/uXI0l5arScvMjZqsUaTMh0Kw0OGI2UZUBQPCopb0jshy7Yw2NUntqsQQWBsBWNtqtkYBtXlEknkkd2I8OjUcEXKvKWUbUbXxlWfzwBMHGY/G09qXS70z1mLHpI24meYdAgtzUzsv9whAU141myhq8YgkWwG7/9g4y9ub0tbLtaaUrYngFcvby74zVqKgKUYtJrZyqtSMATYLbe2qyUSRSCpdbU01d0SSrYANB4WXTk6ysLlhar2Zakq5HBXPxc5YSEFTySP6Wkts5Wb9BSZfNdlHcdGZbdx70+U1t5NnG/lxRlsjsWRy1rbbappaId+pSSod+3wacVOKviFvf8HJiSh7h0bZf2ycE+NR62syWdVkoqhV2QrYUDDAqs4Fsw5HrKapFfLtmKx07PNlzq1SJdzUENhwQDhwYgIUlrc3EU0k63rIuZlZTTY91apN67vZ3Lv7tM7mv3nnmllrR9U0E22+HZOVjn2mz70cnfzlbF4rZRPahtVdbN0xwMqkTm3v5ESModEIm779OJee01FzTbv1SFVJJJVE6m9SSSZJu59QJZlU2poKL+4tUcyhYkZ+VFu7cj59IYXGXqqCdy5G3KSO9qPxBKOROIdHIjzxwjAf33A+n3jLhQVtL/O9lzrherd3ciLGiyMTCJBUrfsO/0rxFvxTBX5iOhEkvQnAvZ/7tguPyxLFHCu0s3mujopLIbOQe2334ryH7ZZ6pFIhn3s+iWrrjgGi8QTHTkUJIIQCQkKVO/ue5RXL2/N67Zne+4IG57Mr1cGCN4EfHZskgIBAYzBQ9x3+pZDtaH+6gHcScqrgTyan71cjSxQ1oprGofsVoNkKuQeeOMi1ly7j0YHjOcdejpFK+RT8+SaqweFxRiNxAgiBgAAQFIglknnHPNN7FxFiiWTJDha8Bx/RRBIBUKGzzbnIZK12+JeTX+GfWh5PTCeA+cISRQ2Zy3HoMxWqsxWgMxVyjw4cz2vyt1I3s+Rb8OebqFZ0tHB4JELITRLgVPUbg4G8Y57pvY9MxLh148UlO1hIP/hwmp3OXNREW5Pz2rXY4Z8PVefIPtWEE59q459u108lBPWsV4/qPlHYGbun8ytUZytAS1HA9+0Z4uREjEMjEzSFgnS2NdLWFC6q4Mq34M/3fWxa380TLwyTUCUoTpJQhUWt4bxj9uvTKfXBQmp7qe88GBBUtaqbNmeSebQfTxX6GW3+yQLa9+tdXScKO2M3O79CdbYCtNhO99R30uK2x0cTSQ4OT7C0LUE4GCy44Mq34M/3fWxY3cXHN5zPnX3PEks4kzMuag0XFHMl+qOqqWkzRT1H+VMJIDF9tJ9q5/d28pryqOtEYWfsZudXqM5WgBZbyKW+k0XNTTSGghwdm2QynuTUZIKvXveKgjukj4xOcnR0MuemlULexyfecmHWKVTAuR5DrrXWShXac9G0OVsbvzcx1MoR/2MDx7lv1yCHTk5w1sJmrnvNCtZ1L650WCVV14mi0uP7q5VfMpitAC22kPN+JwubwyxsDqOqjEzEpppICumQPnNhIwdPRDgwPMGydiUUDPgW/IW+j8zCttBaay3Ni5Q5pDOeTE4186Rux5PJqQQwnzw2cJzbH95LKCAsbApx7NQktz+8l0+yal4li7pOFNV2bkKpFNvv4pcMcilAZyvk/OLzficnJ2JTNYqWhiBfffAZHnjiYMEd0iC8NBrh8MnJnE4oK0VhXYu11nKO5a+Uch3137drkFBApqaISe0v9+0atEQxX9TSuQm5KkW/izcZ7B0aJRpPpk25UUwBOlt8qe/kyGiEY6eiAAjQ2hjkzr5nWdwaZlGzM9tuvh3SC5vDtDWFGJmIzdnlNytZa81a4Cc1bbSP6vwf2VPOo/5DJydYmHHGc1M4wOGTE0Vtt9rUZKKYjCd57ugpAgKCIAKBgCBAQJz74j4WEO8y537q72XdS/jsO17GN3/+HAfdo+ObanzUU6mOYFPrbu7dzaJm54ipFJ39s8WXSlKfuO9Jkqppo56OjEYZGY+xdMH0tOyl7JD2KtVouEJiUHUK7aROd9im7idU0dTondQ/z8la3ufpPCvwC1XOo/6zFjZz7NRk2qSTkViSMxc2F7XdalOTiQLcoyUFcHeGhN/aM7vgjAX83XtenrZs4MgYAZGpBBMIOIklKIKIEHTvC4IEnCNeEXH/OokJTk9WAc8Y+3Ip5RHs1h0DxBIJjo05J2Q1BAMsbA4V1WySa3yT8eTU55gq7xpDASLx3Cf3K7TGWEytTKeaZ0BRPvK6lWz54dMkNUZTKEgkliCaUK5ft4Khk5Hp5pw5ODO3HjpdsynnUf91r1nB7Q/vZSKWoCkcIBJLEk8q171mRdHbriY1myjKLanKzmePlXzHmko+biEYCAhBT00n9ViqdoRMJx9v7Qg8tanU+gEpab/L3qFRRsZjTowBIZ5Ujo5GiSVGC37/s8WXKqQFJwHHE8qLI84O3dYUIj6uvgV/354hvvTj3/LcMSfxdC5ooKEhzMhELK0/JfNkK+/R+50/3UdAoCEYIJFUGoIB4skE//jwPtacvRCFqSN3BZJJ75F8ekF/wRkLuHnDBdy3a5DDJyc40/0dXbxsEWOT8YI/x3xla365bfseOloaOBWNz+vEUc6j/nXdi/kkq077fufb52iJYgblatdMqoJCosia0Ew2vupsbn9oL7FE0jmCjSeIJ5X396zgyOjkVEICpyCG6ZpP2kJgMpZAUecxdR5SlMlYgtFILO11H3nmKHf94jkOnphgWXszH3ndSl6/qjNtHRH40OXn8L9+tIdEMpZW2H9w3TkMn4pyh1tId7Y1cngkAuKcuXb45ARLFzRxw2vP5fHnT/DiyARnLWriDy87lwvPbOPgiQke3XeUL/7ot4yMx0i6r3ngRIS2xhibf28tl3UvIanKc0dP+TbLvDA8zsKmUNoInYagcPDEeEGF+7ruxRUvODKbXxJJ5eREjFOTcc5d0jJvR+tA+Y/6q+H7LTdLFDOo1dEM685bzCevPP0I5xUrFp1WuM8m6DaVJZKKyHQTUCgg/PCpQ9y3a5D9x8aIxJJMxpOEQ8LS1gYOj0S49Ye/5ZNXxlnXvfi0Jo+3vqyLJwdH0uJbs2whw+NRBt1CuiEYoGthI8dPRYnGneT6Z2+6gHXdi7lu3TlpcUbca018e+cLjEWmk0TK6GSCf/rPZ3n1yo6c3vd8bHfObH4ZHo8SEOfARZCK/77L2SxWL0f95WSJYga1PJqhVEc4K5cs4MDwKU5FE8QSScLBAK0NIRY2N3D7w3uJxROMTcaJuyVzIqEMjU7S1dZEKCDct2sQ4LSa2b8//RKfvDL7kau3kG5tCNHaEGIilmBJa+Os7+nQyQlimVnCNZhHH818bHfOTH4xdxLAcHD62mWV+n3PxbkImfvEYwPH+dT9T9Vdf02hLFHMYD4eVWbjdySXKjCXNoRIJJMcPRXl+HiMsckECxqDnIom3CYrp6qRVAiJMDweZXlHM4dPTuRdMyumkD5rYTMvnZws+jOp1BFoOY+qMz/XQEBIJJTFrdPXaq/U77vUtXd1+45UT7+dBPqfO87X/vNZQgGhtSHIkdEIX3nwGW56QzevOrd9quacGkU2tQ2c/5JT20w97jTJploqZ3x9NMu2p5cr003TmrGdVN9X+vOc5zoxOcGp53bSux2Flob0ywHnQ2pxCN3LX3Wp/uAnO8r6Gt6jHG+BNdORcDnj2Pazgakj4hXtzdy0/vyCd6Ck54e6a+A4d/TtI5jxHv90/flc4u4wj+8f5luP7ueF4QlCAVjc2sDQaKrZAkIB0o7iwwFneWdbI+3NDRwZi9DaGJru+nB3jFPRBF/8/Zen7yjuDvH0iyd58LdDHD81yeLWRt50UScXndUGCr89dJKf/u4Ix8ejLG5uYP2FS1l1RhuK8szhMb698/nTmp4Ep1P7T9Z3e3Za747udkp7dzCfHQ+fgiDpliipwgTvcrIUPu7jh05EeHLwhDO6LiBTI6devmwRnW2NMxY63veRKkxUIcl0gZMqfI6finLwxIQ7ogxiCSUUcEbjpc6jWN7RzMLm0KyfwdRnx3Rcme87mS3eLLePn4pO9Zulnpx6XmtjaNb3nXnbzOz5237vcVXtyfd5JUkUInI1cDsQBL6hql/KeLwR+BbwauAY8H5V3e8+9lfAjTjdup9Q1e2zvd4Z3Wv0fV/49mk7a9qOkuV2MsuPNJWN1XNUjFugjkzEODQSIRpP0BAK0tXWyIKmUMZO6ylkZtohSP8he2PLLCS9yyfjScYi8dN+/AI0hgMERU47Ckl7r7bzGFNzUsPsIeMcMPfB1GhH8d4myxB9d3lqhGUwIOz8zFsKShRFNz2JSBC4E3grcADYJSK9qvq0Z7UbgWFVvUBErgNuA94vImuA64C1wNnAgyJyoar6jgU6MRHj3351qNjQ8xJNxOd0OKMfxWkmqGfeM1JEoCEUmDpnJaFKLK4kUQIiNIUCzuPujuTdgaSAHU+Q07eR7bkyvdNP7+ipc3CmbwM88cIJwoHp829S4gnltecvAdKHRme7fdr78Lz29G04PDLJky8MExAhFBTnQjso61YuZllH82nvPe22+xnj87j3MwhI6v2kr+/9DJ47Msb23S8RDDh9JrFEkoTCOy8+i1VnLiCQ7fvx3E7/DKbjQcj63Nsf3MvIRIzG0HT/zGQ8SXtLA//z6ouc7zvjO5767j3fpWS+ZsbvI3Wu1dTt1Ofk91zvl19iHS0NLP5MYc8tRR/FOmCfqg4AiMh9wEbAmyg2Ap93bz8A3CHOJ7IRuE9VJ4HnRGSfu71H/V6wrTHEhgs7EXGq0y8cG2cinqA5HOS8Ja10LnSu0JVtB/L+ANJ+UJ71vCfMpZ885/2RZy8gUjuQd2dK+5HksGOlnndn3z5GJ+IEAqndDFLV/4VNIT711ouy7kBTMZPlyMJz+x8f2sdIxNlhUlufjCcJBYTJRJKgCI3hAC+OTJBMwNK2BlrCzk9mJBIlEksSDgixpBIOBmgMBTg2FkWBhqDTbBAKBrnyok5+c/AkQ6MRuhY28d5LlvPq8zpO/zy8n6/7Hry3r//6ThY2hTyfhVPzGo3E+dc/mZspOcrlU/c/dVqfWKoT/29+b03JX2txa8NprzV8Klby18rNGVyyomOqT2hFR2tZ+4T++PXncfvDe1GYanINBISPvG4l5y1tLctr1rpSJIplwKDn/gHgspnWUdW4iIwAS9zlOzOeu2y2FzyrvZnN71oz1Y+woCnE0nADkViSQycjvK9nfoxg+PGvD/P0oRGU6cI/mYRQ0BmRdMWqpUVt/8YrnB0GnKas1A7TGA4S8HQudi1o4qWTEU6Mx1i0JEwklqQpHOLTb5vur3ls4Dh/0/sbAgHnxL+kwmgkQXuLsOfwGF/9wCVFxQrFDTAo91nJxW5/LkdaVeOIvrk8F8GGy+avZkY9ichNwE0AZy93dp5aPdchV9e9ZgW3bT/FyYkYyHTfw8KGcM4FiF8BNtMO85WHnkkrSBY0hmBhI0fGooxG4ll3rPt2DZJMKsGATNVuku7RfqkKoEIL03IPvyzF9uey8CrXiL5amiKkHk6SK6VSJIqDgHdPXe4uy7bOAREJAYtwOrVzeS4AqroN2AbOqCeoziOjUlrXvZhPv2112qincztyH/V0zy/28+3HXnCnoRCSyeRpBVi2Hab1ZyGePzZOUp0mpcWtDQQDAdaetYgvv/+VWV/r0MkJGkIB4gmdnmJEIJpQzm5vprkhmN5EByDwi73HuGfn87w4MsHZi5r50OXn8rpVS6Z74D1Ntleu6aIpHOSenc9zaGSCs1LrX7BkavXMQQQA3+0fJBwUmsOpiQgDTMTi3N8/yGsvWDI1ZHF6G/l1/5fqgGWuCq9y1F7q5boM9aoUiWIXsEpEzsMp5K8DPpCxTi9wA07fw7XAw6qqItIL/KuIfBmnM3sV8FiuL1wP5zqkCg/v0VrqRLbUDphq1/fOH7Xz2aN857EXUFXCQacp6MR4nI5W4YEnDvC2i8+cnrpjqp8DHtl7lJGJKImkEhCIJ5IcHonQ0RLmv71zDcs6pj/bVF+BCJy3pJXnjo5xfDwG6ixLqtM5+YkrV3HWotO/k749Q3z5wWcIB4UlrQ2cmIjy5Qefob1l7YyT77370mW8+9JZWyfTDI1N0t4cTusoDAfDHB2b5NwlM7dJZxsCmjmmPakwNBZhUVPY+QDdHNPSEJya0yrbzK6/LMM8YrkqR+1lvtfu613RicLtc7gZ2I4zPPYuVd0tIluAflXtBb4J3ON2Vh/HSSa4630Xp+M7Dnx8thFPXvPhDFpvIZ85QWDQvf+LfUe5o2/fVIF6MhLlzr59dLWt5U0v68o6UuKBxw+65zk4o30Qp5Aai8R56WSEDs+JVl7fevR52lsaWNDoFKTRhHMtiqULGrnq4jNnfB8fe+P5bO7dzZJWYTQSZzKeJBgQPr7h/KlCP3Pq7uFTkyW7qE+uF0NKyWWixNRoFPfejOudu7j1tO2PR+Ocu6SVMxc1nbZ+354h7nS/z6We73Nx6xquuHBpWnLKPC/DO7X49LWj859WvNS1l/leu693JemjUNUfAT/KWLbZczsC/MEMz/0i8MVCXreaOqVE3EI+wFQBHwyIU1AHTp/9NTC1/uzD4b6183kaQ4Gpgqi1McB4NM7XH3mOK9eckfU5g8PjNGZpCppMJH0LyNQ04NIgLHSnA09ditTPbFe+yzZ19/5jp1jenl7TKGRK9FwvhlSuC1Tlu/2tOwZoyPJ9/t9f7PdNxrNJJRZvDUY995PKVGJJeGo4qZPtirmORT3U7utZzXRmz6SUR0YzHd0716Nw7qeG0gYkdV0KmTq7tVwKucbEio4W4omkc5W4pJMkEqqEAgHfArKYacr9rnyX7YJF4UCAl0YnWdg8XbspZEr0XC+GVOh1vGeT7/bLddU752AFgj61n1xkTsHuPes8qdkvnPSh157DP/zHM0TiCee6ITVYuzczq8lEMXh8nOu/vnPGtt2pQjyjsE8dwacuPBQQ4ZFnjvCNR55jcHiccxa3lrQAKZVCCu/UUe6S1oYZm4L8nlfqo+9sheMZCxs5cCJS9Gtlbjt1re39x8a5ftvOqe+0nN9rPtuv9mu1TyWcPA5+3vvqFSxpbZxKlsvam/noFedxxYWd09f8SE5fpS/tvuc63KY61eRcT23LL9LX//k2IvEk8YTy2Xe+jA2ru6aafHI9uvc2WXgLqi3XzNyZWgmZcR47NcnxUzEWNAa58IyFMya3VLt9vkfRhT7Pz/XbdmZtx28IBmhvaSjqtbzbPjkRm7rQUTggnNXeXHXfaa387iohkVTiyaRzre+kpy9mhua0zPtmZh0tDSxe0Fi5uZ7mWvs5q/UNf/kNwClsutqauPem/M/MnanwKnR75ZQqvPcOjTIaidPREmbpgkYmYglGJmJ0LmhkdDJe1PWdy6mchaN324dHIkQTSQTh7KzJjNAAABHYSURBVPYm2prCVfmdliMZG6fZLDWhYuaVC5PJ6Sa1qeY1d7m6t2uxPMxVMYmiJpuevIpp2y1XW3E5pJo2MpNbPKGcGI8xFolzQdeCvK7vXCi/EUZ+8Zern8C77f3HxmkMCl0LnSQB1fmdlrsprF6JO39VMbxNYalEk5Zc3NuJzAEC87jprOYTRTFtu9XWVuytNUTjzrDUrrYmVJWxaIIVHS3sHRrlzIXTQy6Pjk1OTYQnIgUNMc2n4J9thJGfchaOMyVSqK72f1P9ggEhiBAu4PIN3iSTNuLM24SWdJKMt2mt2msyNZkoku4HW2xHa7mHTeYjVQDHEglGxmMgcCqqDJ+KEQgIy9qbGBqNMBqJEw5OsnSBkyyi7pXKGjxXKsvnCDrfgn+2EUaVVk3fabUqpEZoclNokklLHp4kEk8m3eax6Way1DpzqSYTRSgQYGQiVnTzRbmHTeYjVQAfG4tPDceNxBIgEBTh6FiU7s4FNIcDHD45yZHRqDNNsipJhM62xqlt5XMEnW/BX+3NddX0nVajYmqEpnwCASGQR4LxnjPjTS4JTz+M9zHV6ZkbClGTiaK7s5WfffrKkmyrGtqK+/YM0b//ODG3jVNwpr5QmJoOI5pIcnIixqloggDQEBQi8QQBEVpCAYIBKaiWlW/BP9fNdYX2h1T6O82mGo7kq71GaHLjPWemkCayfAVmX8WUU9+eIf7ygaeIezrCFCcxgJMkVJ2mpaNjzvWgm8JBzu9qY81ZizhncQvLOlroamtiZCJGV1tTXiOJVnS0MBFLnzXFr+DftL6bWEIZj8ZRdf6Wq2kndfQ7NBpJO/rt2zNU8tcqt2p5L4PD42lnT0N11QhNdbJEUWFbdwwwGokTnOGbEHfY3tIFDUzGk6CkNTM1h4OMTca596bL+dmnr+Temy7P68gw34J/w+outlyztuDElA/v0W+qoz4cFLbuGCj5a5VbtbyXfA8MjIEabXqqJbM1NwwOjxNPJgkFAwjTQ+7AaYK66Mw2VJVT0QQtDUFaG4NTwz6h+J28kDb9uWraqfb+kHxUy3uxzn5TCEsUZZRLx+GKjhaOjk6iCqFggBDOCAgJwCUrOtJOEkttr9Q7eS4FfyXa13PtD6mGtv/ZVMtQbOvsN4WwpqcyyqW5YdP6btqaQu6IhaTzT5UFjaHTEkAhzT59e4a4fttOrrjtYa7ftrOgNvFKta/n0ixWLW3/s5nLvp3ZbFjdVXBTpalPNTmFR09Pj/b391c6jFldcdvDp10wJzVlt3fUVt+eIW779z0MHD0FwHlLWrjl7S8r6dQWxUybUcmpTmab6qIWp2GxI3lTKSJSn1N4VLNcmxvK1eZfqqGQlWxfn+2zqZa2/1xU67BdY2ZjTU9lVOnmhlINhazmkTLVHJsx84UlijKay6Gk2ZSqEK10wqvV2KpZKfquTP2wPop5rJRTe1dz+3o1x1aN7HoY9avQPgpLFPOcFaImUy0NADClZZ3ZJivrQDWZamkAQK2qhXN78mGJwsyZ+bbz1KpqOflvvpqPM/RaZ7aZE7VyYlw9sAEA5VUt83qVkiUKMyfm485Tqyo9Gm++m48z9FrTk5kT1i5eXazvqnyqsWkv1ewb7lz58kKebzUKMyfsxDhTL6qtac/b7Ism44VswxKFmRPVtvOY7OxEvOJVW9Ne5lQ+hbCmJzMnbHrr6jcfR+tUSjU17WVr9s1XUYlCRBYD9wMrgf3A+1R1OMt6NwCfde9+QVXvFpEW4HvA+UAC+DdVvaWYeEx1q6adx0+9DuOtl+tp19v3m63PJF/FNj3dAjykqquAh9z7adxk8jngMmAd8DkR6XAf/t+quhq4BHi9iLy9yHiMKUo9D+Odj6N1MtXj9+tt9i1UsYliI3C3e/tu4N1Z1nkb8BNVPe7WNn4CXK2q46r6UwBVjQJPAMuLjMeYotTzMN56GHBQj9+vt88ECRRUrSg2UZyhqofc24eBM7KsswwY9Nw/4C6bIiLtwLtwaiVZichNItIvIv1HjhwpLmpjZlAPR9UzqYcBB/X6/aauahg7sv/XhTx/1kQhIg+KyG+y/NvoXU+d2QXznmFQRELAvcBXVXXGtK6q21S1R1V7Ojs7830ZY3JSD0fVM6m20TrlUM/fbzFmrYao6ltmekxEXhKRs1T1kIicBWRr6DsIbPDcXw70ee5vA/aq6v/JKWJjymjT+m429+7m6FiEkfEYk4kkoUCAja88u9KhzYlaGXBQqNT3Ox6Np02xPp9qTeVQbNNTL3CDe/sG4AdZ1tkOXCUiHW4n9lXuMkTkC8Ai4L8XGYcxJbFhdRfXXrqM46diRBNKUyhIR0uYB544OK87POtFPdSayqHY8yi+BHxXRG4EngfeByAiPcDHVPWjqnpcRG4FdrnP2eIuWw78NbAHeEJEAO5Q1W8UGZMxRXl04DjLO5pPu17DfBsmWq/me62pHIpKFKp6DHhzluX9wEc99+8C7spY5wAgxby+KVy9jSXPh81LZUw6m8KjDtXjWPJ8WIenMeksUdShehxLno9chonanEimnliiqEP1OpY8V7N1eFqNzNQbmxSwDlXjfPnVxq/Ds17mRDImxWoUdagezsAtJ6uRmXpjiaIO2Vjy4lhnt6k31vRUp2wseeHs7F5Tb6xGYUyerEZm6o3VKIwpgNXITD2xGoUxxhhfliiMMcb4skRhjDHGlyUKY4wxvixRGGOM8WWJwhhjjC9LFMYYY3xZojDGGOPLEoUxxhhfliiMMcb4skRhjDHGlyUKY4wxvixRGGOM8WWJwhhjjC9LFMYYY3xZojDGGOPLEoUxxhhfliiMMcb4skRhjDHGV1HXzBaRxcD9wEpgP/A+VR3Ost4NwGfdu19Q1bszHu8FulX14mLiyaZvzxBbdwwwODzOio4WNq3vtmsdm5piv2FTacXWKG4BHlLVVcBD7v00bjL5HHAZsA74nIh0eB5/DzBWZBxZ9e0ZYnPvboZGI7Q3hxkajbC5dzd9e4bK8XLGlJz9hk01KDZRbARStYO7gXdnWedtwE9U9bhb2/gJcDWAiCwAPgV8ocg4stq6Y4BwUGhpCCHi/A0Hha07BsrxcsaUnP2GTTUoNlGcoaqH3NuHgTOyrLMMGPTcP+AuA7gV+AdgfLYXEpGbRKRfRPqPHDmSU3CDw+M0h4Npy5rDQQ4Mz/pyxlQF+w2bajBrohCRB0XkN1n+bfSup6oKaK4vLCKvAs5X1e/nsr6qblPVHlXt6ezszOk1VnS0MBFLpC2biCVY3tGSa5jGVJT9hk01mDVRqOpbVPXiLP9+ALwkImcBuH+zNZweBFZ47i93l70W6BGR/cAjwIUi0lfc20m3aX03sYQyHo2j6vyNJZRN67tL+TLGlI39hk01KLbpqRe4wb19A/CDLOtsB64SkQ63E/sqYLuq/pOqnq2qK4ErgGdUdUOR8aTZsLqLLdespautiZGJGF1tTWy5Zq2NGDE1w37DphoUNTwW+BLwXRG5EXgeeB+AiPQAH1PVj6rqcRG5FdjlPmeLqh4v8nVztmF1l+1UpqbZb9hUmjhdC7Wlp6dH+/v7Kx2GMcbUFBF5XFV78n2enZltjDHGlyUKY4wxvixRGGOM8WWJwhhjjK+a7MwWkSM4o6xqyVLgaKWDKFKtvweLv7Is/sq7SFXb8n1SscNjK0JVczs1u4qISH8how2qSa2/B4u/siz+yhORgoaLWtOTMcYYX5YojDHG+LJEMXe2VTqAEqj192DxV5bFX3kFvYea7Mw2xhgzd6xGYYwxxpclCmOMMb4sUZSYiFwtIr8TkX0icto1xD3rvVdE1J1pt2rMFr+IfFhEjojIf7n/PlqJOGeSy+cvIu8TkadFZLeI/OtcxzibHL6Dr3g+/2dE5EQl4pxJDvGfIyI/FZEnReRXIvKOSsQ5kxziP1dEHnJj7xOR5ZWIcyYicpeIDInIb2Z4XETkq+77+5WIXDrrRlXV/pXoHxAEngW6gQbgKWBNlvXagB3ATqCn0nHnEz/wYeCOSsdaRPyrgCeBDvd+V6XjLuQ35Fn/z4C7Kh13nt/BNuBP3dtrgP2VjjvP+L8H3ODevhK4p9JxZ8S3HrgU+M0Mj78D+DEgwOXAL2fbptUoSmsdsE9VB1Q1CtwHbMyy3q3AbUBkLoPLQa7xV6tc4v8T4E5VHQZQ1WxXZaykfL+D64F75ySy3OQSvwIL3duLgBfnML7Z5BL/GuBh9/ZPszxeUaq6A/C75s9G4Fvq2Am0p65UOhNLFKW1DBj03D/gLpviVvNWqOoP5zKwHM0av+u9bpX1ARFZkeXxSskl/gtxLrv7cxHZKSJXz1l0ucn1O0BEzgXOY7rQqga5xP954A9F5ADwI5xaUbXIJf6ngPe4t38faBORJXMQW6nk/BtLsUQxh0QkAHwZ+PNKx1KEfwNWquorgJ8Ad1c4nnyFcJqfNuAcjX9dRNorGlHhrgMeUNVEpQPJ0/XAv6jqcpxmkHvcfaNW/AXwRhF5EngjcBCote8gL7X05dSCg4D3CHu5uyylDbgY6BOR/Tjtg71V1KE9W/yo6jFVnXTvfgN49RzFlotZ48c5eupV1ZiqPgc8g5M4qkUu7yHlOqqr2Qlyi/9G4LsAqvoo0IQz4V41yGUfeFFV36OqlwB/7S6rqgEFs8jnNwZYoii1XcAqETlPRBpwduTe1IOqOqKqS1V1paquxOnMvkZVq+W6rr7xA2S0ZV4D/HYO45vNrPED/w+nNoGILMVpihqYyyBnkct7QERWAx3Ao3Mc32xyif8F4M0AIvIynERxZE6jnFku+8BSTw3or4C75jjGYvUCf+SOfrocGFHVQ35PqMnZY6uVqsZF5GZgO87oibtUdbeIbAH6VfW0Hb6a5Bj/J0TkGiCO02H24YoFnCHH+LcDV4nI0zjNBX+pqscqF3W6PH5D1wH3qTuMpVrkGP+f4zT5/Q+cju0PV8v7yDH+DcDfiYjijF78eMUCzkJE7sWJcanbD/Q5IAygqv+M0y/0DmAfMA58ZNZtVsn3Y4wxpkpZ05MxxhhfliiMMcb4skRhjDHGlyUKY4wxvixRGGOM8WWJwhhjjC9LFMbMIRH5vIj8RYVe+zOVeF1T+yxRGFPFRKSUJ8VaojAFsURhaoaIrBSRPSLyL+4Fe74jIm9xZ4LdKyLrRKTVvXDLY+6FcTZ6nvszEXnC/fc6d/kG9+IzD7jb/o6IiE8M+92pPxCRHhHpc29/3n3dPhEZEJFPeJ7z1268jwAXeZafLyL/LiKPu7Gtdpf/i4j8s4j8Evj7GeLwe70/dN//f4nIVhEJisiXgGZ32XcK/xZMPbIpPEytuQD4A+CPcebl+QBwBc68U58BngYeVtU/dmeFfUxEHgSGgLeqakREVuFMppeajPESYC3OdRF+DrweeKSA2FYDb8KZ/PF3IvJPwCtwptt4Fc7+9gTwuLv+NuBjqrpXRC4DvoZzIRxwJmp73Swzw2Z7vQuA9wOvV9WYiHwN+KCq3iIiN6vqqwp4X6bOWaIwteY5Vf01gIjsBh5SVRWRXwMrcQrYazz9AE3AOThJ4A4ReRXOHE8Xerb5mKoecLf5X+52CkkUP3Rn1p0UkSHgDOANwPdVddzdfq/7dwHwOuB7ngpMo2db38th+vBsr/dmnBl9d7nbbcZJksYUzBKFqTWTnttJz/0kzu85AbxXVX/nfZKIfB54CXglTpOr9+qC3m0m8N8v4kw32Tb5xDbbdgLACZ8j/FM+z/V7PQHuVtW/yuH5xuTE+ijMfLMd+LNUP4OIXOIuXwQcUtUk8CGcmUELsZ/pa3C8N4f1dwDvFpFmEWkD3gWgqieB50TkD9w4RUReWWBMXg8B14pIl7vdxeJcCQ8gJiLhEryGqTOWKMx8cyvOlMq/cpumbnWXfw24QUSewmnbz+WIPZu/BW4XkX5yuKqZqj4B3I9z+cwf4/SrpHwQuNGNaTcluPayqj4NfBb4DxH5Fc5VCFPXENmG87lYZ7bJi00zbowxxpfVKIwxxviyzmxjshCR7wPnZSz+tKpun+M4PgJ8MmPxz1W1qq6qZuY3a3oyxhjjy5qejDHG+LJEYYwxxpclCmOMMb4sURhjjPH1/wFMLb01wfBquAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.regplot(y=y, x=dist_dd.mean_under_net)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd.net_ownership"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 446,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f81ce1ec460>"
      ]
     },
     "execution_count": 446,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "malaria_prob(dd.groupby('under_net'), 'malaria_now')\n",
    "\n",
    "sns.barplot(y='under_net', x='malaria_now', data=dd, orient='h', color=base_color)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "forms[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# dd.groupby(\"home\").used_spray.value_counts(normalize=True)\n",
    "\n",
    "sns.barplot(x='used_spray', y='home', data=dd.assign(used_spray = dd.used_spray == 'Yes'), color=base_color)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "forms[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 697,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>fever_treatment_seeking</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>I would seek treatment right away</td>\n",
       "      <td>0.690202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>I would seek treatment if the fever didn’t sub...</td>\n",
       "      <td>0.182558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>I don’t think it is necessary to seek treatmen...</td>\n",
       "      <td>0.070607</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>I would wait and see how the fever develops ov...</td>\n",
       "      <td>0.056633</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               index  fever_treatment_seeking\n",
       "0                  I would seek treatment right away                 0.690202\n",
       "1  I would seek treatment if the fever didn’t sub...                 0.182558\n",
       "2  I don’t think it is necessary to seek treatmen...                 0.070607\n",
       "3  I would wait and see how the fever develops ov...                 0.056633"
      ]
     },
     "execution_count": 697,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 675,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6902022589965853"
      ]
     },
     "execution_count": 675,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.fever_seek_treatment_rightaway.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1334,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "with sns.color_palette(sns.cubehelix_palette(3, start=1.25, rot=.0, dark=0.8, light=.35)):\n",
    "    fig, ax = plt.subplots(figsize=(8,2.5))\n",
    "    dat = dd.fever_treatment_seeking.value_counts(normalize=True)\n",
    "    dat = dat.reset_index()\n",
    "    sns.barplot(y='index', x='fever_treatment_seeking', data=dat, color=base_color)\n",
    "    ax.set_ylabel('')\n",
    "    ax.set_xlabel('Share of respondents')\n",
    "\n",
    "plt.savefig('treatment-seeking.png', bbox_inches='tight', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 688,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# worries about covid seem to be positively correlated with treatment seeking attitudes. \n",
    "\n",
    "with sns.color_palette(sns.cubehelix_palette(3, start=1.25, rot=.0, dark=0.8, light=.35)):\n",
    "    fig, ax = plt.subplots(figsize=(15,3))\n",
    "    sns.barplot(x='fever_seek_treatment_rightaway', y='worried_covid', color=base_color, data=dd, ax=ax)\n",
    "    ax.set_ylabel('Worried about getting COVID-19')\n",
    "    ax.set_xlabel('Would seek treatment right away in the advent of fever')\n",
    "\n",
    "plt.savefig('covid-treatment-seeking.png', bbox_inches='tight', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sns.barplot(x='fever_seek_treatment_rightaway', y='home', data=dd, color=base_color)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sns.barplot(x='fever_seek_treatment_rightaway', y='worried_malaria', data=dd, color=base_color)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "fever_seek_treatment_rightaway\n",
       "False    0.200064\n",
       "True     0.193914\n",
       "Name: mean_malaria_5_year, dtype: float64"
      ]
     },
     "execution_count": 332,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# treatment seeking same by district risk\n",
    "\n",
    "dd.groupby('fever_seek_treatment_rightaway').mean_malaria_5_year.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sns.swarmplot(x = 'if_malaria_tested', y = 'mean_malaria_5_year', data=dd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sns.swarmplot(x = 'if_malaria_seek_treatment', y = 'mean_malaria_5_year', data=dd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1577,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "home                                     dist_medical             \n",
       "Kutcha (made of mud, tin, straw)         Less than 15 minutes         0.025169\n",
       "                                         Between 15 and 30 minutes    0.025962\n",
       "                                         Between 30 and 60 minutes    0.029273\n",
       "                                         More than 60 minutes         0.030473\n",
       "Semi-pucca                               Less than 15 minutes         0.010277\n",
       "                                         Between 15 and 30 minutes    0.011218\n",
       "                                         Between 30 and 60 minutes    0.015094\n",
       "                                         More than 60 minutes         0.023014\n",
       "Pucca (have cement/brick wall and floor  Less than 15 minutes         0.010996\n",
       "                                         Between 15 and 30 minutes    0.010010\n",
       "                                         Between 30 and 60 minutes    0.011124\n",
       "                                         More than 60 minutes         0.021390\n",
       "Name: malaria_now, dtype: float64"
      ]
     },
     "execution_count": 1577,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.groupby(['home', 'dist_medical']).malaria_now.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sns.barplot(x='malaria_now', y='home', hue='dist_medical', data=dd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1559,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5294117647058824"
      ]
     },
     "execution_count": 1559,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(dd[dd.malaria_now].if_malaria_tested == 'Yes').mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1572,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Less than 24 hours    0.654971\n",
       "24-48 hours           0.198830\n",
       "More than 2 days      0.146199\n",
       "Name: if_malaria_seek_treatment, dtype: float64"
      ]
     },
     "execution_count": 1572,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd[dd.malaria_now & dd.if_malaria_seek_treatment.notna()].if_malaria_seek_treatment.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "malaria_prob(dd.groupby('home'), 'malaria_now')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd[['mean_bednet', 'mean_malaria_5_year', 'disthash']].groupby('disthash').head(1).corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# district fixed effects of kutcha\n",
    "\n",
    "def kutcha_dif(df):\n",
    "    a = df[df.kutcha == True].iloc[0]['count']\n",
    "    b = df[df.kutcha == False].iloc[0]['count']\n",
    "    return a - b\n",
    "\n",
    "malaria_prob(dd.groupby(['disthash', 'kutcha']), 'malaria_now').reset_index() \\\n",
    "    .groupby('disthash') \\\n",
    "    .filter(lambda df: df.shape[0] > 1) \\\n",
    "    .groupby('disthash') \\\n",
    "    .apply(kutcha_dif).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# Spatial Autocorrelation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1759,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import libpysal\n",
    "import esda\n",
    "\n",
    "\n",
    "gdf = gpd.read_file('outs/cluster.shp')\n",
    "gdf = gdf[gdf.geometry.notna()]\n",
    "gdf = gdf.to_crs(3857) \n",
    "\n",
    "dist_shapes = gdf.groupby('disthash').apply(lambda df: df.geometry.unary_union)\n",
    "dist_shapes = gpd.GeoDataFrame(dist_shapes.reset_index().rename(columns={0: 'geometry'}))\n",
    "\n",
    "dist_stats = dd[dd.disthash.isin(final_clusters)].groupby('disthash').mean().reset_index()\n",
    "dist_shapes = dist_shapes.merge(dist_stats)\n",
    "wnn = libpysal.weights.KNN.from_dataframe(dist_shapes, k=3)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1768,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.3878159251073957"
      ]
     },
     "execution_count": 1768,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mi = esda.moran.Moran(dist_shapes['mean_malaria'], wnn)\n",
    "mi.I"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1769,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.19119790275382118"
      ]
     },
     "execution_count": 1769,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mi = esda.moran.Moran(dist_shapes['mean_malaria_now'], wnn)\n",
    "mi.I"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# Effects of Recruitment Optimization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dat = dd[dd.disthash.isin(final_clusters)].reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filter_time(dat, (2020, 7,1), (2020,8,1)).groupby('disthash').kutcha.mean().quantile(.63)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 255,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.color_palette('Set2'):\n",
    "    _, ax = plt.subplots(figsize=(16, 5))\n",
    "    sns.distplot(filter_time(dat, (2020, 7,1), (2020,8,1)).groupby('disthash').kutcha.mean())\n",
    "    sns.distplot(filter_time(dat, (2020, 7,1), (2020,9,1)).groupby('disthash').kutcha.mean())\n",
    "    leg = ax.legend(title='Recruitment Strategy', labels=['Naive', 'Optimized'])    \n",
    "    ax.set_xlabel('Proportion of respondents living in kutcha dwellings')\n",
    "\n",
    "plt.savefig('optimization-kutcha-proportion.png', bbox_inches='tight', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dat['recruitment'] = 'Naive'\n",
    "dat.loc[in_time(dat, (2020, 8,1), (2020,9,1)), 'recruitment'] = 'Optimized'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "home\n",
       "Kutcha (made of mud, tin, straw)           0.035211\n",
       "Semi-pucca                                 0.013875\n",
       "Pucca (have cement/brick wall and floor    0.008420\n",
       "Name: malaria_now, dtype: float64"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filter_time(dat, (2020, 7,1), (2020,8,1)).groupby('home').malaria_now.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.18450930975065866"
      ]
     },
     "execution_count": 256,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filter_time(dat, (2020, 7,1), (2020,8,1)).groupby('disthash').kutcha.mean().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.24964053340133172"
      ]
     },
     "execution_count": 257,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filter_time(dat, (2020, 7,1), (2020,9,1)).groupby('disthash').kutcha.mean().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13"
      ]
     },
     "execution_count": 270,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(filter_time(dat, (2020, 7,1), (2020,8,1)).groupby('disthash').kutcha.mean() < 0.1).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 271,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(filter_time(dat, (2020, 7,1), (2020,9,1)).groupby('disthash').kutcha.mean() < 0.1).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 265,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.056405163853028795"
      ]
     },
     "execution_count": 265,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filter_time(dat, (2020, 7,1), (2020,8,1)).groupby('disthash').kutcha.mean().quantile(0.05)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.12272727272727273"
      ]
     },
     "execution_count": 267,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filter_time(dat, (2020, 7,1), (2020,9,1)).groupby('disthash').kutcha.mean().quantile(0.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "filter_time(dat, (2020, 8,1), (2020,9,1)).groupby('disthash').kutcha.mean().quantile(.9)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.16852599098029908"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filter_time(dat, (2020, 7,1), (2020,8,1)).kutcha.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 314,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pucca (have cement/brick wall and floor    9437\n",
       "Semi-pucca                                 4955\n",
       "Kutcha (made of mud, tin, straw)           4792\n",
       "Name: home, dtype: int64"
      ]
     },
     "execution_count": 314,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.home.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 316,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dd['semipucca'] = dd['home'] == 'Semi-pucca'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 318,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.1434108527131783"
      ]
     },
     "execution_count": 318,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dd.groupby('disthash').semipucca.mean().min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2717681715420977"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filter_time(dat, (2020, 8,1), (2020,9,1)).kutcha.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from adopt.clustering import get_budget_lookup, prep_df_for_budget, calc_price\n",
    "from adopt.malaria import window, days_left, get_df, get_cluster_from_adset\n",
    "from adopt.facebook.state import CampaignState, BudgetWindow\n",
    "\n",
    "w = BudgetWindow(datetime(2020,8,1), datetime.now())\n",
    "state = CampaignState(env, w)\n",
    "\n",
    "spend = {get_cluster_from_adset(n): i\n",
    "         for n, i in state.spend.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df = get_df(cnf)\n",
    "stratum = cnf['strata'][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "completed_df = prep_df_for_budget(df, stratum)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "price = calc_price(completed_df, w, spend)\n",
    "price = {k:v/100 for k, v in price.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from adopt.malaria import get_cluster_from_adset\n",
    "\n",
    "def prep_facebook_data(path, dd, final_clusters):\n",
    "    capture_count = dd[(dd.timestamp > datetime(2020,7,23,tzinfo=timezone.utc)) & \n",
    "                       (dd.timestamp < datetime(2020,7,24,tzinfo=timezone.utc))] \\\n",
    "        .groupby('disthash') \\\n",
    "        .userid.count() \\\n",
    "        .reset_index() \\\n",
    "        .rename(columns={'userid': 'completed_survey'})\n",
    "\n",
    "    df = pd.read_csv(path)\n",
    "    df['disthash'] = df['Ad Set Name'].map(get_cluster_from_adset)\n",
    "    df = df.merge(capture_count, how='left')\n",
    "    df = df[df.disthash.isin(final_clusters)].reset_index(drop=True)\n",
    "\n",
    "    spend = 'Amount Spent (INR)'\n",
    "    df['cost_per_completion'] = df.apply(lambda r: r[spend] / r.completed_survey, 1)\n",
    "    df = df.rename(columns = {'CTR (All)': 'CTR', \n",
    "                              'Cost per Results': 'cost_per_message', \n",
    "                              'CPM (Cost per 1,000 Impressions) (INR)': 'CPM'})\n",
    "\n",
    "    return df[['disthash', 'cost_per_completion', 'cost_per_message', 'CTR', 'CPM']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "completed_df['disthash'] = completed_df.cluster"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "naive_costs = prep_facebook_data('outs/fb-export-vlab-1.csv', completed_df, final_clusters)\n",
    "naive_costs = naive_costs[['disthash', 'cost_per_completion']].rename(columns={'cost_per_completion': 'base_cost'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "optimized_costs = pd.DataFrame(price.items(), columns=['disthash', 'optimized_cost'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "costs = optimized_costs.merge(naive_costs)\n",
    "costs['premium'] = costs.optimized_cost - costs.base_cost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "c = costs.merge(districts)\n",
    "# c[(c.include_audience == True) & (c.creative_group == 'exclusions')][0].mean()\n",
    "\n",
    "c['optimized'] = (c.include_audience == True) & (c.creative_group == 'exclusions')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "640780.3590426595"
      ]
     },
     "execution_count": 240,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8673116685646688"
      ]
     },
     "execution_count": 243,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "((c.optimized_cost * 250).sum() - (c[c.optimized].premium * 250).sum()) / (c.optimized_cost * 250).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "c[(c.include_audience == True) & (c.creative_group == 'exclusions')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "c[(c.include_audience == True) & (c.creative_group == 'exclusions')][0].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "di_shapes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 310,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "cluster_shapes = gpd.read_file('outs/cluster.shp')\n",
    "\n",
    "shapes = cluster_shapes.merge(c)\n",
    "shapes[shapes.disthash.isin(final_clusters)]\n",
    "\n",
    "# shapes['base_cost'] = np.sqrt(shapes.base_cost)\n",
    "# shapes['optimized_cost'] = np.sqrt(shapes.optimized_cost)\n",
    "\n",
    "shapes['log_base_cost'] = np.log(shapes.base_cost)\n",
    "shapes['log_optimized_cost'] = np.log(shapes.optimized_cost)\n",
    "plot_incidence_map(shapes, {'var': 'base_cost', 'title': 'costs'}, 'plasma')\n",
    "# plt.savefig('spending-map-naive.png', bbox_inches='tight', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 311,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_incidence_map(shapes, {'var': 'optimized_cost', 'title': 'costs'})"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "argv": [
    "python",
    "-m",
    "ipykernel_launcher",
    "-f",
    "{connection_file}"
   ],
   "display_name": "Python 3",
   "env": null,
   "interrupt_mode": "signal",
   "language": "python",
   "metadata": null,
   "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": "3.8.2"
  },
  "name": "descriptives.ipynb"
 },
 "nbformat": 4,
 "nbformat_minor": 4
}