9.1.2 Logistic regression (Labour market statistics) .ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Plot for group of biases ====> 1st VARIANT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Anaconda\\lib\\site-packages\\matplotlib\\__init__.py:872: UserWarning: axes.color_cycle is deprecated and replaced with axes.prop_cycle; please use the latter.\n",
      "  warnings.warn(self.msg_depr % (key, alt_key))\n"
     ]
    }
   ],
   "source": [
    "%pylab inline\n",
    "import seaborn as sns\n",
    "#sns.set_style(\"whitegrid\")\n",
    "plt.rcParams['patch.linewidth'] = 0.5  \n",
    "plt.rcParams['patch.edgecolor'] =\"black\"# 'white' \n",
    "plt.rcParams['lines.markeredgewidth']=0.5\n",
    "plt.rcParams['boxplot.showfliers']=False\n",
    "#'markerfacecolor'\n",
    "#print rcParams.keys()\n",
    "#plt.rcParams['figure.edgecolor'] ='white' "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Male bias [n=799].\n",
      "Female bias [n=6]:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Berufsgruppe</th>\n",
       "      <th>Frauen</th>\n",
       "      <th>KldB</th>\n",
       "      <th>KldB5</th>\n",
       "      <th>Männer</th>\n",
       "      <th>Overall</th>\n",
       "      <th>name</th>\n",
       "      <th>Bias</th>\n",
       "      <th>page_exist</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Berufe in der Hauswirtschaft</td>\n",
       "      <td>93.691910</td>\n",
       "      <td>8321</td>\n",
       "      <td>83212</td>\n",
       "      <td>6.308090</td>\n",
       "      <td>190866</td>\n",
       "      <td>Haushälter</td>\n",
       "      <td>female</td>\n",
       "      <td>only female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Betriebsinterne Wahlämter</td>\n",
       "      <td>32.042925</td>\n",
       "      <td>7123</td>\n",
       "      <td>71234</td>\n",
       "      <td>67.957075</td>\n",
       "      <td>12580</td>\n",
       "      <td>Gleichstellungsbeauftragter</td>\n",
       "      <td>female</td>\n",
       "      <td>only female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Berufe Geburtshilfe, Entbindungspflege</td>\n",
       "      <td>99.843137</td>\n",
       "      <td>8135</td>\n",
       "      <td>81353</td>\n",
       "      <td>0.156863</td>\n",
       "      <td>10200</td>\n",
       "      <td>Entbindungspfleger</td>\n",
       "      <td>female</td>\n",
       "      <td>only female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Berufe Gesundheits-, Krankenpflege (o.S)</td>\n",
       "      <td>84.959672</td>\n",
       "      <td>8130</td>\n",
       "      <td>81301</td>\n",
       "      <td>15.040328</td>\n",
       "      <td>807755</td>\n",
       "      <td>Pflegediensthelfer</td>\n",
       "      <td>female</td>\n",
       "      <td>only female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Berufe in der Hauswirtschaft</td>\n",
       "      <td>93.691910</td>\n",
       "      <td>8321</td>\n",
       "      <td>83212</td>\n",
       "      <td>6.308090</td>\n",
       "      <td>190866</td>\n",
       "      <td>Beschließer</td>\n",
       "      <td>female</td>\n",
       "      <td>only female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Berufe Geburtshilfe, Entbindungspflege</td>\n",
       "      <td>99.843137</td>\n",
       "      <td>8135</td>\n",
       "      <td>81353</td>\n",
       "      <td>0.156863</td>\n",
       "      <td>10200</td>\n",
       "      <td>Belegentbindungspfleger</td>\n",
       "      <td>female</td>\n",
       "      <td>only female</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               Berufsgruppe     Frauen  KldB  KldB5  \\\n",
       "0              Berufe in der Hauswirtschaft  93.691910  8321  83212   \n",
       "1                 Betriebsinterne Wahlämter  32.042925  7123  71234   \n",
       "2    Berufe Geburtshilfe, Entbindungspflege  99.843137  8135  81353   \n",
       "3  Berufe Gesundheits-, Krankenpflege (o.S)  84.959672  8130  81301   \n",
       "4              Berufe in der Hauswirtschaft  93.691910  8321  83212   \n",
       "5    Berufe Geburtshilfe, Entbindungspflege  99.843137  8135  81353   \n",
       "\n",
       "      Männer  Overall                         name    Bias   page_exist  \n",
       "0   6.308090   190866                   Haushälter  female  only female  \n",
       "1  67.957075    12580  Gleichstellungsbeauftragter  female  only female  \n",
       "2   0.156863    10200           Entbindungspfleger  female  only female  \n",
       "3  15.040328   807755           Pflegediensthelfer  female  only female  \n",
       "4   6.308090   190866                  Beschließer  female  only female  \n",
       "5   0.156863    10200      Belegentbindungspfleger  female  only female  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as pyplot\n",
    "from scipy import stats  \n",
    "import json\n",
    "def load_simple_json(filename):\n",
    "    with open(filename, 'r') as f:\n",
    "        return json.load(f)\n",
    "f_bias_prop=load_simple_json('de/f_bias_labour.json')\n",
    "n_bias_prop=load_simple_json('de/n_bias_labour.json')\n",
    "n_n_bias_prop=load_simple_json('de/n_n_bias_labour.json')\n",
    "m_bias_prop=load_simple_json('de/m_bias_labour.json')\n",
    "n_pages_lev=load_simple_json('de/wiki/m_neutral_lev.json')\n",
    "                       \n",
    "f_df=pd.DataFrame.from_dict(f_bias_prop)\n",
    "n_df=pd.DataFrame.from_dict(n_bias_prop)\n",
    "n_n_df=pd.DataFrame.from_dict(n_n_bias_prop)\n",
    "m_df=pd.DataFrame.from_dict(m_bias_prop)\n",
    "print \"Male bias [n={0}].\".format(len(m_df))\n",
    "print \"Female bias [n={0}]:\".format(len(f_df))\n",
    "f_df[\"Bias\"]=\"female\"\n",
    "m_df[\"Bias\"]=\"male\"\n",
    "n_df[\"Bias\"]=\"neutral\"\n",
    "n_n_df[\"Bias\"]=\"neutral\"\n",
    "f_df[\"page_exist\"]=\"only female\"#female form of profession\n",
    "m_df[\"page_exist\"]=\"only male\"\n",
    "n_df[\"page_exist\"]=\"female&male\"\n",
    "n_df.loc[n_df[\"name\"].isin(n_pages_lev.keys()),\"page_exist\"]=\"only neutral\"\n",
    "n_n_df[\"page_exist\"]=\"only neutral\"\n",
    "all_together = pd.concat([n_df,n_n_df,f_df,m_df],ignore_index=True)\n",
    "f_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Neutral(male&female profession label on Wiki) [n=12]:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Berufsgruppe</th>\n",
       "      <th>Frauen</th>\n",
       "      <th>KldB</th>\n",
       "      <th>KldB5</th>\n",
       "      <th>Männer</th>\n",
       "      <th>Overall</th>\n",
       "      <th>name</th>\n",
       "      <th>Bias</th>\n",
       "      <th>page_exist</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Obj.Pers.Brandschutz,Arbeitssicherh(ssT)</td>\n",
       "      <td>37.700505</td>\n",
       "      <td>5318</td>\n",
       "      <td>53182</td>\n",
       "      <td>62.299495</td>\n",
       "      <td>8915</td>\n",
       "      <td>Hilfspolizist</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Führung - Allgemeinbildende Schulen</td>\n",
       "      <td>62.997543</td>\n",
       "      <td>8419</td>\n",
       "      <td>84194</td>\n",
       "      <td>37.002457</td>\n",
       "      <td>2035</td>\n",
       "      <td>Direktor</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Berufe für Post- und Zustelldienste</td>\n",
       "      <td>47.894039</td>\n",
       "      <td>5132</td>\n",
       "      <td>51321</td>\n",
       "      <td>52.105961</td>\n",
       "      <td>205963</td>\n",
       "      <td>Bote</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Angehörige gesetzgebender Körperschaften</td>\n",
       "      <td>18.156809</td>\n",
       "      <td>7121</td>\n",
       "      <td>71214</td>\n",
       "      <td>81.843191</td>\n",
       "      <td>2908</td>\n",
       "      <td>Bundeskanzler</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Berufe in der Gemeindearbeit</td>\n",
       "      <td>58.613996</td>\n",
       "      <td>8332</td>\n",
       "      <td>83323</td>\n",
       "      <td>41.386004</td>\n",
       "      <td>10303</td>\n",
       "      <td>Diakon</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Angehörige geistl. Orden, Mutterhäuser</td>\n",
       "      <td>62.571104</td>\n",
       "      <td>8333</td>\n",
       "      <td>83332</td>\n",
       "      <td>37.428896</td>\n",
       "      <td>879</td>\n",
       "      <td>Mönch</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Berufe im Veranstaltungsservice,-manag.</td>\n",
       "      <td>58.326084</td>\n",
       "      <td>6340</td>\n",
       "      <td>63401</td>\n",
       "      <td>41.673916</td>\n",
       "      <td>27588</td>\n",
       "      <td>Host</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Berufe in der Landwirtschaft (o.S.)</td>\n",
       "      <td>21.576590</td>\n",
       "      <td>1110</td>\n",
       "      <td>11101</td>\n",
       "      <td>78.423410</td>\n",
       "      <td>124357</td>\n",
       "      <td>Knecht</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Apotheker, Pharmazeuten</td>\n",
       "      <td>83.810209</td>\n",
       "      <td>8180</td>\n",
       "      <td>81804</td>\n",
       "      <td>16.189791</td>\n",
       "      <td>37536</td>\n",
       "      <td>Apotheker</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Angehörige geistl. Orden, Mutterhäuser</td>\n",
       "      <td>62.571104</td>\n",
       "      <td>8333</td>\n",
       "      <td>83332</td>\n",
       "      <td>37.428896</td>\n",
       "      <td>879</td>\n",
       "      <td>Ordensbruder</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Berufskraftfahrer(Personentransport/PKW)</td>\n",
       "      <td>13.962924</td>\n",
       "      <td>5211</td>\n",
       "      <td>52112</td>\n",
       "      <td>86.037076</td>\n",
       "      <td>74927</td>\n",
       "      <td>Taxifahrer</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Mannequins, Dressmen, sonstige Models</td>\n",
       "      <td>88.775510</td>\n",
       "      <td>9423</td>\n",
       "      <td>94232</td>\n",
       "      <td>11.224490</td>\n",
       "      <td>98</td>\n",
       "      <td>Dressman</td>\n",
       "      <td>neutral</td>\n",
       "      <td>female&amp;male</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                Berufsgruppe     Frauen  KldB  KldB5  \\\n",
       "0   Obj.Pers.Brandschutz,Arbeitssicherh(ssT)  37.700505  5318  53182   \n",
       "3        Führung - Allgemeinbildende Schulen  62.997543  8419  84194   \n",
       "4        Berufe für Post- und Zustelldienste  47.894039  5132  51321   \n",
       "5   Angehörige gesetzgebender Körperschaften  18.156809  7121  71214   \n",
       "6               Berufe in der Gemeindearbeit  58.613996  8332  83323   \n",
       "9     Angehörige geistl. Orden, Mutterhäuser  62.571104  8333  83332   \n",
       "10   Berufe im Veranstaltungsservice,-manag.  58.326084  6340  63401   \n",
       "12       Berufe in der Landwirtschaft (o.S.)  21.576590  1110  11101   \n",
       "13                   Apotheker, Pharmazeuten  83.810209  8180  81804   \n",
       "16    Angehörige geistl. Orden, Mutterhäuser  62.571104  8333  83332   \n",
       "17  Berufskraftfahrer(Personentransport/PKW)  13.962924  5211  52112   \n",
       "18     Mannequins, Dressmen, sonstige Models  88.775510  9423  94232   \n",
       "\n",
       "       Männer  Overall           name     Bias   page_exist  \n",
       "0   62.299495     8915  Hilfspolizist  neutral  female&male  \n",
       "3   37.002457     2035       Direktor  neutral  female&male  \n",
       "4   52.105961   205963           Bote  neutral  female&male  \n",
       "5   81.843191     2908  Bundeskanzler  neutral  female&male  \n",
       "6   41.386004    10303         Diakon  neutral  female&male  \n",
       "9   37.428896      879          Mönch  neutral  female&male  \n",
       "10  41.673916    27588           Host  neutral  female&male  \n",
       "12  78.423410   124357         Knecht  neutral  female&male  \n",
       "13  16.189791    37536      Apotheker  neutral  female&male  \n",
       "16  37.428896      879   Ordensbruder  neutral  female&male  \n",
       "17  86.037076    74927     Taxifahrer  neutral  female&male  \n",
       "18  11.224490       98       Dressman  neutral  female&male  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print \"Neutral(male&female profession label on Wiki) [n={0}]:\".format(len(all_together[all_together.page_exist==\"female&male\"]))\n",
    "all_together[all_together.page_exist==\"female&male\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Neutral (only neutral label exist) [n=42]:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Berufsgruppe</th>\n",
       "      <th>Frauen</th>\n",
       "      <th>KldB</th>\n",
       "      <th>KldB5</th>\n",
       "      <th>Männer</th>\n",
       "      <th>Overall</th>\n",
       "      <th>name</th>\n",
       "      <th>Bias</th>\n",
       "      <th>page_exist</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Musik-,Gesangs-, Dirigententätigk. (ssT)</td>\n",
       "      <td>35.508637</td>\n",
       "      <td>9418</td>\n",
       "      <td>94183</td>\n",
       "      <td>64.491363</td>\n",
       "      <td>521</td>\n",
       "      <td>Audiodesigner</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Aufsicht,Führung-TheaterFilm,Fernsehprod</td>\n",
       "      <td>40.187288</td>\n",
       "      <td>9449</td>\n",
       "      <td>94494</td>\n",
       "      <td>59.812712</td>\n",
       "      <td>5019</td>\n",
       "      <td>Filmgeschäftsführer</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Berufe Informations,Telekommunikationst.</td>\n",
       "      <td>11.438161</td>\n",
       "      <td>2631</td>\n",
       "      <td>26313</td>\n",
       "      <td>88.561839</td>\n",
       "      <td>191884</td>\n",
       "      <td>Medientechniker</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Berufe Informations,Telekommunikationst.</td>\n",
       "      <td>11.438161</td>\n",
       "      <td>2631</td>\n",
       "      <td>26313</td>\n",
       "      <td>88.561839</td>\n",
       "      <td>191884</td>\n",
       "      <td>Projektassistent</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Berufe i.d. Bühnen- und Kostümbildnerei</td>\n",
       "      <td>70.973668</td>\n",
       "      <td>9461</td>\n",
       "      <td>94612</td>\n",
       "      <td>29.026332</td>\n",
       "      <td>3266</td>\n",
       "      <td>Dekorationenmaler</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Berufe in der Bautischlerei</td>\n",
       "      <td>2.720148</td>\n",
       "      <td>3333</td>\n",
       "      <td>33332</td>\n",
       "      <td>97.279852</td>\n",
       "      <td>23859</td>\n",
       "      <td>Messebauer</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Aufsicht,Führung-Krankenpfl.,Rettungsd.</td>\n",
       "      <td>78.356648</td>\n",
       "      <td>8139</td>\n",
       "      <td>81394</td>\n",
       "      <td>21.643352</td>\n",
       "      <td>36815</td>\n",
       "      <td>Pflegedienstleiter</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Berufe in der Elektrotechnik (o.S.)</td>\n",
       "      <td>25.599873</td>\n",
       "      <td>2630</td>\n",
       "      <td>26302</td>\n",
       "      <td>74.400127</td>\n",
       "      <td>220505</td>\n",
       "      <td>Elektrofachkraft</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Berufe Wasserversorgung,Abwassertechnik</td>\n",
       "      <td>3.556357</td>\n",
       "      <td>3431</td>\n",
       "      <td>34312</td>\n",
       "      <td>96.443643</td>\n",
       "      <td>27725</td>\n",
       "      <td>Fachkraft für Wasserversorgungstechnik</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Athleten, Berufssportler</td>\n",
       "      <td>8.496605</td>\n",
       "      <td>9424</td>\n",
       "      <td>94243</td>\n",
       "      <td>91.503395</td>\n",
       "      <td>5155</td>\n",
       "      <td>Jockey</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>Berufe im Holz-, Möbel-, Innenausbau</td>\n",
       "      <td>3.901460</td>\n",
       "      <td>2234</td>\n",
       "      <td>22342</td>\n",
       "      <td>96.098540</td>\n",
       "      <td>143049</td>\n",
       "      <td>Fachkraft für Möbel-, Küchen- und Umzugsservice</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Berufe in der Wasserwirtschaft</td>\n",
       "      <td>29.759646</td>\n",
       "      <td>3114</td>\n",
       "      <td>31142</td>\n",
       "      <td>70.240354</td>\n",
       "      <td>3162</td>\n",
       "      <td>Fachkraft für Wasserwirtschaft</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Berufe in der Metallbearbeitung (o.S.)</td>\n",
       "      <td>18.449343</td>\n",
       "      <td>2420</td>\n",
       "      <td>24202</td>\n",
       "      <td>81.550657</td>\n",
       "      <td>396735</td>\n",
       "      <td>Fachkraft für Metalltechnik</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Berufe im Objekt-,Werte-, Personenschutz</td>\n",
       "      <td>31.087103</td>\n",
       "      <td>5311</td>\n",
       "      <td>53112</td>\n",
       "      <td>68.912897</td>\n",
       "      <td>212059</td>\n",
       "      <td>Fachkraft für Schutz und Sicherheit</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Berufe im Gastronomieservice (o.S.)</td>\n",
       "      <td>65.264831</td>\n",
       "      <td>6330</td>\n",
       "      <td>63302</td>\n",
       "      <td>34.735169</td>\n",
       "      <td>391439</td>\n",
       "      <td>Fachkraft im Gastgewerbe</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>Berufe i.d. Lebensmittelherstellung (oS)</td>\n",
       "      <td>38.666751</td>\n",
       "      <td>2920</td>\n",
       "      <td>29202</td>\n",
       "      <td>61.333249</td>\n",
       "      <td>126398</td>\n",
       "      <td>Fachkraft für Lebensmitteltechnik</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>Berufe in der Abfallwirtschaft</td>\n",
       "      <td>8.069793</td>\n",
       "      <td>3433</td>\n",
       "      <td>34332</td>\n",
       "      <td>91.930207</td>\n",
       "      <td>11004</td>\n",
       "      <td>Fachkraft für Kreislauf- und Abfallwirtschaft</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Berufe Veranstaltungs- und Bühnentechnik</td>\n",
       "      <td>12.590514</td>\n",
       "      <td>9451</td>\n",
       "      <td>94512</td>\n",
       "      <td>87.409486</td>\n",
       "      <td>17815</td>\n",
       "      <td>Fachkraft für Veranstaltungstechnik</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>Aufsicht und Führung - Verkauf</td>\n",
       "      <td>51.144516</td>\n",
       "      <td>6219</td>\n",
       "      <td>62193</td>\n",
       "      <td>48.855484</td>\n",
       "      <td>96460</td>\n",
       "      <td>Kassenaufsicht</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>Berufe Arbeitssicherh. Sicherheitstech.</td>\n",
       "      <td>13.171087</td>\n",
       "      <td>5312</td>\n",
       "      <td>53123</td>\n",
       "      <td>86.828913</td>\n",
       "      <td>35707</td>\n",
       "      <td>Fachkraft für Arbeitssicherheit</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>Berufe für Post- und Zustelldienste</td>\n",
       "      <td>47.894039</td>\n",
       "      <td>5132</td>\n",
       "      <td>51322</td>\n",
       "      <td>52.105961</td>\n",
       "      <td>205963</td>\n",
       "      <td>Fachkraft für Kurier-, Express- und Postdienst...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>Berufe im Bestattungswesen</td>\n",
       "      <td>26.259416</td>\n",
       "      <td>8240</td>\n",
       "      <td>82402</td>\n",
       "      <td>73.740584</td>\n",
       "      <td>8496</td>\n",
       "      <td>Bestattungsfachkraft</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>Berufe im Holz- und Bautenschutz</td>\n",
       "      <td>1.222494</td>\n",
       "      <td>3324</td>\n",
       "      <td>33242</td>\n",
       "      <td>98.777506</td>\n",
       "      <td>1636</td>\n",
       "      <td>Fachkraft für Holz- und Bautenschutzarbeiten</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>Berufe in der Landwirtschaft (o.S.)</td>\n",
       "      <td>21.576590</td>\n",
       "      <td>1110</td>\n",
       "      <td>11102</td>\n",
       "      <td>78.423410</td>\n",
       "      <td>124357</td>\n",
       "      <td>Fachkraft für Agrarservice</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>Berufe im Güter- und Warenumschlag</td>\n",
       "      <td>14.357637</td>\n",
       "      <td>5133</td>\n",
       "      <td>51332</td>\n",
       "      <td>85.642363</td>\n",
       "      <td>16486</td>\n",
       "      <td>Fachkraft für Hafenlogistik</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>Berufe in der Lederherstellung</td>\n",
       "      <td>34.473198</td>\n",
       "      <td>2831</td>\n",
       "      <td>28312</td>\n",
       "      <td>65.526802</td>\n",
       "      <td>1082</td>\n",
       "      <td>Fachkraft für Lederverarbeitung</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>Berufe in der Lagerwirtschaft</td>\n",
       "      <td>24.976831</td>\n",
       "      <td>5131</td>\n",
       "      <td>51312</td>\n",
       "      <td>75.023169</td>\n",
       "      <td>1265705</td>\n",
       "      <td>Fachkraft für Lagerwirtschaft</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>Berufe Wasserversorgung,Abwassertechnik</td>\n",
       "      <td>3.556357</td>\n",
       "      <td>3431</td>\n",
       "      <td>34312</td>\n",
       "      <td>96.443643</td>\n",
       "      <td>27725</td>\n",
       "      <td>Fachkraft für Rohr-, Kanal- und Industrieservice</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>Berufe i.d. Kinderbetreuung, -erziehung</td>\n",
       "      <td>91.017849</td>\n",
       "      <td>8311</td>\n",
       "      <td>83111</td>\n",
       "      <td>8.982151</td>\n",
       "      <td>732486</td>\n",
       "      <td>Au-pair</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>Berufe in der Chemie- und Pharmatechnik</td>\n",
       "      <td>26.783835</td>\n",
       "      <td>4131</td>\n",
       "      <td>41312</td>\n",
       "      <td>73.216165</td>\n",
       "      <td>179627</td>\n",
       "      <td>Produktionsfachkraft Chemie</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>Berufe im Dialogmarketing</td>\n",
       "      <td>60.041787</td>\n",
       "      <td>9212</td>\n",
       "      <td>92122</td>\n",
       "      <td>39.958213</td>\n",
       "      <td>147414</td>\n",
       "      <td>Servicefachkraft für Dialogmarketing</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>Berufe in der Bauelektrik</td>\n",
       "      <td>1.461264</td>\n",
       "      <td>2621</td>\n",
       "      <td>26212</td>\n",
       "      <td>98.538736</td>\n",
       "      <td>224121</td>\n",
       "      <td>Elektrofachkraft für festgelegte Tätigkeiten</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>Berufe Bauplanung Verkehrswege, -anlagen</td>\n",
       "      <td>22.241135</td>\n",
       "      <td>3113</td>\n",
       "      <td>31132</td>\n",
       "      <td>77.758865</td>\n",
       "      <td>3525</td>\n",
       "      <td>Fachkraft für Straßen- und Verkehrstechnik</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>Bus-, Straßenbahnfahrer/innen</td>\n",
       "      <td>11.630906</td>\n",
       "      <td>5213</td>\n",
       "      <td>52132</td>\n",
       "      <td>88.369094</td>\n",
       "      <td>119406</td>\n",
       "      <td>Fachkraft im Fahrbetrieb</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>Berufe Sozialverwaltung, versicherung</td>\n",
       "      <td>76.993323</td>\n",
       "      <td>7321</td>\n",
       "      <td>73214</td>\n",
       "      <td>23.006677</td>\n",
       "      <td>129106</td>\n",
       "      <td>Aufsichtsperson</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>Rechtsberatung,-sprechung,-ordnung (ssT)</td>\n",
       "      <td>57.932782</td>\n",
       "      <td>7318</td>\n",
       "      <td>73183</td>\n",
       "      <td>42.067218</td>\n",
       "      <td>26838</td>\n",
       "      <td>Rechtsbeistand</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>Berufe Wasserversorgung,Abwassertechnik</td>\n",
       "      <td>3.556357</td>\n",
       "      <td>3431</td>\n",
       "      <td>34312</td>\n",
       "      <td>96.443643</td>\n",
       "      <td>27725</td>\n",
       "      <td>Fachkraft für Abwassertechnik</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>Berufe i.d. spanenden Metallbearbeitung</td>\n",
       "      <td>2.737703</td>\n",
       "      <td>2423</td>\n",
       "      <td>24233</td>\n",
       "      <td>97.262297</td>\n",
       "      <td>195675</td>\n",
       "      <td>CNC-Fachkraft</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>Berufe Informations,Telekommunikationst.</td>\n",
       "      <td>11.438161</td>\n",
       "      <td>2631</td>\n",
       "      <td>26312</td>\n",
       "      <td>88.561839</td>\n",
       "      <td>191884</td>\n",
       "      <td>Elektrogerätefachkraft</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>Büro- und Sekretariatskräfte (o.S.)</td>\n",
       "      <td>79.444358</td>\n",
       "      <td>7140</td>\n",
       "      <td>71402</td>\n",
       "      <td>20.555642</td>\n",
       "      <td>1969075</td>\n",
       "      <td>Bürokraft</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>Berufe in der Süßwarenherstellung</td>\n",
       "      <td>57.693876</td>\n",
       "      <td>2926</td>\n",
       "      <td>29262</td>\n",
       "      <td>42.306124</td>\n",
       "      <td>12263</td>\n",
       "      <td>Fachkraft für Süßwarentechnik</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>Kaufleute im Automatenservice</td>\n",
       "      <td>31.104167</td>\n",
       "      <td>6114</td>\n",
       "      <td>61142</td>\n",
       "      <td>68.895833</td>\n",
       "      <td>4800</td>\n",
       "      <td>Fachkraft für Automatenservice</td>\n",
       "      <td>neutral</td>\n",
       "      <td>only neutral</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                Berufsgruppe     Frauen  KldB  KldB5  \\\n",
       "1   Musik-,Gesangs-, Dirigententätigk. (ssT)  35.508637  9418  94183   \n",
       "2   Aufsicht,Führung-TheaterFilm,Fernsehprod  40.187288  9449  94494   \n",
       "7   Berufe Informations,Telekommunikationst.  11.438161  2631  26313   \n",
       "8   Berufe Informations,Telekommunikationst.  11.438161  2631  26313   \n",
       "11   Berufe i.d. Bühnen- und Kostümbildnerei  70.973668  9461  94612   \n",
       "14               Berufe in der Bautischlerei   2.720148  3333  33332   \n",
       "15   Aufsicht,Führung-Krankenpfl.,Rettungsd.  78.356648  8139  81394   \n",
       "19       Berufe in der Elektrotechnik (o.S.)  25.599873  2630  26302   \n",
       "20   Berufe Wasserversorgung,Abwassertechnik   3.556357  3431  34312   \n",
       "21                  Athleten, Berufssportler   8.496605  9424  94243   \n",
       "22      Berufe im Holz-, Möbel-, Innenausbau   3.901460  2234  22342   \n",
       "23            Berufe in der Wasserwirtschaft  29.759646  3114  31142   \n",
       "24    Berufe in der Metallbearbeitung (o.S.)  18.449343  2420  24202   \n",
       "25  Berufe im Objekt-,Werte-, Personenschutz  31.087103  5311  53112   \n",
       "26       Berufe im Gastronomieservice (o.S.)  65.264831  6330  63302   \n",
       "27  Berufe i.d. Lebensmittelherstellung (oS)  38.666751  2920  29202   \n",
       "28            Berufe in der Abfallwirtschaft   8.069793  3433  34332   \n",
       "29  Berufe Veranstaltungs- und Bühnentechnik  12.590514  9451  94512   \n",
       "30            Aufsicht und Führung - Verkauf  51.144516  6219  62193   \n",
       "31   Berufe Arbeitssicherh. Sicherheitstech.  13.171087  5312  53123   \n",
       "32       Berufe für Post- und Zustelldienste  47.894039  5132  51322   \n",
       "33                Berufe im Bestattungswesen  26.259416  8240  82402   \n",
       "34          Berufe im Holz- und Bautenschutz   1.222494  3324  33242   \n",
       "35       Berufe in der Landwirtschaft (o.S.)  21.576590  1110  11102   \n",
       "36        Berufe im Güter- und Warenumschlag  14.357637  5133  51332   \n",
       "37            Berufe in der Lederherstellung  34.473198  2831  28312   \n",
       "38             Berufe in der Lagerwirtschaft  24.976831  5131  51312   \n",
       "39   Berufe Wasserversorgung,Abwassertechnik   3.556357  3431  34312   \n",
       "40   Berufe i.d. Kinderbetreuung, -erziehung  91.017849  8311  83111   \n",
       "41   Berufe in der Chemie- und Pharmatechnik  26.783835  4131  41312   \n",
       "42                 Berufe im Dialogmarketing  60.041787  9212  92122   \n",
       "43                 Berufe in der Bauelektrik   1.461264  2621  26212   \n",
       "44  Berufe Bauplanung Verkehrswege, -anlagen  22.241135  3113  31132   \n",
       "45             Bus-, Straßenbahnfahrer/innen  11.630906  5213  52132   \n",
       "46     Berufe Sozialverwaltung, versicherung  76.993323  7321  73214   \n",
       "47  Rechtsberatung,-sprechung,-ordnung (ssT)  57.932782  7318  73183   \n",
       "48   Berufe Wasserversorgung,Abwassertechnik   3.556357  3431  34312   \n",
       "49   Berufe i.d. spanenden Metallbearbeitung   2.737703  2423  24233   \n",
       "50  Berufe Informations,Telekommunikationst.  11.438161  2631  26312   \n",
       "51       Büro- und Sekretariatskräfte (o.S.)  79.444358  7140  71402   \n",
       "52         Berufe in der Süßwarenherstellung  57.693876  2926  29262   \n",
       "53             Kaufleute im Automatenservice  31.104167  6114  61142   \n",
       "\n",
       "       Männer  Overall                                               name  \\\n",
       "1   64.491363      521                                      Audiodesigner   \n",
       "2   59.812712     5019                                Filmgeschäftsführer   \n",
       "7   88.561839   191884                                    Medientechniker   \n",
       "8   88.561839   191884                                   Projektassistent   \n",
       "11  29.026332     3266                                  Dekorationenmaler   \n",
       "14  97.279852    23859                                         Messebauer   \n",
       "15  21.643352    36815                                 Pflegedienstleiter   \n",
       "19  74.400127   220505                                   Elektrofachkraft   \n",
       "20  96.443643    27725             Fachkraft für Wasserversorgungstechnik   \n",
       "21  91.503395     5155                                             Jockey   \n",
       "22  96.098540   143049    Fachkraft für Möbel-, Küchen- und Umzugsservice   \n",
       "23  70.240354     3162                     Fachkraft für Wasserwirtschaft   \n",
       "24  81.550657   396735                        Fachkraft für Metalltechnik   \n",
       "25  68.912897   212059                Fachkraft für Schutz und Sicherheit   \n",
       "26  34.735169   391439                           Fachkraft im Gastgewerbe   \n",
       "27  61.333249   126398                  Fachkraft für Lebensmitteltechnik   \n",
       "28  91.930207    11004      Fachkraft für Kreislauf- und Abfallwirtschaft   \n",
       "29  87.409486    17815                Fachkraft für Veranstaltungstechnik   \n",
       "30  48.855484    96460                                     Kassenaufsicht   \n",
       "31  86.828913    35707                    Fachkraft für Arbeitssicherheit   \n",
       "32  52.105961   205963  Fachkraft für Kurier-, Express- und Postdienst...   \n",
       "33  73.740584     8496                               Bestattungsfachkraft   \n",
       "34  98.777506     1636       Fachkraft für Holz- und Bautenschutzarbeiten   \n",
       "35  78.423410   124357                         Fachkraft für Agrarservice   \n",
       "36  85.642363    16486                        Fachkraft für Hafenlogistik   \n",
       "37  65.526802     1082                    Fachkraft für Lederverarbeitung   \n",
       "38  75.023169  1265705                      Fachkraft für Lagerwirtschaft   \n",
       "39  96.443643    27725   Fachkraft für Rohr-, Kanal- und Industrieservice   \n",
       "40   8.982151   732486                                            Au-pair   \n",
       "41  73.216165   179627                        Produktionsfachkraft Chemie   \n",
       "42  39.958213   147414               Servicefachkraft für Dialogmarketing   \n",
       "43  98.538736   224121       Elektrofachkraft für festgelegte Tätigkeiten   \n",
       "44  77.758865     3525         Fachkraft für Straßen- und Verkehrstechnik   \n",
       "45  88.369094   119406                           Fachkraft im Fahrbetrieb   \n",
       "46  23.006677   129106                                    Aufsichtsperson   \n",
       "47  42.067218    26838                                     Rechtsbeistand   \n",
       "48  96.443643    27725                      Fachkraft für Abwassertechnik   \n",
       "49  97.262297   195675                                      CNC-Fachkraft   \n",
       "50  88.561839   191884                             Elektrogerätefachkraft   \n",
       "51  20.555642  1969075                                          Bürokraft   \n",
       "52  42.306124    12263                      Fachkraft für Süßwarentechnik   \n",
       "53  68.895833     4800                     Fachkraft für Automatenservice   \n",
       "\n",
       "       Bias    page_exist  \n",
       "1   neutral  only neutral  \n",
       "2   neutral  only neutral  \n",
       "7   neutral  only neutral  \n",
       "8   neutral  only neutral  \n",
       "11  neutral  only neutral  \n",
       "14  neutral  only neutral  \n",
       "15  neutral  only neutral  \n",
       "19  neutral  only neutral  \n",
       "20  neutral  only neutral  \n",
       "21  neutral  only neutral  \n",
       "22  neutral  only neutral  \n",
       "23  neutral  only neutral  \n",
       "24  neutral  only neutral  \n",
       "25  neutral  only neutral  \n",
       "26  neutral  only neutral  \n",
       "27  neutral  only neutral  \n",
       "28  neutral  only neutral  \n",
       "29  neutral  only neutral  \n",
       "30  neutral  only neutral  \n",
       "31  neutral  only neutral  \n",
       "32  neutral  only neutral  \n",
       "33  neutral  only neutral  \n",
       "34  neutral  only neutral  \n",
       "35  neutral  only neutral  \n",
       "36  neutral  only neutral  \n",
       "37  neutral  only neutral  \n",
       "38  neutral  only neutral  \n",
       "39  neutral  only neutral  \n",
       "40  neutral  only neutral  \n",
       "41  neutral  only neutral  \n",
       "42  neutral  only neutral  \n",
       "43  neutral  only neutral  \n",
       "44  neutral  only neutral  \n",
       "45  neutral  only neutral  \n",
       "46  neutral  only neutral  \n",
       "47  neutral  only neutral  \n",
       "48  neutral  only neutral  \n",
       "49  neutral  only neutral  \n",
       "50  neutral  only neutral  \n",
       "51  neutral  only neutral  \n",
       "52  neutral  only neutral  \n",
       "53  neutral  only neutral  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print \"Neutral (only neutral label exist) [n={0}]:\".format(len(all_together[all_together.page_exist==\"only neutral\"]))\n",
    "all_together[all_together.page_exist==\"only neutral\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### var1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Median male bias:  35.5086372361\n",
      "Median neutral bias:  28.271740593\n",
      "Median female bias:  93.6919095072\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmoAAAFTCAYAAACJTyCHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEOyUSNQEQRAEQRDs1P8BgXdOQ8So32kAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11aa7510>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, big_axes = plt.subplots( figsize=(10,5) , nrows=1, ncols=2)\n",
    "for row, big_ax in enumerate(big_axes, start=1):\n",
    "    #big_ax.set_title(groups.size().index[row-1].title()+\" dominnated professions\", fontsize=12)\n",
    "    # Turn off axis lines and ticks of the big subplot \n",
    "    # obs alpha is 0 in RGBA string!\n",
    "    big_ax.tick_params(labelcolor=(1.,1.,1., 0.0), top='off', bottom='off', left='off', right='off')\n",
    "ax1 = fig.add_subplot(1,2,1) \n",
    "\n",
    "\n",
    "colors = [\"pink\",\"green\",\"windows blue\", \"amber\", \"greyish\", \"faded green\", \"dusty purple\"]\n",
    "with sns.xkcd_palette(colors):#sns.color_palette(\"pastel\", len(set(all_together.Bias.values))):\n",
    "    sns.stripplot(x=\"Bias\",y=\"Frauen\",order=[\"female\",\"neutral\", \"male\"],\n",
    "               data=all_together,jitter=True,edgecolor='white',linewidth=0.7)#,hue=\"smoker\",split=True)\n",
    "    ax1.set_ylabel('women in the labour market,%')\n",
    "    ax1.set_xlabel('Redirection bias')\n",
    "    ax1.set_ylim(0,100)\n",
    "    \n",
    "f_df2=all_together[all_together.page_exist==\"only female\"].Frauen\n",
    "m_df2=all_together[all_together.page_exist==\"only male\"].Frauen\n",
    "n_df2=pd.concat([n_df.Frauen,n_n_df.Frauen])\n",
    "\n",
    "ax1.set_xticklabels([\"female\\n[n={}]\".format(len(f_df2)),\"neutral\\n[n={}]\".format(len(n_df2)),\"male\\n[n={}]\".format(len(m_df2))])\n",
    "\n",
    "\n",
    "ax2= fig.add_subplot(1,2,2)\n",
    "ax2.boxplot([f_df2,n_df2,m_df2], )\n",
    "ax2.set_xticklabels([\"female\\n[n={}]\".format(len(f_df2)),\"neutral\\n[n={}]\".format(len(n_df2)),\"male\\n[n={}]\".format(len(m_df2))])\n",
    "#ax2.set_xticks([1,2,3],[\"female bias\\n[n={}]\".format(len(f_df2)),\"neutral\\n[n={}]\".format(len(n_df2)),\"male bias\\n[n={}]\".format(len(m_df2))])\n",
    "#ax2.ylabel(\"women in labour market, %\")\n",
    "\n",
    "ax2.scatter(x=[1,3,3,2,2],y=[all_together[all_together.name==\"Entbindungspfleger\"].Frauen.values[0],\n",
    "                all_together[all_together.name==u\"Sekretär\"].Frauen.values[0],\n",
    "                all_together[all_together.name==u\"Projektleiter\"].Frauen.values[0],\n",
    "                           all_together[all_together.name==u\"Host\"].Frauen.values[0],\n",
    "                            all_together[all_together.name==u\"Projektassistent\"].Frauen.values[0]],\n",
    "                 c=[\"r\",\"r\",\"r\",\"r\"])\n",
    "ax2.annotate('Hebamme', xy=(1, all_together[all_together.name==\"Entbindungspfleger\"].Frauen.values[0]), xytext=(1.2, 85),\n",
    "             arrowprops=dict(width=0.5,shrink=0.05,headwidth=4,facecolor='black'))\n",
    "ax2.annotate(u\"Sekretär\", xy=(3, all_together[all_together.name==u\"Sekretär\"].Frauen.values[0]), xytext=(2.3, 70),\n",
    "             arrowprops=dict(width=0.5,shrink=0.05,headwidth=4,facecolor='black'))\n",
    "ax2.annotate(u\"Projektleiter\", xy=(3, all_together[all_together.name==u\"Projektleiter\"].Frauen.values[0]), xytext=(2.05, 7),\n",
    "             arrowprops=dict(width=0.5,shrink=0.05,headwidth=4,facecolor='black'))\n",
    "ax2.annotate(u\"Host/Hostess\", xy=(2, all_together[all_together.name==u\"Host\"].Frauen.values[0]), xytext=(0.9, 48),\n",
    "             arrowprops=dict(width=0.5,shrink=0.05,headwidth=4,facecolor='black'))\n",
    "ax2.annotate(u\"Projektassistent/in\", xy=(2, all_together[all_together.name==u\"Projektassistent\"].Frauen.values[0]),\n",
    "             xytext=(0.6, 2),   arrowprops=dict(width=0.5,shrink=0.05,headwidth=4,facecolor='black'))\n",
    "ax2.set_xlabel(\"Redirection Bias\")\n",
    "ax2.set_ylabel('women involved in profession,%')\n",
    "ax2.set_ylim(-0.5,100.5)\n",
    "fig.show()\n",
    "fig.savefig(\"de/labour_market_redirect.png\",dpi=500)\n",
    "\n",
    "print \"Median male bias: \",m_df2.median()\n",
    "print \"Median neutral bias: \",n_df2.median()\n",
    "print \"Median female bias: \",f_df2.median()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# significanse test "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we check whether all groups follow the same distribution. In order to do so, we perform Mann-Whitney-Wilcoxon RankSum test. The null hypothesis is that all instances of two groups are independent and identically distributed. The alternative hypothesis is that all instances of the first group are independent and identically distributed according to some underlying distribution F_X and all instances of the second group are independent and identically distributed according to some underlying distribution F_Y but F_X differs from F_Y."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MWW RankSum P for male and female distributions = 0.000888464276433 z_stat= -3.32365376474\n",
      "MWW RankSum P for male and neutral distributions = 0.401687143991 z_stat= 0.838611876286\n",
      "MWW RankSum P for neutral and female distributions = 0.000804806421507 z_stat= -3.35113642833\n",
      "MWW RankSum P for all together and female distributions = 0.000915003594479 z_stat= -3.31543578854\n",
      "MWW RankSum P for all together and neutral distributions = 0.393021634532 z_stat= 0.854150448439\n",
      "MWW RankSum P for all together and male distributions = 0.965894830531 z_stat= 0.0427575157314\n"
     ]
    }
   ],
   "source": [
    "f_=all_together[all_together.page_exist==\"only female\"].Frauen.tolist()\n",
    "n_=all_together[all_together.page_exist==\"female&male\"].Frauen.tolist()\n",
    "m_=all_together[all_together.page_exist==\"only male\"].Frauen.tolist()\n",
    "n_n_=all_together[all_together.page_exist==\"only neutral\"].Frauen.tolist()\n",
    "\n",
    "z_stat, p_val = stats.ranksums(m_, f_)  \n",
    "print \"MWW RankSum P for male and female distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(m_, n_+n_n_)   \n",
    "print \"MWW RankSum P for male and neutral distributions =\", p_val ,\"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_+n_n_, f_)  \n",
    "print \"MWW RankSum P for neutral and female distributions =\", p_val  , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_+n_n_+m_+f_, f_)  \n",
    "print \"MWW RankSum P for all together and female distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_+n_n_+f_+m_, n_+n_n_)  \n",
    "print \"MWW RankSum P for all together and neutral distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_+n_n_+f_+m_, m_)  \n",
    "print \"MWW RankSum P for all together and male distributions =\", p_val , \"z_stat=\",z_stat\n",
    "\n",
    "\n",
    "#with Bonferroni correction (3*2/2=3)=> alpha=0.05/3=0.0167 # in order to controlling the familywise error rate\n",
    "#p-val should be smaller than 0.0167"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "| | p-val | z | \n",
    "| :-|:-:|\n",
    "| male bias & female bias |    8.88e-04     |    -3.32    |\n",
    "| male bias & neutral |     0.40     |    0.83    | \n",
    "| neutral & female bias |     8.04e-04     |      -3.35    |"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Two stage FDR correction of Benjamini-Hochberg:\n",
      "---------------------------------------------------------------------------\n",
      " ======================================== | p-value_old | reject | p-value_corrected\n",
      "---------------------------------------------------------------------------\n",
      "male bias & female bias                         8.88e-04    True   0.00044\n",
      "male bias & neutral                             4.02e-01   False   0.13390\n",
      "neutral & female bias                           8.05e-04    True   0.00044\n"
     ]
    }
   ],
   "source": [
    "from statsmodels.sandbox.stats.multicomp import multipletests\n",
    "p_all=[0.000888464276433,0.401687143991,0.000804806421507]\n",
    "names_c=[\"male bias & female bias\",\"male bias & neutral\",\"neutral & female bias\"]\n",
    "print \"Two stage FDR correction of Benjamini-Hochberg:\"\n",
    "res=multipletests(p_all, alpha=0.05, method='fdr_tsbh')[:2]\n",
    "print \"-\"*75\n",
    "print \" ======================================== | p-value_old | reject | p-value_corrected\"\n",
    "print \"-\"*75\n",
    "for i in range(0,len(names_c)):\n",
    "    print \"{:<46}  {:<3.2e} {!r:>7}   {:<3.5f}\".format(names_c[i],p_all[i],res[0][i],res[1][i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "| | p-val | z | reject null |\n",
    "| :-|:-:|:-:|\n",
    "| male bias & female bias |    8.88e-04     |    -3.32    |  4.44e-04 |\n",
    "| male bias & neutral |     0.40     |    0.83    |  1.34e-01 |  \n",
    "| neutral & female bias |     8.04e-04     |      -3.35    |   4.44e-04 |\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "H-statistic:\t11.8479647609\n",
      "P-value:\t0.00267452795285\n",
      "Reject NULL hypothesis - Significant differences exist between groups.\n"
     ]
    }
   ],
   "source": [
    "#we can use Kruskal–Wallis test (for comparing two or more independent samples of equal or different sample sizes)\n",
    "from scipy.stats.mstats import kruskalwallis\n",
    "H, pval = kruskalwallis(m_,f_,n_+n_n_)\n",
    "print \"H-statistic:\\t%s\\nP-value:\\t%s\" % (str(H),str(pval))\n",
    "if pval < 0.05:\n",
    "    print(\"Reject NULL hypothesis - Significant differences exist between groups.\")\n",
    "if pval > 0.05:\n",
    "    print(\"Accept NULL hypothesis - No significant difference between groups.\")\n",
    "# A significant Kruskal-Wallis test indicates that at least one sample stochastically dominates one other sample.\n",
    "#The test does not identify where this stochastic dominance occurs or for how many pairs of groups stochastic\n",
    "#dominance obtains. Dunn's test[5] would help analyze the specific sample pairs for stochastic dominance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "dependent variable: the percentage of women involved in profession;\n",
    "independent variable: bias category\n",
    "\n",
    "The results suggest that there is a statistically significant difference between the underlying distributions of:\n",
    "\n",
    "1) the percentage of women involved in male bias professions and the percentage of women involved in female bias professions (z = -3.32, p = 8.88e-04). \n",
    "\n",
    "2) the percentage of women involved in no bias (neutral) professions and the percentage of women involved in female bias professions (z = -3.35, p = 8.04e-04).\n",
    "\n",
    "There is no statistically significant difference between the underlying distributions of the percentage of women involved in male bias professions and the percentage of women involved in no bias (neutral) professions (z = 0.83, p = 0.4)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fit logistic regression model (predict female bias)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "%pylab inline\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import r2_score,confusion_matrix,classification_report\n",
    "from sklearn.cross_validation import train_test_split \n",
    "from sklearn.learning_curve import learning_curve\n",
    "from sklearn.metrics import roc_curve, auc\n",
    "from statsmodels.tools import categorical\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "from sklearn import cross_validation\n",
    "import statsmodels.api as sm\n",
    "from statsmodels.tools.tools import add_constant\n",
    "import numpy as np\n",
    "\n",
    "def reject_outliers2(data, m=2):\n",
    "    return data[abs(data - np.mean(data)) < m * np.std(data)]\n",
    "def load_data():\n",
    "    df=pd.read_csv(\"de/all_data_bias_labour_gender_all_persons.csv\",sep=\"\\t\",index_col=0)\n",
    "    df2=df[[\"labor_market_woman\",\"Bias\"]]\n",
    "    df2=df2.dropna()\n",
    "    df2.drop([\"Apothekerin\",\"Botin\",\"Bundeskanzlerin\",\"Diakonin\",\"Direktrice\",\"Politesse\",\"Hostess\",\"Magd\",\n",
    "         \"Mannequin\",\"Nonne\",\"Ordensschwester\",\"Taxifahrerin\"],inplace=True)\n",
    "    print len(df),len(df2)\n",
    "    df2[\"Bias2\"]=df2.Bias.apply(lambda x: 0 if x==\"male\" else 1 if x==\"neutral\" else 2 )\n",
    "\n",
    "\n",
    "    dff_m=reject_outliers2(df2[df2.Bias==\"male\"].labor_market_woman.values)\n",
    "    dff_f=reject_outliers2(df2[df2.Bias==\"female\"].labor_market_woman.values)\n",
    "    dff_n=reject_outliers2(df2[df2.Bias==\"neutral\"].labor_market_woman.values)\n",
    "    df_df = pd.DataFrame({'labor_market_woman':np.append(dff_m,values=np.append(dff_f,dff_n)),\n",
    "                          'Bias':[\"male\"]*len(dff_m)+[\"female\"]*len(dff_f)+[\"neutral\"]*len(dff_n)})\n",
    "    df_df[\"Bias2\"]=df_df.Bias.apply(lambda x: 0 if x==\"male\" else 1 if x==\"neutral\" else 2 )\n",
    "    return df_df\n",
    "\n",
    "def predict_model(predict_case,remove=False,train_split=True):\n",
    "    \n",
    "    df_df=load_data()\n",
    "    if remove==\"neutral\":\n",
    "        df_df=df_df[(df_df[\"Bias2\"]==0)|(df_df[\"Bias2\"]==2)]\n",
    "    elif remove==\"male\":\n",
    "        df_df=df_df[(df_df[\"Bias2\"]==1)|(df_df[\"Bias2\"]==2)]\n",
    "    df_df[\"Bias2\"]=df_df.Bias.apply(lambda x: 1 if x==predict_case else 0 )\n",
    "    \n",
    "    data = df_df[[\"labor_market_woman\",\"Bias2\"]]\n",
    "    data=sm.add_constant(data, prepend=False)\n",
    "    train_cols = data.columns[0]\n",
    "    y=data['Bias2']\n",
    "    X=data[train_cols]\n",
    "    X=sm.add_constant(X, prepend=False)\n",
    "    X=X.rename(columns = {'const':'intercept'})\n",
    "\n",
    "    #res1 = sm.MNLogit(y, X) \n",
    "    if train_split==True:\n",
    "        X_train, X_test, Y_train, Y_test = train_test_split(X, y, test_size=0.2)\n",
    "        print len(X_train), len(X_test), len(X_train) + len(X_test)\n",
    "    else:\n",
    "        X_train,Y_train=X,y\n",
    "    res1=sm.Logit(Y_train, X_train) \n",
    "    mlogit_res = res1.fit()\n",
    "    print mlogit_res.params\n",
    "    print mlogit_res.summary()\n",
    "    print mlogit_res.pred_table()\n",
    "    acc=float(sum(mlogit_res.pred_table().diagonal()))/float(len(Y_train))\n",
    "    print \"Accuracy:\",acc\n",
    "    print \"Odds:\\n\", np.exp(mlogit_res.params)\n",
    "    print \"p:\\n\", np.exp(mlogit_res.params)/(1+np.exp(mlogit_res.params))\n",
    "\n",
    "    return mlogit_res\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "885 859\n",
      "Optimization terminated successfully.\n",
      "         Current function value: 0.014341\n",
      "         Iterations 14\n",
      "labor_market_woman     0.364305\n",
      "intercept            -35.534682\n",
      "dtype: float64\n",
      "                           Logit Regression Results                           \n",
      "==============================================================================\n",
      "Dep. Variable:                  Bias2   No. Observations:                  840\n",
      "Model:                          Logit   Df Residuals:                      838\n",
      "Method:                           MLE   Df Model:                            1\n",
      "Date:                Tue, 30 Aug 2016   Pseudo R-squ.:                  0.6064\n",
      "Time:                        20:10:22   Log-Likelihood:                -12.047\n",
      "converged:                       True   LL-Null:                       -30.605\n",
      "                                        LLR p-value:                 1.113e-09\n",
      "======================================================================================\n",
      "                         coef    std err          z      P>|z|      [95.0% Conf. Int.]\n",
      "--------------------------------------------------------------------------------------\n",
      "labor_market_woman     0.3643      0.137      2.669      0.008         0.097     0.632\n",
      "intercept            -35.5347     12.709     -2.796      0.005       -60.443   -10.626\n",
      "======================================================================================\n",
      "[[ 835.    0.]\n",
      " [   3.    2.]]\n",
      "Accuracy: 0.996428571429\n",
      "Odds:\n",
      "labor_market_woman    1.439513e+00\n",
      "intercept             3.693888e-16\n",
      "dtype: float64\n",
      "p:\n",
      "labor_market_woman    5.900822e-01\n",
      "intercept             3.693888e-16\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "df_df=load_data()\n",
    "female_pred_full=predict_model(predict_case=\"female\",train_split=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "probability of being in female bias group given 2% of woman in labor market:\n",
      "7.65447166167e-16\n",
      "probability of being in female bias group given 20% of woman in labor market:\n",
      "5.3925996984e-13\n",
      "probability of being in female bias group given 80% of woman in labor market:\n",
      "0.00167499434126\n",
      "probability of being in female bias group given 90% of woman in labor market:\n",
      "0.0602439875119\n",
      "probability of being in female bias group given 95% of woman in labor market:\n",
      "0.283799589728\n",
      "probability of being in female bias group given 98% of woman in labor market:\n",
      "0.54170879189\n",
      "probability of being in female bias group given 99% of woman in labor market:\n",
      "0.629839638539\n"
     ]
    }
   ],
   "source": [
    "print \"probability of being in female bias group given 2% of woman in labor market:\"\n",
    "print 1.0/(1+np.exp(-(female_pred_full.params[1]+female_pred_full.params[0]*2)))\n",
    "\n",
    "print \"probability of being in female bias group given 20% of woman in labor market:\"\n",
    "print 1.0/(1+np.exp(-(female_pred_full.params[1]+female_pred_full.params[0]*20)))\n",
    "\n",
    "print \"probability of being in female bias group given 80% of woman in labor market:\"\n",
    "print 1.0/(1+np.exp(-(female_pred_full.params[1]+female_pred_full.params[0]*80)))\n",
    "\n",
    "print \"probability of being in female bias group given 90% of woman in labor market:\"\n",
    "print 1.0/(1+np.exp(-(female_pred_full.params[1]+female_pred_full.params[0]*90)))\n",
    "\n",
    "print \"probability of being in female bias group given 95% of woman in labor market:\"\n",
    "print 1.0/(1+np.exp(-(female_pred_full.params[1]+female_pred_full.params[0]*95)))\n",
    "\n",
    "print \"probability of being in female bias group given 98% of woman in labor market:\"\n",
    "print 1.0/(1+np.exp(-(female_pred_full.params[1]+female_pred_full.params[0]*98)))\n",
    "\n",
    "print \"probability of being in female bias group given 99% of woman in labor market:\"\n",
    "print 1.0/(1+np.exp(-(female_pred_full.params[1]+female_pred_full.params[0]*99)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "885 859\n",
      "672 168 840\n",
      "Optimization terminated successfully.\n",
      "         Current function value: 0.017014\n",
      "         Iterations 14\n",
      "labor_market_woman     0.344288\n",
      "intercept            -33.503123\n",
      "dtype: float64\n",
      "                           Logit Regression Results                           \n",
      "==============================================================================\n",
      "Dep. Variable:                  Bias2   No. Observations:                  672\n",
      "Model:                          Logit   Df Residuals:                      670\n",
      "Method:                           MLE   Df Model:                            1\n",
      "Date:                Tue, 30 Aug 2016   Pseudo R-squ.:                  0.6122\n",
      "Time:                        01:21:33   Log-Likelihood:                -11.433\n",
      "converged:                       True   LL-Null:                       -29.485\n",
      "                                        LLR p-value:                 1.871e-09\n",
      "======================================================================================\n",
      "                         coef    std err          z      P>|z|      [95.0% Conf. Int.]\n",
      "--------------------------------------------------------------------------------------\n",
      "labor_market_woman     0.3443      0.128      2.683      0.007         0.093     0.596\n",
      "intercept            -33.5031     11.941     -2.806      0.005       -56.908   -10.098\n",
      "======================================================================================\n",
      "[[ 667.    0.]\n",
      " [   3.    2.]]\n",
      "Accuracy: 0.995535714286\n",
      "Odds:\n",
      "labor_market_woman    1.410984e+00\n",
      "intercept             2.816945e-15\n",
      "dtype: float64\n",
      "p:\n",
      "labor_market_woman    5.852317e-01\n",
      "intercept             2.816945e-15\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "#use only 80% of data\n",
    "fem_pred=predict_model(predict_case=\"female\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## consider in neutral group only professions with 2 articles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfMAAAFgCAYAAACv/ZCmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/wfPdS8yqeDPHgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1f2b7668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f_df2=all_together[all_together.page_exist==\"only female\"].Frauen\n",
    "m_df2=all_together[all_together.page_exist==\"only male\"].Frauen\n",
    "n_df2=all_together[all_together.page_exist==\"female&male\"].Frauen\n",
    "pyplot.boxplot([f_df2,n_df2,m_df2])\n",
    "pyplot.xticks([1,2,3],[\"female bias\\n[n={}]\".format(len(f_df2)),\"no bias(female&male page)\\n[n={}]\".format(len(n_df2)),\"male bias\\n[n={}]\".format(len(m_df2))])\n",
    "pyplot.ylabel(\"women in labour market, %\")\n",
    "#pyplot.xlabel(\"Profession labels of Wiki pages\")\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MWW RankSum P for male and female distributions = 0.000888464276433 z_stat= -3.32365376474\n",
      "MWW RankSum P for male and neutral distributions = 0.0847385046498 z_stat= -1.72383017978\n",
      "MWW RankSum P for neutral and female distributions = 0.0148867238335 z_stat= -2.43512311011\n",
      "MWW RankSum P for all together and female distributions = 0.000982283024057 z_stat= -3.29555262954\n",
      "MWW RankSum P for all together and neutral distributions = 0.0980107548006 z_stat= -1.65457491801\n",
      "MWW RankSum P for all together and male distributions = 0.726986270702 z_stat= 0.349137252466\n"
     ]
    }
   ],
   "source": [
    " z_stat, p_val = stats.ranksums(m_, f_)   \n",
    "print \"MWW RankSum P for male and female distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(m_, n_)  \n",
    "print \"MWW RankSum P for male and neutral distributions =\", p_val ,\"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_, f_)  \n",
    "print \"MWW RankSum P for neutral and female distributions =\", p_val  , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_+m_+f_, f_)  \n",
    "print \"MWW RankSum P for all together and female distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_+f_+m_, n_)  \n",
    "print \"MWW RankSum P for all together and neutral distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_+f_+m_, m_)  \n",
    "print \"MWW RankSum P for all together and male distributions =\", p_val , \"z_stat=\",z_stat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "| | p-val | z |\n",
    "| :-|:-:|:-:|\n",
    "| male bias & female bias |    9.42e-04     |    -3.32    | \n",
    "| male bias & neutral |     0.084     |    -1.72    |  \n",
    "| neutral & female bias |     0.015     |      -2.43    | \n",
    "\n",
    "| all_together & female bias |     9.84e-04    |     -3.29    | \n",
    "| all_together & neutral |     0.098    |     -1.65    | \n",
    "| all_together & male bias |     0.727    |     0.35    | \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can reject null hypothesis for pairs male-female,neutral-female and all_together-female, i.e., two distributions are significantly different. In other words female professions that are represented in Wikipedia has significantly different distribution of involved in profession women. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2nd VARIANT ====>  Group articles by gender of title"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "all_together = pd.concat([n_df,n_n_df,f_df,m_df],ignore_index=True)\n",
    "f_val=load_simple_json('de/wiki/f_page_validated.json')\n",
    "f_amb=load_simple_json('de/wiki/f_page_ambigious.json')\n",
    "n_pages=load_simple_json('de/wiki/n_page_validated.json')\n",
    "n_pages_lev=load_simple_json('de/wiki/m_neutral_lev.json')\n",
    "\n",
    "words=load_simple_json('de/occupation_all.json')\n",
    "words_revert={words[i][1]:i for i in words}#feminine:masculine\n",
    "f_val_m=[]\n",
    "f_val_m_onlyval=[]\n",
    "for i in f_val:\n",
    "    f_val_m.append(words_revert[i])\n",
    "    f_val_m_onlyval.append(words_revert[i])\n",
    "for i in f_amb:\n",
    "    f_val_m.append(words_revert[i])\n",
    "all_together.loc[all_together[\"name\"].isin(n_pages),\"label\"]=\"neutral\"\n",
    "all_together.loc[all_together[\"name\"].isin(n_pages_lev.keys()),\"label\"]=\"neutral\"\n",
    "all_together.loc[all_together[\"name\"].isin(f_val_m),\"label\"]=\"female\"\n",
    "all_together.loc[all_together[\"name\"].isin(m_df.name.values),\"label\"]=\"male\"\n",
    "#m_all_df=all_together[all_together.page_exist.isin([\"only male\",\"female&male\"])]#811\n",
    "#m_all_df.loc[:,\"label\"]=\"male\"\n",
    "m_f_df=all_together[all_together.page_exist==\"female&male\"]#12\n",
    "m_f_df.loc[:,\"label\"]=\"male\"\n",
    "all_df=pd.concat([all_together,m_f_df],ignore_index=True)#871\n",
    "all_df.to_csv('de/labour_market_for_all_pages.csv', sep='\\t', encoding='utf-8')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "f_=all_df[all_df[\"label\"]==\"female\"].Frauen.tolist()\n",
    "f_onlyval=all_together[all_together[\"name\"].isin(f_val_m_onlyval)].Frauen.tolist() #f_val_m_onlyval\n",
    "n_=all_df[all_df[\"label\"]==\"neutral\"].Frauen.tolist()\n",
    "m_=all_df[all_df[\"label\"]==\"male\"].Frauen.tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYQAAAEbCAYAAADajfNFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQAlBREQiJQQREQHg/wFJtCKpyRnVGgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1dd0dd68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f_df2=all_df[all_df[\"label\"]==\"female\"].Frauen\n",
    "m_df2=all_df[all_df[\"label\"]==\"male\"].Frauen\n",
    "n_df2=all_df[all_df[\"label\"]==\"neutral\"].Frauen\n",
    "pyplot.boxplot([f_df2,n_df2,m_df2])\n",
    "pyplot.xticks([1,2,3],[\"female\\n[n={0}]\".format(len(f_df2)),\"neutral\\n[n={0}]\".format(len(n_df2)),\"male\\n[n={0}]\".format(len(m_df2))])\n",
    "pyplot.ylabel(\"women in labour market, %\")\n",
    "pyplot.xlabel(\"Gender of profession name of Wiki article\")\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1dd0d160>"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAECCAYAAAARlssoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1dd0da90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(m_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x23d020f0>"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAECCAYAAAARlssoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x23cf8c18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(n_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x23edcdd8>"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAECCAYAAAARlssoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xt0m/d93/E3wAt4A+/QlaLu+om6WDfbkmXLlmI7rl07\ncbJ1ibuebMlc53jpThuv2anb43WXtsvqxU2zzdkWJ07O1mZp3Mrd6vmWOrZlWZYlyxdJln66ixRF\nUaR4AwGSAIhnfwCkYFkUIRLkAwKf1zk+JvB7HuCLRwA+eH7P7/k9HsdxEBERAfC6XYCIiGQPhYKI\niIxSKIiIyCiFgoiIjFIoiIjIKIWCiIiMKhxvAWOMB3gaWAcMAg9ba0+ltD8APAFEgWettc+ktG0G\nvm2t3XHFY/468FvW2q0ZeRUiIpIR6ewpPAj4kl/gjwNPjTQYYwqTt+8CtgOPGGMCybZvAT8AfKkP\nZozZAHwtE8WLiEhmpRMKtwEvAVhr9wI3prQ1AcettX3W2ijwFnB7su0E8IXUBzLG1AJ/BPz2JOsW\nEZEpkE4oVAK9KbdjxhjvGG1BoArAWrsTiI00JNf5IfAYEAI8Ey9bRESmQjqh0Af4U9ex1sZT2ipT\n2vxAzxiPswlYBnwf+CnQZIx5aoxlRUTEBeMeaAZ2A/cDzxljtgAHU9qOAMuMMdVAmETX0ZNXrO8B\nsNbuA9YCGGMWAj+11j52rSd2HMfxeLRDISJynSb8xZlOKOwE7jbG7E7e/qox5iGg3Fr7jDHmMeCV\nZBHPWGvbrlh/wjPueTweOjqCE109pwQCfm2LJG2Ly7QtLtO2uCwQ8I+/0Bg8WT5LqqN/5AS94S/T\ntrhM2+IybYvLAgH/hPcUdPKaiIiMUiiIiMiodI4piOQ1x3EIBvvcLuOq6usr3C5BcoxCQWQcwWAf\nr+49QWlZudulfMJAOMRD9X60wy+ZpFAQSUNpWTll5RMf0SEyU+gnhoiIjFIoiIjIKIWCiIiMUiiI\niMgohYKIiIxSKIiIyCiFgoiIjFIoiIjIKIWCiIiMUiiIiMgohYKIiIxSKIiIyCiFgoiIjFIoiIjI\nKIWCiIiMUiiIiMgohYKIiIxSKIiIyCiFgoiIjBr3Gs3GGA/wNLAOGAQettaeSml/AHgCiALPWmuf\nSWnbDHzbWrsjeXs98D0gBgwBX7HWdmTu5YiIyGSks6fwIOCz1m4FHgeeGmkwxhQmb98FbAceMcYE\nkm3fAn4A+FIe67vAN6y1nwF2Ar+XgdcgIiIZkk4o3Aa8BGCt3QvcmNLWBBy31vZZa6PAW8DtybYT\nwBeueKwvWWsPJv8uBAYmWriIiGReOqFQCfSm3I4ZY7xjtAWBKgBr7U4S3USjrLXtAMaYrcA3gD+b\nWNkiIjIVxj2mAPQB/pTbXmttPKWtMqXND/Rc68GMMV8i0Q11n7X20nhPHgj4x1skb2hbXDad26K4\nOE5FeRflFSXT9pzp8BIB9L5IpW0xeemEwm7gfuA5Y8wW4GBK2xFgmTGmGgiT6Dp68or1PSN/GGN+\nA3gE2G6tvWZ4jOjoCKazWM4LBPzaFknTvS36+oL0h4aIMzhtz5mOcGgI0GdkhD4jl00mHNMJhZ3A\n3caY3cnbXzXGPASUW2ufMcY8BrxC4sv/GWtt2xXrOwDJLqc/B84CO40xDvCGtfbfTrh6ERHJqHFD\nwVrrAI9ecfexlPYXgBfGWPcssDX5dxyom3ClIiIy5XTymoiIjEqn+0hkyjmOQzDYl9ayxcVx+vqm\nr+84GOxLdoKK5D6FgmSFYLCPV/eeoLSsfNxlK8q76E8eZJ0OXZ3tlJVXUlahkS2S+xQKkjVKy8op\nKx//i7e8omRaRwKFQ/3T9lwibtMxBRERGaVQEBGRUQoFEREZpVAQEZFRCgURERmlUBARkVEKBRER\nGaVQEBGRUQoFEREZpVAQEZFRCgURERmlUBARkVGaEE8kTY7jMBgZJjwYYzgeZzjuEI8n7i8q9FJc\nVEBFaRFFhfqtJTOXQkHkCo7j0BeKcKlviEu9g1zsCjEQcRiM9BB3xr+wQqmvkEB1CbNrymicXUF5\nadE0VC2SGQoFEWAwEqO1I0TLxX7aLoWJxuKfaPcVeait9FFeUkhZSREFBR4KvB68Hg8eD0SHHYYi\nMYLhKL39EZrb+2lu72ff0YvMri1l9aJa5gfK8Xg8Lr1CkfQoFCRvRWNxzlzo42RrHx3dA6MXV6so\nLWLBrArqKkuorfLhDPZQXFRIbf2stB7XcRxCAzHOd4Y43dZHe9cA7V2t1FWWsHnVLOqrS6fuRYlM\nkkJB8k54MMbHZ7o43tJLdDixRxCoLmXBrHIaZlVQVV78iV/0nRd7r+vxPR4PFWVFrGisZkVjNd3B\nIT46eYmzF4K8+E4zqxbXsn55PQVe7TVI9lEoSN4Yig5z8OQljjb3EI87lPoKaFpUx/KGqint96/x\n+7hj/TwuXArz9qELHD7dRWfPAHdsmE9JccGUPa/IRCgUJOc5jsOJ1j7esxeJROOUlxSydmkdS+dV\nUlAwfSOF5tSV8cCti9h9sI3m9n5efOcsd9+0gAodiJYsolCQnNY/EOXtgxe40BWmsMDDxhX1NC2s\nmdYwSFVU6OWO9fM4cKyTw6e7eHVfC/fc3EhZiT6Kkh3GfScaYzzA08A6YBB42Fp7KqX9AeAJIAo8\na619JqVtM/Bta+2O5O2lwI+BOHDIWvuNzL0UkU86eyHInkMXiMTiNATK2bxqdlYMD/V4PGwyAbwe\nOHiqi1/sb+GezY34itSVJO5L5+fSg4DPWrsVeBx4aqTBGFOYvH0XsB14xBgTSLZ9C/gB4Et5rKeA\n37fW3gF4jTGfz8SLEEkVjzvsP3qRNz44T9xxuGXNbHZsnJ8VgZBq/fJ6VjZW09Mf4a2P2nDSOAdC\nZKqlEwq3AS8BWGv3AjemtDUBx621fdbaKPAWcHuy7QTwhSsea5O1dlfy7xdJhIlIxkSiw7x24Bwf\nn+mmsryY+25ZyPKG6qw8P8Dj8XBj0yzm1pXR2hHi4Kkut0sSSSsUKoHUMXkxY4x3jLYgUAVgrd0J\nxK7xuKPLimRCeDDGy++2cL4zzPxAOfdtaaS6wjf+ii7yejxsWzeXspJCPjjeyYWusNslSZ5L5+hW\nH+BPue211sZT2ipT2vxAzzUeK/U00fGWBSAQ8I+3SN7I5W1RXBynoryL8oqStJb3X7FcT/8QL7/b\nTDAcZc2SOrZtmI83Q3sHA6FivN6iTz1npviBe29ZxN/88gTvHG7ny3ebtOZP8hIBcvt9cb20LSYv\nnVDYDdwPPGeM2QIcTGk7AiwzxlQDYRJdR09esX7qJ/N9Y8zt1to3gXuB18Z78o6OYBol5r5AwJ/T\n26KvL0h/aIg4g+Mu668oIdh/ebm+UIRX3m0hPBRj3bI6blhaRyg0lLHaQqEIXu8wvtLxa5uocl8B\nTYtq+PhMN299cI4bV45/9nQ4+Rpz+X1xPXL9M3I9JhOO6YTCTuBuY8zu5O2vGmMeAsqttc8YYx4D\nXiHx5f+MtbbtivVTj579LvADY0wRiUB5bsKVi5AIhJffbWZgaJhNJsDqxbVulzRh65fX03KxnyNn\nulk0t5L6qqnZMxG5Fk+Wj3hwlPwJuf4rqK+vl7cOtlFWPv4vnJE9hfBglJf2ttA/EOXGlQFWLZqa\nQOi82IbXW5D23EeTceFSmFf2tTCrppR7bl5wzQPk4VCQB3esIBLRVN2Q+5+R6xEI+Cfcd6ozZvJQ\nNBp1u4RPiUZjOPH4+AsmDUWG+cX+c/QPRFm/rG7KAmG6zakrY8GsClouJmZZXThHfeQyvRQKeSYW\ni/G3r+yhpDy7Bn6Fw/109gyxatX4dUVjw/z9e+fo6Y+wcmE1a5fWTUOF02eTCXCuo58DxzpomFVO\ngVd7AjJ9FAp5qKSiioqqerfL+ARvkY9LwfZxl4vHHV7ac5bO3kEWz/Vz08pZWXkOwmRUlhezsrGG\nI2e7OXGuF9NY43ZJkkf0E0RmDMdxePdIO83tQeYHyrl17dycC4QRa5bUUuD1cPBUF8PxrD7uJzlG\noSAzhm3u4VhLL3VVJdy+bh7eHL4eQamvkBULqgkPxjjZen3XcxCZDIWCzAhtl0LsO3qRkuIC7tu6\nOK2Tu2a61Ytr8Xo9HDrVRVx7CzJNcv+TJTNeXyjCGx+cxwNs3zCPyvJit0uaFmUlhSxvqKJ/IMqZ\nC31ulyN5QqEgWS0SHeaXB1qJRONsXj2HWTVlbpc0rVYtShxkPnKmR7OoyrRQKEjWijsOuz5qozcU\noWlhDcsbsmsY7XTwlxWzYFYFl/oG6egZcLscyQMKBcla7x/roLUjxLz6MjYlLtORl5pG9xa6Xa5E\n8oFCQbLSydZeDp/uprKsKOdHGo1ndk0pNX4fze399A9k39noklsUCpJ1OnoG2HOonaJCLzs2NlCc\n55ep9Hg8NC2swQFOnNPwVJlaCgXJKqHBKL880IrjONyxfh5VFfkx0mg8C+f4KSr0cuJcr4anypRS\nKEjWGI47/PJAK4ORYTatDDCvvtztkrJGUaGXxXP9hIdinO8MuV2O5DCFgmQFx3E40+nQ1TfEsvlV\nNC3UfD9XWt5QDcAxdSHJFFIoSFY42tJPV8ghUF3K5tW5N8ldJtRVlVBb6aO1o5/w4LUufy4ycQoF\ncV1ze5DDZ4MUFyTOWNZU0WNb3lCF48CpNp3hLFNDnz5xVXdwkLc+aqPA62H5bC+lPs3mfi0L51Ti\n9cDp8woFmRoKBXHNwFCM195rJTbscLOppsynLqPxlBQXMD9QQXdwiJ6QzlmQzFMoiCuG4w5vfHCe\n0GCMdcvqmF9f6nZJM8aSeZUANLeHXa5EcpFCQaad4zjs/bidi90DLJzj54Ycu5zmVGuYVU5xoZfm\njgGdsyAZp1CQaXe0uYcT53qprfRx69o5Gml0nQq8XhbO8TMYiXPkbI/b5UiOUSjItDrfGWJ/8mI5\nOzbMp7BAb8GJWDw30YW09+OLLlciuUafSJk2faEIb35wHg8edmyYT3lpkdslzVizakvxFXnZd6ST\n4Xjc7XIkh4w7/s8Y4wGeBtYBg8DD1tpTKe0PAE8AUeBZa+0zY61jjFkH/LfkssestQ9n+gVJdhq9\nWE4szq1r5xCo0YHlyfB6PDTUl3CyLczR5h5WL6p1uyTJEensKTwI+Ky1W4HHgadGGowxhcnbdwHb\ngUeMMYFrrPOHwL+x1t4OlBhjfjVTL0SyV9xxePPDxMVyVi2qYen8/LtYzlRoCCSCdd8RdSFJ5qQT\nCrcBLwFYa/cCN6a0NQHHrbV91toosAu44yrrbEou/z5Qn9yT8JPYY5Acd8B2cL4zxPz6cjbm8cVy\nMq2+spjqimIOHOsgNqwuJMmMdEKhEkidgStmjPGO0dYPVJH4wk+9fzi5znHge8BhYBbw+sTKlpni\nZGsvH5/ppqq8mG3r5uLVSKOM8Xg83NQUoH8gim3WKCTJjHTmFOgj8SU/wmutjae0Vaa0+YHusdYx\nxvw5cKu19qgx5p+T6Fb6rWs9eSDgv1ZzXsnEtojFYvgrfFRUlGSgomtr6wyx53A7vqIC7r9tCdV+\n35jLeolQWlqEP8260l0uEwZCxXi96dc2XbxEWL6oklf3tXL0XC/bb17odkmu0/fF5KUTCruB+4Hn\njDFbgIMpbUeAZcaYaiAMbAOeTLZdbZ1LQDD593lg63hP3tERHG+RvBAI+DOyLWKxGMH+IZyCwQxU\nNbbQQJT/t+csjuOwbd1cCjwOwf6xnzMcGmJgIHrNZUb4K0rSWi5TQqEIXu8wvtLpe850hENDmE1V\nlPkK2XPwPF+8bVFen/ORqc9ILphMOKYTCjuBu40xu5O3v2qMeQgoT440egx4BfAAP7TWthljPrVO\n8v8PAz8zxkSBCPCbE65cslY0FueX7yculnNT0yxdLGcKFRZ4Wbesjj2H22lu72fhHP1SlskZNxSs\ntQ7w6BV3H0tpfwF4IY11sNa+TeIgtOQox3F4+9AFuvqGWN5QxcrGardLynkblgfYc7id9493KBRk\n0nTymmTURycvcfZCkNk1pdy8anZed2dMl9WLayks8PD+8U63S5EcoFCQjDl7IciHJy5RUVrEHRvm\nUeBVIEyHUl8hTQtrabnYT2fPgNvlyAynUJCM6A4OsftgG4UFHnZsnEdJsS6WM502LK8H0N6CTJpC\nQSZtKDrM6+8nLpZz69q51Piza+hmPli3bCQUOlyuRGY6hYJMiuM4vPVRG8FwlLVLanWg0yU1fh+L\n51ZyrKWX/gFNFCATp1CQSfnwxCVaO0LMqy9jXbILQ9yxcUU9ccfh4MlLbpciM5hCQSas5WI/H51M\nHFjedsM8TWHhsvXLE/NKqQtJJkOhIBMSDEd466M2Crwetm+Yh6+4wO2S8t68ujJm1ZRy8HQX0Zgm\nyJOJUSjIdYvHHXZ92EY0FmfL6tnUVurAcjbweDxsWF7PUGSYo83dbpcjM5RCQa7bB8c76ewdZMm8\nSl0bIcvcsKQOQMcVZMIUCnJdzneGOHS6C39ZEZtXzXa7HLnC8gXV+IoLOHhKoSATo1CQtA0Mxdh9\nsA2PB7atm0dRod4+2aawwMuqhTW0dw/Q3h12uxyZgfSplrSMTHQ3MDTMxhUB6qt0HCFb3bBUXUgy\ncQoFScuJ1l5aO0LMrStj1aIat8uRa1ibPK7wkbqQZAIUCjKu8GCU/Uc7KCr0snXtHM18muVqK0to\nCJRz9GwPQ9Fht8uRGUahINfkOA7vHG4nGouzyQQoLylyuyRJw9qldcSG41gNTZXrpFCQazrTFuRc\nR4g5tWUsb9Dw05liZGjqRzquINdJoSBjGozEePfIRQoLPNyyRhfMmUmWzq+i1JcYmuo4jtvlyAyi\nUJAxvfvxRYaiw6xfXo+/rNjtcuQ6FBZ4Wb2olo6eQS50aWiqpE+hIFd1vjPEmQtB6qtKWLlQo41m\norU6u1kmQKEgnxKPO+w7ehGAzatna/bTGWrNSChoaKpcB4WCfIpt7qG3P8LyhirqNNndjFXj99E4\nqwLb0sNgJOZ2OTJDKBTkEwYjMT480UlRoZcNK3TRnJkuMTTV4ejZHrdLkRlCoSCf8MHxTiKxOOuX\n1VNSXOh2OTJJo8cVTqsLSdIz7qfeGOMBngbWAYPAw9baUyntDwBPAFHgWWvtM2OtY4wJAD8AqoEC\n4CvW2tMZfk0yQV19gxxr6aWqvBjTWO12OZIBS+ZVJoamnkwMTdWwYhlPOnsKDwI+a+1W4HHgqZEG\nY0xh8vZdwHbgkeQX/1jr/Cnwv6y120kEycrMvAyZLMdx2HckcXD5pqZZeL368sgFhQVeVi2qpbN3\nkIvdA26XIzNAOqFwG/ASgLV2L3BjSlsTcNxa22etjQK7gDuuss6m5PK3Ag3GmFeBXwdez8BrkAxo\n7QjR3j1AQ6CcefXlbpcjGbRWo5DkOqQTCpVAb8rtmDHGO0ZbP1AF+K+4f9gYUwAsArqstXcDLcDv\nTbBuySDHcXj/eCcAG1cEXK5GMm3N4loADp7qcrkSmQnSOZLYR+JLfoTXWhtPaatMafMD3WOsM2yM\nuQT83+R9/xf4o/GePBDwj7dI3sjEtojFYvgrfFRUXB5qerylm+7gECsaa1gw1535jbxEKC0twl+R\n3hDYdJfLhIFQMV5v+rVNFy8RYPz3RSDgp3GOH9vSQ2V1Gb6igukozxX6vpi8dEJhN3A/8JwxZgtw\nMKXtCLDMGFMNhIFtwJPJtqutswu4D/gL4Hbg8HhP3tERTKPE3BcI+DOyLWKxGMH+IZyCQSBxoto7\nyauprV5UTbB/cNLPMRHh0BADA9G0nt9fUTKtdYZCEbzeYXyl7mybsYRDQ0B6n5GmxmqaLwR5+/0W\n1iyum+rSXJGpz0gumEw4ptN9tBMYMsbsBr4DfNMY85Ax5mFrbQx4DHiFRHj80FrbdrV1ko/1u8A/\nMca8BdwD/MmEK5eMOHm+j75wlOUNVZrfKIeNHFc4pC4kGce4ewrWWgd49Iq7j6W0vwC8kMY6WGub\ngc9OqFLJuOF4nI9OdOL1ekYv4Si5aXlDNcVFXg6eusSX71zudjmSxXTyWh471tJLaDDGysZqynTx\nnJxWVOilqbGGtkthOns1NFXGplDIU7HhOAdPXqKwwMOaJbVulyPTYI26kCQNCoU8deJcL4ORYZoW\n1mg6izyxdqnOV5DxKRTyUDzucOh0FwVeD02LdK2EfDGrupTZNaUcOdtNbDg+/gqSlxQKeai5c4jw\nYIwVC6q1l5Bn1iypYzAyzMnW3vEXlrykUMgz8bjDsdYBvB5Ypb2EvLM2efzoI3UhyRgUCnnmwPFO\n+gfjLJlfRXmpRhzlG9NYQ2GBVwebZUwKhTziOA4v7m0GLs+HI/nFV1SAaaym5WI/3cEht8uRLKRQ\nyCMHT3XRcjFEQ10xleU6ezlfrU3+IDh8WnsL8mkKhTzywp4zAJj5pa7WIe4aPV9BV2OTq1Ao5ImT\n53s5fq6XNYtrqSrXiKN8NreujLpKH4dPdzEc19BU+SSFQp54dV8LAJ+9ab7LlYjbPB4Pa5bUERqM\ncbpNs4rKJykU8kBX3yD7j3bQECjHLNC1lyV11lR1IcknKRTywGsHWok7DnffuEAXbhcAmhbWUOD1\n6Gps8ikKhRw3FB3mjQ9aqSgtYsvq2W6XI1mi1FfIsvlVnGnrIxiOuF2OZBGFQo7bc+gCocEYOzbM\np6gwdy/DKNdvzZJaHDQ0VT5JoZDD4o7Dq/tbKPB62LFRB5jlk0aOK6gLSVIpFHLYx6e7aLsU5uam\n2VRX+NwuR7LMglkVVJUXc/j0JeKO43Y5kiUUCjnslf2JYah339TgciWSjRJDU2vpC0c5e0FDUyVB\noZCjzneGOHSqixUNVSyaU+l2OZKl1i2tB+DDE50uVyLZQqGQo34xupewwOVKJJutXlxLgdfDBwoF\nSVIo5KD+gShvH7pAfVUJG5YH3C5Hslipr5CVjdU0t/fT1TfodjmSBRQKOeiND1qJxOLctakBr1cn\nq8m1rVuW6EL66KTObhaFQs6JDcd57UArvuICbrthntvlyAywPhkK6kISgHGnyzTGeICngXXAIPCw\ntfZUSvsDwBNAFHjWWvtMGuv8OvBb1tqtmXwxAu/ZDrqDQ9y1qYGyEs2GKuOrry5lfqCcI2e7GYoO\n4yvSSY75LJ09hQcBX/IL/HHgqZEGY0xh8vZdwHbgEWNMYJx1NgBfy9QLkE96dX8LHuDOGzUMVdK3\nflk90Vicj8/oRLZ8l04o3Aa8BGCt3QvcmNLWBBy31vZZa6PALuCOsdYxxtQBfwT8dqZegFx2srWX\nU+f7WLesntk1ZW6XIzPIyHEFDU2VdEKhEuhNuR0zxnjHaOsHqgD/VdYpBp4BHgNCgI6AZtirGoYq\nE7RkbiUVpUV8eEJnN+e7dDqd+0h8yY/wWmvjKW2pZ0b5ge6rrQPcACwDvg+UAk3GmKestY9d68kD\nAf+1mvPKtbbFxe4w+20Hi+ZWsm3T2FNkx2Ix/BU+KipKpqrMCfESobS0CH+adaW7XCYMhIrxetOv\nbbp4ScxumqnPyM2r5/Da/hb6hoZZvqAmI4853fR9MXnphMJu4H7gOWPMFuBgStsRYJkxphoIA9uA\nJ5Ntn1jHWrsfWAtgjFkI/HS8QADo6NDp95B4s19rWzz3yxPE4w47Nsyjs7N/zOVisRjB/iGcguwa\nkx4ODTEwECXYP35d/oqStJbLlFAogtc7jK80+7YZZO4zsrKhitf2t/D6vmaqZ+AghfE+I/lkMuGY\nTvfRTmDIGLMb+A7wTWPMQ8aYh621MRLdQa+QCI8fWmvbrrbOhCuUcQ1Fhnnjg/NUlhWxZZWumSAT\ns3pxLYUFOrs53437c8Ba6wCPXnH3sZT2F4AX0lgntf0soOGoGfL2oTbCQzE+d+siXTNBJqzUV4hp\nrOHw6S66+gaprcyu7jKZHjp5bYZLXDPhHIUFHnZs0DUTZHI2rkhMi/Ke7XC5EnGLQmGGO3Sqiwtd\nYTY3zaZK10yQSdq4IoDHA/vsRbdLEZcoFGa4kWGod92oYagyeVXlxZgF1Zw410t3cMjtcsQFCoUZ\nrLWjn8OnuzALqlk4R0PxJDM2mVkAHDimLqR8pFCYwX7x3jlAJ6tJZm1cEcAD7D+qLqR8pFCYofrC\nkdFrJozMcimSCTV+H8saqjjW0kNvv7qQ8o1CYYZ6/f1WorE4d9+0QNdMkIy7ceUsHNSFlI8UCjNQ\nNDbMa++do9RXyG1r57pdjuSgTcmhqfs1NDXvKBRmoD2H2+kLR9m+fh6lvpk3HYFkv9rKEpbOr+Ro\nczd94Yjb5cg0UijMMI7j8Mq+Fgq8Hu7cpGsmyNS50czCceB9dSHlFYXCDHPodBfnO0Pc1DRL0xDI\nlNpkkl1IGoWUVxQKM8wr7zYDcM9NjS5XIrmuvqqUxXMrOXK2h6C6kPKGQmEGabnYz+Ez3axs1Mlq\nMj1ubppF3HF494j2FvKFQmEGGdlL+OzN2kuQ6bF51Ww8Hnjn8AW3S5FpolCYITq6B3jn43bm1JZx\nw9I6t8uRPFFd4WPVolpOnu+jvSvsdjkyDRQKM8TON04wHHe4d0sj3jEutSkyFW5Znbhw0x7tLeQF\nhcIM0BeK8PI7Z6mr9HHL6jlulyN5ZuOKAMVFXvYcvoDjOG6XI1NMoTADvLq/hUh0mF/ZvJDCAv2T\nyfQqKS5k44oAHT2DHD/X63Y5MsX0DZPlwoNRXjtwjuoKH9tu0JQW4o5tN8wDYNeH512uRKaaQiHL\nvXaglYGhYT5/x1KKi3T9ZXGHaawmUF3CvqMXCQ/G3C5HppBCIYsNRYZ5ZV8LZb5C7tu6yO1yJI95\nPR623TCPSCzOu0fa3S5HppBCIYu9+eF5+gei3LmpgbKSIrfLkTx369q5eDyJ96XkLoVClopEh3lx\n71l8RQW6sppkhRq/jxuW1HHmQpCzF4JulyNTRKGQpV470EpPf4Q7NzVQUaq9BMkOOzYmZub9+wPn\nXK5Epsq4k/EbYzzA08A6YBB42Fp7KqX9AeAJIAo8a619Zqx1jDHrge8BMWAI+Iq1uorHlcKDMV7Y\nc4YyXyGr4RNqAAANn0lEQVT3btGUFpI91iypZVZ1KXs/bucf7VimHyw5KJ09hQcBn7V2K/A48NRI\ngzGmMHn7LmA78IgxJnCNdb4LfMNa+xlgJ/B7GXodOeWld5sJDca4d0sj5TqWIFnE6/HwmY3zicbi\n7PpIxxZyUTqhcBvwEoC1di9wY0pbE3DcWttnrY0Cu4A7rrLOpuTyX7LWHkz+XQgMTPoV5JjeUIRX\n97VQVV7MXZt0LEGyz203zKW4yMsvD7QSj+sM51yTTihUAqmnMcaMMd4x2vqBKsB/xf3DxhivtbYd\nwBizFfgG8GcTLTxXvfD2GYaiwzxw6yJ8xTovQbJPWUkRW9fMpbN3kPd0Vback84FfvtIfMmP8Fpr\n4yltlSltfqD7WusYY75EokvpPmvtpfGePBDIn+sGtHeFef2DVmbXlvHFOw1FhZ/M7Exsi1gshr/C\nR0VFdl21zUuE0tIi/GnWle5ymTAQKsbrTb+26eIlceEbNz4jD92zkjc+aOXV/S3ce9sSPFkySWM+\nfV9MlXRCYTdwP/CcMWYLcDCl7QiwzBhTDYSBbcCTybZPrWOM+Q3gEWC7tbYnnQI7OvJn6NuP/u5j\nYsMOn9u6iJ7u0CfaAgF/RrZFLBYj2D+EUzA46cfKpHBoiIGBKMH+8evyV5SktVymhEIRvN5hfKXZ\nt83Anc9IEYmJ8t6zHeza30zTotppr+FKmfqM5ILJhGM63Uc7gSFjzG7gO8A3jTEPGWMettbGgMeA\nV0iExw+ttW1XWed3kl1Ofw5UADuNMa8ZY/5wwpXnmNNtfew5dIGGQDmbV812uxyRcd27eSEAL+5t\ndrkSyaRx9xSstQ7w6BV3H0tpfwF4IY11AHR1mKtwHIe//MUxHOChu1bg9WbHrrjItSyZV8nKxmoO\nne7idFsfi+dWjr+SZD2dvJYF3jnczsnWPjaZAE0La9wuRyRtD9y6GIDnd512uRLJFIWCy8KDMf7q\n9RMUFXr50o5lbpcjcl2aFtawsrGag6cucaJV11rIBQoFl/31Gyfp7Y9w/y0Lqa8udbsckev24LYl\nAOx889Q4S8pMoFBw0YnWXl5/v5V59eXcu2Wh2+WITMiKBdWsXlTDkbPdHDo97ihzyXIKBZdEY3F+\n8uJRHOAr9xhdZlNmtF/bsQwP8LO/P8FwPD7u8pK99E3kkud3naK1M8SODfNZsaDa7XJEJqVxtp9t\n6+bS2hnizQ80J9JMplBwwfFzPby0t5lZ1aX82o6lbpcjkhFf2LYEX3EBO3edpn8g6nY5MkEKhWkW\nHozxzN99DMDXfrWJkuJ0TioXyX5VFT4+f+ti+gei/Oy1426XIxOkUJhGjuPw4xeP0NEzyH23LFS3\nkeScu29qoHF2BbsPXuDjM11ulyMToFCYRr98v5X9toMVDVU8uG2x2+WIZFyB18tX723C6/Hwk5eO\nMhiJuV2SXCeFwjSxzd389BfHqSgt4uufX0OBV5tectPCOX5+ZXMjHT2D/MWrx8ZfQbKKvpmmQUfP\nAP915yEA/vmDa6jx+1yuSGRqPbhtMYvm+Nl98AJ7P253uxy5DgqFKdY/EOW7P/+Q/oEo//izK1ip\nuY0kDxQWePn651bjKyrgJy8d5XxnaPyVJCsoFKbQYCTGd3/+IW2Xwnz2pgVsXz/f7ZJEps3s2jL+\nyb2Gwcgw3/vrjzRMdYZQKEyRocgw//mvD3LqfB9b18zhH31Gk91J/tmyag73bVnIxe4Bvv/8IaIx\nne2c7RQKU2BgKMaf/dUHHDnbzYbl9fzTe1fizZLLFYpMty/esYQNy+s5crab//F/DmsajCynUMiw\nnv4h/vSn73PsXC83rZzFow+u0bxGkte8Hg9f/9xqVjZW896xDn70whEFQxbTt1UGNbcH+fc/2c/Z\nC0G23TCXr39utQJBBCguKuBf/IMbWDKvkj2H23l65yEi0WG3y5Kr0DdWBjiOw5sfnudP/ud7dAeH\n+Ad3LEl0GemymiKjSn2F/MsvradpYQ3vH+/kP/3sA3r7h9wuS66gUJikvnCE//5/DvPjF49SWODl\nt764ll+9ZREeHUMQ+ZRSXyG/82vruLlpFifO9fJvfryPYy09bpclKTQb2wTFHYc9hy7ws9dO0D8Q\nZen8Sr7+wGpdPU1kHEWFiXMYFs2p5LnXT/If//IA99zcyIO3Laa4qMDt8vKeQuE6OY7Dx2e6ee71\nk5xtD1Jc5OXLdy7nrk0N6i4SSZPH4+FXNjeyZF4lP3rhCC/tbeaA7eAfbl/KJhPQnraLFAppig3H\nOXCsg5ffbeZ0WxCAW1bP5ou3L6WuqsTl6kRmphULqvm3/+xmnt91il/sP8fTzx9i8dxK7tuykA3L\n6/VDywXjhoIxxgM8DawDBoGHrbWnUtofAJ4AosCz1tpnxlrHGLMU+DEQBw5Za7+R4deTUY7jcOp8\nH/uOXmTP4QsEw1E8wCYT4P5bFrFwjt/tEkVmPF9RAV/6zHK2r5/Pc2+c5D3bwX/deZBAdQm3rp3L\n1tVz1C07jdLZU3gQ8FlrtxpjNgNPJe/DGFOYvL0JGAB2G2P+FrhtjHWeAn7fWrvLGPN9Y8znrbV/\nm/mXNTFxx6GtM8TR5h5scze2pYdgOHFqfnlJYWKqig3zmVNb5nKlIrlndm0Z3/jCWs53hnj53Wb2\nHmnn+V2neX7XaRpnVbBmSR1L51eyZF4VVeXFbpebs9IJhduAlwCstXuNMTemtDUBx621fQDGmF3A\nHcAtV6yzKbn8JmvtruTfLwJ3A1MaCkPRYYaHHeKOw3DcIRIdJhiOEgxHCIajdAUHabsUpu1SiAtd\nYSLRyyfV1Ph93Lp2DjeaWaxeXKtzDkSmwbz6cr56XxNfvnM5+45eZL+9yNGz3TRf7B9dpr6qhAWz\nKghUl1JXVUJ9VQmLBmJEBiPUVZboszoJ6YRCJdCbcjtmjPFaa+NXaesHqgD/FfcPG2MKgNQOwmBy\n2Snzd2+f4W/ePDX+gkBxoZc5tWXMD1RgGqtZ2VhNoLpUB7xEXFLqK+T2dfO4fd08BoZinDrfx8nz\nvZw638ep8328f7zzquutbKzmX/36xmmuNnekEwp9JL7kR4wEwkhbZUqbH+geY51hY0z8imWndIDy\n4rmVrFtah8fjocDrwev1UFToxV9WhL+sGH9pEdV+H3Nry6itKsmb+Yki4T7CZNc0A+FQP4MD/YRD\nwXGX9RIhHJq+k54GB0J4vYVp1TadBsL5Mx11qa+Q1YtrWb24Fkgc7wsORLnUO0hn7yCdPQNEHejs\nCmMadZnbyUgnFHYD9wPPGWO2AAdT2o4Ay4wx1UAY2AY8mWy72joHjDG3W2vfBO4FXhvnuT2BwMQP\n5m4P+Nl+88IJr59tJrMtUn39K/dl5HFE3DQLWOp2ETnI4zjONRdIGUl0Q/Kur5I4sFyeHGn0q8Af\nkuga+qG19r9dbR1r7TFjzHLgB0ARiUD5TWvttQsQEZFpM24oiIhI/tAhehERGaVQEBGRUQoFEREZ\npVAQEZFRWTkhnjHmHHAseXOPtfYPkkNbv0tijqVXrbX/zrUCp9F4c0/lA2PMe1w+GfI08CfMoDm0\nMiE5Xcy3rbU7xppDzBjzm8AjJD4jf2ytfcGteqfSFdtiPfB3XP6++L619ue5vi2SUwz9CFgEFAN/\nDHxMBt4XWbenkHzDv2et/Uzyvz9INn0f+LK1dhuw2Rizzr0qp9Xo3FPA4yTmj8obxhgfQMr74Z9x\neQ6tOwCvMebzrhY5xYwx3yIxlNuXvOtTr98YMxv4FySmmPkV4D8YY4pcKXgKXWVbbAK+k/L++Hme\nbIvfADqttbeTeI3/hQy9L7JxT2ET0GCMeY3ECXHfBC4AxdbaM8llXgbuAj50pcLpda25p/LBOqDc\nGPMyUAD8AbBxuufQctkJ4AvA/0zevnIOsc+S+HX4lrU2BvQZY46TOE/ovekudop9alsAK4wxD5LY\nW/gmcDO5vy3+Cvh58u8CIManPxcTel+4uqdgjPmaMeagMeajkf8DbcCfWGs/A/wH4C9ITKXRl7Lq\nlM+blEWuOveUW8W4IAw8aa29B3iUxPthWufQcpu1dieJD/2IK19/JZ+eb2xkHrKccpVtsRf4VvLX\n8SkSJ9KONSdbzrDWhq21IWOMn0Q4/AEZel+4uqdgrf0RiX6xUcaYUpL/6Nba3caYuVx9jqV8ubDr\nteaeygfHSPw6xFp73BhzCUid7Syf3gsjrjaHWL5+Rp631o586T0PfA94gzzYFsaYBcDfAP/FWvu/\njTF/mtI84fdFNv7i/EPgdwCSxw1arLVBYMgYszh54PUeYNc1HiOX7AbuA7jK3FP54GvAdwCMMfNI\nvMFfMcbckWy/l/x5L4w4YIy5Pfn3yOvfB9xmjCk2xlQBK4FDbhU4jV5O6VK9k0S3SM5vi+SxgpeB\nf2Wt/Uny7vcz8b7IxmMK3wb+V3JOpSjwT5P3Pwr8JYkge8Vau8+d8qbdTuBuY8zu5O2vulmMC34I\nPJu8VkecxPvhEvBM8oDZEeA598pzxe8CP0h9/dZaxxjzPeAtEt0Iv2+tjbhZ5DR5FPjPxpgIiWOP\nj1hr+/NgWzwOVANPGGP+NeAAv01iW0zqfaG5j0REZFQ2dh+JiIhLFAoiIjJKoSAiIqMUCiIiMkqh\nICIioxQKIiIySqEgIiKjFAoiIjLq/wO+R/IyYANtTwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x23edc940>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(f_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x23cf8a20>"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAECCAYAAAARlssoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x240dd518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(f_+m_+n_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### All data for Germany. Distribution of involved women"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x22cdb358>"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAECCAYAAAARlssoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkaj/D7MLOepHvwykAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2339ce80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_all_subgroups=pd.read_csv(\"de/labourmarket_subgroups.csv\",sep='\\t', encoding='utf-8',index_col=0)\n",
    "all_frauen_ratio=df_all_subgroups.Frauen_ratio.values\n",
    "sns.distplot(all_frauen_ratio)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Tests of Significance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The null hypothesis is that all instances of X and Y are independent and identically distributed (iid).\n",
    "\n",
    "Let us take as the alternate hypothesis that (a) all instances of X are iid according to some underlying distribution F_X and (b) all instances of Y are iid according to some underlying distribution F_Y but (c) F_X differs from F_Y. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Mann-Whitney-Wilcoxon (MWW) RankSum test\n",
    "The MWW RankSum test is a useful test to determine if two distributions are significantly different or not."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MWW RankSum P for male and female distributions = 0.00102121367586 z_stat= -3.28461622826\n",
      "MWW RankSum P for male and neutral distributions = 0.0557839582946 z_stat= 1.91271960265\n",
      "MWW RankSum P for neutral and female distributions = 0.000207128266371 z_stat= -3.7101600418\n",
      "MWW RankSum P for all together and female distributions = 0.00106014461881 z_stat= -3.27405903129\n",
      "MWW RankSum P for all together and neutral distributions = 0.0546253733428 z_stat= 1.92184413219\n",
      "MWW RankSum P for all together and male distributions = 0.974152243293 z_stat= 0.0324010272497\n"
     ]
    }
   ],
   "source": [
    "from scipy import stats  \n",
    "  \n",
    "z_stat, p_val = stats.ranksums(m_, f_)  \n",
    "print \"MWW RankSum P for male and female distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(m_, n_)  \n",
    "print \"MWW RankSum P for male and neutral distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_, f_)  \n",
    "print \"MWW RankSum P for neutral and female distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_+m_+f_, f_)  \n",
    "print \"MWW RankSum P for all together and female distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_+f_+m_, n_)  \n",
    "print \"MWW RankSum P for all together and neutral distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_+f_+m_, m_)  \n",
    "print \"MWW RankSum P for all together and male distributions =\", p_val , \"z_stat=\",z_stat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MWW RankSum P for those that have wikipedia page and all Germany labour market distributions = 0.492401303559 z_stat= 0.686494544192\n",
      "MWW RankSum P for those that have female wikipedia page and all Germany labour market distributions = 0.000750774810638 z_stat= 3.37033269226\n",
      "MWW RankSum P for those that have male wikipedia page and all Germany labour market distributions = 0.518994425105 z_stat= 0.644896245275\n",
      "MWW RankSum P for those that have neutral wikipedia page and all Germany labour market distributions = 0.0963858092114 z_stat= -1.6626335592\n"
     ]
    }
   ],
   "source": [
    "z_stat, p_val = stats.ranksums(n_+f_+m_, all_frauen_ratio)  \n",
    "print \"MWW RankSum P for those that have wikipedia page and all Germany labour market distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(f_, all_frauen_ratio)  \n",
    "print \"MWW RankSum P for those that have female wikipedia page and all Germany labour market distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(m_, all_frauen_ratio)  \n",
    "print \"MWW RankSum P for those that have male wikipedia page and all Germany labour market distributions =\", p_val , \"z_stat=\",z_stat\n",
    "z_stat, p_val = stats.ranksums(n_, all_frauen_ratio)  \n",
    "print \"MWW RankSum P for those that have neutral wikipedia page and all Germany labour market distributions =\", p_val , \"z_stat=\",z_stat\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "|pair  | p-val | z |\n",
    "| :-|:-:|:-:|\n",
    "| male & female |    0.001     |    -3.28    | \n",
    "| male & neutral |     0.056     |    1.91    |  \n",
    "| neutral & female |     2.07e-04     |      -3.71    | \n",
    "| all_that_have_Wiki_page & female |     0.001    |     -3.27    | \n",
    "| all_that_have_Wiki_page & neutral |     0.055    |     1.92    | \n",
    "| all_that_have_Wiki_page & male |     0.974    |     0.03    | \n",
    "| all_that_have_Wiki_page & German labor market |     0.492    |     0.69    | \n",
    "| female & German labor market |     7.50e-04    |     3.37    | \n",
    "| male & German labor market |     0.51    |     0.64    | \n",
    "| neutral & German labor market |     0.096    |     -1.66    | "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "|pair  | p-val | z |\n",
    "| :-|:-:|:-:|\n",
    "| male & female |    0.001     |    -3.28    | \n",
    "| male & neutral |     0.056     |    1.91    |  \n",
    "| neutral & female |     2.07e-04     |      -3.71    | \n",
    "| all_that_have_Wiki_page & German labor market |     0.492    |     0.69    | \n",
    "| female & German labor market |     7.50e-04    |     3.37    | \n",
    "| male & German labor market |     0.51    |     0.64    | \n",
    "| neutral & German labor market |     0.096    |     -1.66    | "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# the Kolmogorov-Smirnov test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ks_2sampResult(statistic=0.20788562033938118, pvalue=0.054473445111893663)\n",
      "Ks_2sampResult(statistic=0.3867653103164817, pvalue=0.0070933955390278169)\n",
      "Ks_2sampResult(statistic=0.5, pvalue=0.0021245199617610125)\n",
      "Ks_2sampResult(statistic=0.0048076038285288503, pvalue=1.0000000000000002)\n",
      "Ks_2sampResult(statistic=0.38423268274014549, pvalue=0.0075745918499930624)\n",
      "Ks_2sampResult(statistic=0.20307801651085233, pvalue=0.063497984844993915)\n"
     ]
    }
   ],
   "source": [
    "from scipy.stats import ks_2samp,ttest_1samp,kstest\n",
    "print ks_2samp(m_, n_)\n",
    "print ks_2samp(m_, f_)\n",
    "print ks_2samp(f_, n_)\n",
    "print ks_2samp(m_+n_+f_, m_)\n",
    "print ks_2samp(m_+n_+f_, f_)\n",
    "print ks_2samp(m_+n_+f_, n_)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}