jing-jin-mc/A_Simple_AB_test.ipynb

## A_Simple_AB_test.ipynb
{
  "cells": [
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "# A Simple AB Test Analysis "
    },
    {
      "metadata": {
        "code_folding": [
          0
        ],
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Imports the Google Cloud client library\nfrom google.cloud import bigquery\nimport pandas as pd\nimport numpy as np\nimport scipy.stats\nimport statsmodels.stats.power as smp\nimport statsmodels.stats.proportion as sms\nimport matplotlib.pyplot as plt\nfrom IPython.display import display\npd.options.display.float_format = '{:.2f}'.format",
      "execution_count": 22,
      "outputs": []
    },
    {
      "metadata": {
        "code_folding": [],
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Instantiates a Bigquery client\nbigquery_client = bigquery.Client()",
      "execution_count": 2,
      "outputs": []
    },
    {
      "metadata": {
        "code_folding": [],
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Congif the test ID: 'Y_sXp3zqTuS02_sp31o8Pg' and set test date range\nstart_date = '20181027'\nend_date = '20181116'\nexperiment_id = 'Y_sXp3zqTuS02_sp31o8Pg'",
      "execution_count": 3,
      "outputs": []
    },
    {
      "metadata": {
        "code_folding": [
          0
        ],
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Write the BigQuery with the correct table name\n# Change the table name accordingly \nQUERY = \"\"\"\n    WITH\n      dataset AS(\n      SELECT\n        fullvisitorId AS clientId,\n        visitId AS sessionId,\n        test.experimentVariant AS variants,\n        device.deviceCategory AS device,\n        MAX (IF (h.eCommerceAction.action_type = '3',\n            1,\n            0)\n            ) AS add_to_cart,\n        IF (MAX(totals.transactions)>0,\n          1,\n          0) AS make_a_trans,\n        IF (MAX(totals.transactionRevenue)>0,\n          MAX(totals.transactionRevenue/1000000),\n          0) AS trans_rev\n      FROM\n        `bigquery-176409.109577992.ga_sessions_*`,\n        UNNEST (hits) AS h,\n        UNNEST(h.experiment) AS test\n      WHERE\n        _TABLE_SUFFIX BETWEEN @START_DATE\n        AND @END_DATE\n        AND test.experimentId = @EXPERIMENT_ID\n      GROUP BY\n        fullvisitorId,\n        visitId,\n        variants,\n        device)\n    SELECT\n      clientId,\n      variants,\n      device,\n      SUM(add_to_cart) AS num_atc,\n      COUNT(sessionId) AS num_sess,\n      SUM(make_a_trans) AS num_trans,\n      SUM(trans_rev) AS rev\n    FROM\n      dataset\n    GROUP BY\n      clientId,\n      variants,\n      device\n    \"\"\"",
      "execution_count": 11,
      "outputs": []
    },
    {
      "metadata": {
        "code_folding": [],
        "trusted": true
      },
      "cell_type": "code",
      "source": "# config parameters and get the data from BigQuery \nquery_params = [\n    #bigquery.ScalarQueryParameter('TABLE_NAME', 'STRING', table_name),\n    bigquery.ScalarQueryParameter('START_DATE', 'STRING', start_date),\n    bigquery.ScalarQueryParameter('END_DATE', 'STRING', end_date),\n    bigquery.ScalarQueryParameter('EXPERIMENT_ID', 'STRING', experiment_id)\n]\n\njob_config = bigquery.QueryJobConfig()\njob_config.query_parameters = query_params\nquery_job = bigquery_client.query(QUERY,job_config=job_config) \ndf = query_job.to_dataframe()\ndf.head()",
      "execution_count": 12,
      "outputs": [
        {
          "data": {
            "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>clientId</th>\n      <th>variants</th>\n      <th>device</th>\n      <th>num_atc</th>\n      <th>num_sess</th>\n      <th>num_trans</th>\n      <th>rev</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>3456892674124422100</td>\n      <td>0</td>\n      <td>mobile</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0.00</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>4960553732248368850</td>\n      <td>0</td>\n      <td>mobile</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0.00</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>3369147557328129456</td>\n      <td>0</td>\n      <td>mobile</td>\n      <td>0</td>\n      <td>4</td>\n      <td>0</td>\n      <td>0.00</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>2474884608469966745</td>\n      <td>0</td>\n      <td>mobile</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0.00</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>2667782236950514782</td>\n      <td>1</td>\n      <td>mobile</td>\n      <td>0</td>\n      <td>2</td>\n      <td>0</td>\n      <td>0.00</td>\n    </tr>\n  </tbody>\n</table>\n</div>",
            "text/plain": "              clientId variants  device  num_atc  num_sess  num_trans  rev\n0  3456892674124422100        0  mobile        0         1          0 0.00\n1  4960553732248368850        0  mobile        0         1          0 0.00\n2  3369147557328129456        0  mobile        0         4          0 0.00\n3  2474884608469966745        0  mobile        0         1          0 0.00\n4  2667782236950514782        1  mobile        0         2          0 0.00"
          },
          "execution_count": 12,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "code_folding": [
          0
        ],
        "trusted": true
      },
      "cell_type": "code",
      "source": "######### Function: Get data overview ##########\ndef get_data_overview(df,test_names):    \n    result_df = []\n    result_dic = {}\n    for name in test_names:\n        test_tmp = df[df['variants'] == name]\n        test_tmp_con = test_tmp[test_tmp['num_trans']>0]\n        test_tmp_atc_con = test_tmp[test_tmp['num_atc']>0]\n        cr = float(test_tmp_con.shape[0])/float(test_tmp.shape[0])\n        atcr = float(test_tmp_atc_con.shape[0])/float(test_tmp.shape[0])\n        tmp = {\"Test\":name,\n               \"#Users\":test_tmp.shape[0],\n               \"#Sess\":sum(test_tmp['num_sess']),\n               \"∑Rev.\":sum(test_tmp['rev']),\n               \"∑ATC\":sum(test_tmp['num_atc']),\n               \"∑Trans.\":sum(test_tmp['num_trans']),\n               \"#ATC./User\":test_tmp['num_atc'].mean(),\n               \"Rev./Trans\":float(sum(test_tmp['rev']))/float(sum(test_tmp['num_trans'])),\n               \"Rev./User\":test_tmp['rev'].mean(),\n               \"#Trans/User.\":test_tmp['num_trans'].mean(),\n               \"#Sess/User\":test_tmp['num_sess'].mean(),\n               \"#ATC\":test_tmp_atc_con.shape[0],\n               \"#Conveted\":test_tmp_con.shape[0],\n               \"CR for User\":cr}\n        result_df.append(tmp)\n        result_dic[name] = tmp\n    result_df = pd.DataFrame(result_df)\n    result_df = result_df[['Test',\n                           '#Sess',\n                           '#Users',\n                           \"#ATC\",\n                           '#Conveted',\n                           '∑ATC',\n                           '∑Trans.',\n                           '∑Rev.',\n                           '#Sess/User',\n                           '#ATC./User',\n                           '#Trans/User.',\n                           'Rev./User',\n                           'Rev./Trans',\n                           'CR for User']]\n    return (result_df,result_dic)        ",
      "execution_count": 6,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "* ## KPIs for this AB test"
    },
    {
      "metadata": {
        "code_folding": [],
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Get the over view for the AB test\n# Step 1 get the general one\ntest_names = list(df['variants'].unique())\nprint ('AB test overview for all devices')\nresult_df,result_dic = get_data_overview(df,test_names)\ndisplay(result_df)\n# Step 2 get the overview for different devices\ndevice_list = list(df['device'].unique())\nfor i in range(len(device_list)):\n    df_tmp = df[df['device']==device_list[i]]\n    result_df_tmp,result_dic_tmp = get_data_overview(df_tmp,test_names)\n    print (device_list[i] + ' : ' + 'AB test overview')\n    display(result_df_tmp)",
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": "AB test overview for all devices\n",
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "  Test   #Sess  #Users   #ATC  #Conveted   ∑ATC  ∑Trans.       ∑Rev.  \\\n0    0  110697   42534  26442      19062  51377    33506 38398788.37   \n1    1  112065   42221  26172      18895  51458    32732 36974001.34   \n\n   #Sess/User  #ATC./User  #Trans/User.  Rev./User  Rev./Trans  CR for User  \n0        2.60        1.21          0.79     902.78     1146.03         0.45  \n1        2.65        1.22          0.78     875.73     1129.60         0.45  ",
            "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Test</th>\n      <th>#Sess</th>\n      <th>#Users</th>\n      <th>#ATC</th>\n      <th>#Conveted</th>\n      <th>∑ATC</th>\n      <th>∑Trans.</th>\n      <th>∑Rev.</th>\n      <th>#Sess/User</th>\n      <th>#ATC./User</th>\n      <th>#Trans/User.</th>\n      <th>Rev./User</th>\n      <th>Rev./Trans</th>\n      <th>CR for User</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>110697</td>\n      <td>42534</td>\n      <td>26442</td>\n      <td>19062</td>\n      <td>51377</td>\n      <td>33506</td>\n      <td>38398788.37</td>\n      <td>2.60</td>\n      <td>1.21</td>\n      <td>0.79</td>\n      <td>902.78</td>\n      <td>1146.03</td>\n      <td>0.45</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1</td>\n      <td>112065</td>\n      <td>42221</td>\n      <td>26172</td>\n      <td>18895</td>\n      <td>51458</td>\n      <td>32732</td>\n      <td>36974001.34</td>\n      <td>2.65</td>\n      <td>1.22</td>\n      <td>0.78</td>\n      <td>875.73</td>\n      <td>1129.60</td>\n      <td>0.45</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "text": "mobile : AB test overview\n",
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "  Test  #Sess  #Users  #ATC  #Conveted   ∑ATC  ∑Trans.       ∑Rev.  \\\n0    0  46280   15271  9160       5958  19731    10371 11584811.53   \n1    1  46467   15345  9092       5934  19654    10169 11353816.45   \n\n   #Sess/User  #ATC./User  #Trans/User.  Rev./User  Rev./Trans  CR for User  \n0        3.03        1.29          0.68     758.62     1117.04         0.39  \n1        3.03        1.28          0.66     739.90     1116.51         0.39  ",
            "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Test</th>\n      <th>#Sess</th>\n      <th>#Users</th>\n      <th>#ATC</th>\n      <th>#Conveted</th>\n      <th>∑ATC</th>\n      <th>∑Trans.</th>\n      <th>∑Rev.</th>\n      <th>#Sess/User</th>\n      <th>#ATC./User</th>\n      <th>#Trans/User.</th>\n      <th>Rev./User</th>\n      <th>Rev./Trans</th>\n      <th>CR for User</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>46280</td>\n      <td>15271</td>\n      <td>9160</td>\n      <td>5958</td>\n      <td>19731</td>\n      <td>10371</td>\n      <td>11584811.53</td>\n      <td>3.03</td>\n      <td>1.29</td>\n      <td>0.68</td>\n      <td>758.62</td>\n      <td>1117.04</td>\n      <td>0.39</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1</td>\n      <td>46467</td>\n      <td>15345</td>\n      <td>9092</td>\n      <td>5934</td>\n      <td>19654</td>\n      <td>10169</td>\n      <td>11353816.45</td>\n      <td>3.03</td>\n      <td>1.28</td>\n      <td>0.66</td>\n      <td>739.90</td>\n      <td>1116.51</td>\n      <td>0.39</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "text": "tablet : AB test overview\n",
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "  Test  #Sess  #Users  #ATC  #Conveted  ∑ATC  ∑Trans.      ∑Rev.  #Sess/User  \\\n0    0  14845    5949  4286       3421  7733     5866 6664929.71        2.50   \n1    1  15851    6004  4260       3443  8219     5907 6607298.65        2.64   \n\n   #ATC./User  #Trans/User.  Rev./User  Rev./Trans  CR for User  \n0        1.30          0.99    1120.34     1136.20         0.58  \n1        1.37          0.98    1100.48     1118.55         0.57  ",
            "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Test</th>\n      <th>#Sess</th>\n      <th>#Users</th>\n      <th>#ATC</th>\n      <th>#Conveted</th>\n      <th>∑ATC</th>\n      <th>∑Trans.</th>\n      <th>∑Rev.</th>\n      <th>#Sess/User</th>\n      <th>#ATC./User</th>\n      <th>#Trans/User.</th>\n      <th>Rev./User</th>\n      <th>Rev./Trans</th>\n      <th>CR for User</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>14845</td>\n      <td>5949</td>\n      <td>4286</td>\n      <td>3421</td>\n      <td>7733</td>\n      <td>5866</td>\n      <td>6664929.71</td>\n      <td>2.50</td>\n      <td>1.30</td>\n      <td>0.99</td>\n      <td>1120.34</td>\n      <td>1136.20</td>\n      <td>0.58</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1</td>\n      <td>15851</td>\n      <td>6004</td>\n      <td>4260</td>\n      <td>3443</td>\n      <td>8219</td>\n      <td>5907</td>\n      <td>6607298.65</td>\n      <td>2.64</td>\n      <td>1.37</td>\n      <td>0.98</td>\n      <td>1100.48</td>\n      <td>1118.55</td>\n      <td>0.57</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "text": "desktop : AB test overview\n",
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "  Test  #Sess  #Users   #ATC  #Conveted   ∑ATC  ∑Trans.       ∑Rev.  \\\n0    0  49572   21314  12996       9683  23913    17269 20149047.13   \n1    1  49747   20872  12820       9518  23585    16656 19012886.24   \n\n   #Sess/User  #ATC./User  #Trans/User.  Rev./User  Rev./Trans  CR for User  \n0        2.33        1.12          0.81     945.34     1166.78         0.45  \n1        2.38        1.13          0.80     910.93     1141.50         0.46  ",
            "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Test</th>\n      <th>#Sess</th>\n      <th>#Users</th>\n      <th>#ATC</th>\n      <th>#Conveted</th>\n      <th>∑ATC</th>\n      <th>∑Trans.</th>\n      <th>∑Rev.</th>\n      <th>#Sess/User</th>\n      <th>#ATC./User</th>\n      <th>#Trans/User.</th>\n      <th>Rev./User</th>\n      <th>Rev./Trans</th>\n      <th>CR for User</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>49572</td>\n      <td>21314</td>\n      <td>12996</td>\n      <td>9683</td>\n      <td>23913</td>\n      <td>17269</td>\n      <td>20149047.13</td>\n      <td>2.33</td>\n      <td>1.12</td>\n      <td>0.81</td>\n      <td>945.34</td>\n      <td>1166.78</td>\n      <td>0.45</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1</td>\n      <td>49747</td>\n      <td>20872</td>\n      <td>12820</td>\n      <td>9518</td>\n      <td>23585</td>\n      <td>16656</td>\n      <td>19012886.24</td>\n      <td>2.38</td>\n      <td>1.13</td>\n      <td>0.80</td>\n      <td>910.93</td>\n      <td>1141.50</td>\n      <td>0.46</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "code_folding": [
          0
        ],
        "trusted": true
      },
      "cell_type": "code",
      "source": "######### Function: Data Visualization ######### \ndef visionlise_data(df,percentile,test_names,kpi,bin):\n    plt.figure(figsize=(10,5))\n    upper = df[kpi].max()\n    pp = int(np.percentile(df[kpi],percentile))\n    lower = df[kpi].min()\n    print (\"The maximun \"+ kpi + \" of a user : %d\"%upper)\n    print (str(percentile) + \"% users with \"+ kpi + \" below: %d\"%pp)\n    print (\"The minimun \"+ kpi + \" of a user : %d\"%lower)\n    print ('********** Data Visualization for ' + str(percentile) + '% users ******')\n    df_c = df[df[kpi] <= pp]\n    bins = range(int(lower+bin),int(pp+bin),bin)\n    for i in range(len(test_names)):\n        data_tmp = df_c[df_c['variants'] == test_names[i]]\n        data_tmp[kpi].plot(kind = 'hist',bins = bins,alpha = 0.5,label = test_names[i])  \n    plt.ylabel('Frequency' ,\n               fontsize = 14)\n    plt.title('AB Test: Distribution of '+kpi,\n              fontsize = 14)  \n    plt.grid()\n    plt.legend()\n    plt.show()\n    return (lower,pp,upper)",
      "execution_count": 13,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "######## Function: Z-test for Conversion Rate #######\ndef z_test_for_cr(result_dic,Hypothesis,alpha,test_names):\n    print (\"The hypothes is: CR in the variant part is \" + Hypothesis + \" than the control part.\")\n    count = np.array([result_dic[test_names[0]][\"#Conveted\"],result_dic[test_names[1]][\"#Conveted\"]])\n    nobs = np.array([result_dic[test_names[0]][\"#Users\"],result_dic[test_names[1]][\"#Users\"]])\n    z_score,p_value = sms.proportions_ztest(count,nobs,value = 0,alternative = Hypothesis)\n    print (\"z_score : %.4f\"%z_score)\n    print (\"p_value : %.4f\"%p_value)\n    if p_value > alpha:\n        print (\"The result is not significant. The hypothesis can not be nagted!\")\n    else: \n        print (\"The result is significant. The hypothesis can be nagted! \")\n        \n    es = sms.proportion_effectsize(float(count[0])/nobs[0], float(count[1])/nobs[1], method='normal')\n    power = smp.NormalIndPower().solve_power(es, nobs1=nobs[0], ratio = float(nobs[1])/nobs[0], alpha = alpha, alternative=Hypothesis)\n    print (\"power : %.4f\"%power)\n    return ",
      "execution_count": 25,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Check the conversion rate for the desktop \ndf_tmp = df[df['device']=='desktop']\nresult_df_tmp,result_dic_tmp = get_data_overview(df_tmp,test_names)\n# Significant level \nalpha = 0.1\n# Define hypothesis\nHypothesis = 'smaller'\nz_test_for_cr(result_dic,Hypothesis,alpha,test_names)",
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": "The hypothes is: CR in the variant part is smaller than the control part.\nz_score : 0.1853\np_value : 0.5735\nThe result is not significant. The hypothesis can not be nagted!\npower : 0.0712\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "code_folding": [
          0
        ],
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Mannwhitney_test \ndef Mannwhitney_test_for_values(df,Hypothesis,alpha,test_names,kpi):\n    print (\"The hypothes is: \" + kpi +\" in the variant part is \" + Hypothesis + \" than the control part.\")\n    for i in range(len(test_names)-1):\n        data_A = df[df['variants'] == test_names[0]]\n        data_B = df[df['variants'] == test_names[i+1]]\n        u,p_value = scipy.stats.mannwhitneyu(data_A[kpi],data_B[kpi],alternative = Hypothesis)\n        print (\"The p-value is: %.4f\"%p_value)\n        if p_value > alpha:\n            print (\"The result is not significant. The hypothesis can not be nagted!\")\n        else: \n            print (\"The result is significant. The hypothesis can be nagted! \")\n    return ",
      "execution_count": 14,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "* ## Significance Analysis for Revenue per User"
    },
    {
      "metadata": {
        "code_folding": [],
        "scrolled": false,
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Check the KPI (rev per user) distribution for thr test \npercentile = 98\nkpi = 'rev'\nbins = 100\n# Significant level \nalpha = 0.1\n# Define hypothesis\nHypothesis = \"greater\" #\"less\", \"two-sided\", \"greater\"\n# Step 1 get the general one\nprint ('For all devices')\nlower,pp,upper = visionlise_data(df,percentile,test_names,kpi,bins)\n\ndf_c = df[df[kpi] <= pp] # Here you remove the 1 percent data\nMannwhitney_test_for_values(df_c,Hypothesis,alpha,test_names,kpi)\nprint ()\n\n# Step 2 get the overview for different devices\nfor i in range(len(device_list)):\n    df_tmp = df[df['device']==device_list[i]]\n    print ('For ' + device_list[i] )\n    #plt.subplot(1,3,i+1)\n    lower_tmp,pp_tmp,upper_tmp = visionlise_data(df_tmp,percentile,test_names,kpi,bins)\n    df_c_tmp = df_tmp[df_tmp[kpi] <= pp] # Here you remove the 1 percent data\n    Mannwhitney_test_for_values(df_c_tmp,Hypothesis,alpha,test_names,kpi)\n    print ()",
      "execution_count": 16,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": "For all devices\nThe maximun rev of a user : 59820\n98% users with rev below: 5427\nThe minimun rev of a user : 0\n********** Data Visualization for 98% users ******\n"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<Figure size 720x360 with 1 Axes>"
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": "The hypothes is: rev in the variant part is greater than the control part.\nThe p-value is: 0.2654\nThe result is not significant. The hypothesis can not be nagted!\n\nFor mobile\nThe maximun rev of a user : 33698\n98% users with rev below: 5097\nThe minimun rev of a user : 0\n********** Data Visualization for 98% users ******\n"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmsAAAFACAYAAADjzzuMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3XucXVV58PHfYwggSRpIMDGQ1IBcLHgJEDGpBRIRQVTQVilUBQUb28JbRa2CttZLEeyr0lqv0aB4IyJoQYtiiom+1gglgFwLhIgwBImES5MggcTn/WPvwcMwk9nnzJxz9sz8vp/P/pyz11577+eclZl5svZee0VmIkmSpHp6WrcDkCRJ0sBM1iRJkmrMZE2SJKnGTNYkSZJqzGRNkiSpxkzWJEmSasxkTdKoExFLI+KiNhx3x4jIiHhluf6ccv25w32u8vht+RytiIjXRcQdEbE1Ij7X7XikscRkTRqBIuKA8o/mfw2wPRuWxyPizog4JyK2G6D+gj779Le8aRji/quIuL/Ffc9piGVLRKyPiJ9GxLsjYqc+1d8KvGW4Y8rMR4EZwLLmoh80hqPKzzWxz6bKn6OdIuJpwBLga8As4N3djUgaW/r9xS2p9v4S+AxwYkT8UWbeMkCd7wHjgYOA84EHgY/2U/dnFElIr48AzwH+tKHs4WGIe6h+ARxF8R/NKcAhwPuAN0fEIZl5P0BmDnusEbF9Zj6Wmb8e7mMPpB2fo0XTgEnADzJzbdWder+z9oUljQ32rEkjTEQ8HfgL4AvARcApA1R9KDN/nZl3Z+a/U/QGHdhfxd4kpHcBHgGeVJaZvy3Pf2jZo/XbiLg7Iv6tsUcoIg6PiKsiYlNEPBQRP4+IfSPiKOCzwNSGHrIzmvz4W8pY1mbmjZn5WWA+sBtwVkMMT7p82EpMEfHriHhvRHwlIh4Gzut7GbTB/hGxMiIejYibImJhw7mf0mvWePk0Ip4DfL/ctKEs/9wAn+PpEfGpiPhNea7/ioh5/ZxrQURcHRGPRMSVEfG8bX2pEbFrRHw9Ih4s97k8IvbtPSZwb1n1Z+Xx5w1wnKd8Z2X5syLiW+V3vz4iLo2IPcptzy+PuXefY/1tRKyNiHHbil0aC0zWpJHntcCvMvN64KsUvWvjt7VDROwH/DFw5VBOHBEHUSQWFwLPA46jSJZ6k4sdgH8H/rPcPh/4NPA74EfAe4AHKHrxZgD/Vu53TkQ82kpMmXk38E3gzwaIuaWYSu8GrqVIcj+wjTD+Gfi/wAHAT4HvRsS0ih/hdorkG+DZZQwDXWb8F+DVwBvLmG4HfhARu/ap9xHgHRQ9qo9QXL7clq8DLwBeRfH9JPD98rv7Eb9P8l9RxrdqG8d60ncWEZOAFRS9uocAfwI8BCyLiB3Kf8c3AK/vc5zXAxdk5tZBYpdGv8x0cXEZQQvwY+Bd5fsA7gT+rE+dBH4LbAQeLde/BYyreI5PASv6Kb8Q+HSfsnnl8f+AoocrgRcNcNy/Au7vp/wdwC8Giekc4OoBtr29N4ZyfSlwUfm+1Zh+DXyrT9mO5bFeWa4/p1x/Z0OdcWWb/H25flRZZ2JDnd79njtQnX4+xy7AFuC4hu3jgbv7OddhDXUOL8t2HeDzP6/cfnBD2VRgE/CGcn1mWWfeIG3U33f2N8CNfcrGAxuAY8r19wC3N2zfqzzfAd3+eXNxqcNiz5o0gkTEXsCLgW8AZGZS9Ir0dxP63wFzKHpMXgk8n+K+taE4CHhLRGzsXYArym3PzuJ+pqXAioj4bkS8PSJ2H+ygmfmJzHzBEOKK3kP1c+yWYipdXbHeyobzbQX+G9iv4r5V7U2RCD4xqCQzH6foLe17rusb3vfeYzZQT98fAY9RxNx73PXALf0ct4q+39lBwHP6/Jt5EJhA0ZMIxb/hPSPi4HL99cDNmXltC+eXRh0HGEgjy1so/mDfFdGbnxSJSkTMyuKSYK9fZ+bq8v2t5eWoCyLiHzPzjhbP/zSKS4if6Wfb3QCZeUJEHAgcSXFp8iMR8YrMXN7iOavYj6J3bEN/G4cQ06ZhiO135Ws0lG3zsvUABkxI+yl7vJ9tA/3nPAYoH+hcg+n7nT2NIqE8qZ+6vQNCeiLiJxRJ2lUUl4W/3MK5pVHJnjVphIjisRsnAWdS9Jj1Li+g6El58yCH6L33p+9jLppxDbB/Zq7uZ9ncWykzr8nMszPzEIo/1G8sNz1GkWwOm4iYRXHv3DafR9bmmBpv8h8HzKXomQL4TfnaONp2Tp/9e0dMbiuO2yja8E8azjUeeBFwc/MhP+FmYHvghQ3HnUrR4zaU4/a6BtgHuK+ffzMPNdT7GvDnETGfohfxG8NwbmlUMFmTRo5XALsCX8hiJOQTC8VlvpOjeB5Wr50j4pkRsVtEHAa8n+IPfn+P+ajqI8CCiPhkRMyJiL0j4piI+DRAOcLyrIiYX44AfClFr1fvH/07gckRcVg5AvHp5X7viIhfVDj/duVnmlGOpPxrikuQa4F/6G+HVmNq0t9GxKvLkZ2forjkuLjcdjPFvVwfKr+vlwN9R8HeWb6+MiKeERET+p4gMx8Evgh8PCKOLAeNfIHikRqfbyHm3uPeAFwOLImIF0fECygSp/so7nMcqvMp7k/794g4JCL2KL/rf42IZzXUuwiYTDE69/9l5q+G4dzSqGCyJo0cpwDLy/uJ+voW8CzgpQ1lX6B45EIPcAFwE/DyzNzSagCZuQo4jKLX5acUo/7+iSIZgWJAw/7AxRSJ4RfL5dxy+3LgS8C3KXqc3laWTwP2rRDCC8rPdDfwE+ANFKM352b5jLV+tBpTM84A3kvxHLjDKG6cvw+g7HE8vozheornwr23cefMXEPx6JFPUCRJHx/gPKcDl1IkU709Vkdt47NX9YYytv+gSH6fRvFvZcjPSMvM/6XoDVxL8R3fQvF970TDs/uyeKbcdynaeLDRq9KYEsX9yZIkSaoje9YkSZJqzGRNkiSpxkzWJEmSasxkTZIkqcZM1iRJkmpsVM1gsOuuu+bs2bNb2nfTpk1MmPCURxupy2yX+rFN6sl2qR/bpJ7q1C6rVq26PzOfMVi9UZWszZ49m6uvrjqV35OtWLGCBQsWDG9AGjLbpX5sk3qyXerHNqmnOrVLRFR6+LOXQSVJkmrMZE2SJKnGTNYkSZJqbFTdsyZJksaWxx9/nJ6eHh599NFK9SdPnswtt9zS5qiebMcdd2TmzJmMHz++pf1N1iRJ0ojV09PDpEmTmD17NhExaP0NGzYwadKkDkRWyEzWr19PT08Pe+yxR0vH8DKoJEkasR599FGmTp1aKVHrhohg6tSplXv++mOyJkmSRrS6Jmq9hhqfyZokSdIQ/OAHP2Dfffdlr7324pxzzhn243vPmiRJGjXOXXbbNrc/9thmtt9+h8rHO/2Ifba5fevWrZx66qksW7aMmTNn8sIXvpBjjjmG/fbbr/I5BmPPmiRJUouuuuoq9tprL/bcc0+23357jj/+eC655JJhPYfJmiRJUovuueceZs2a9cT6zJkzueeee4b1HF4G1fBafna1egvPrFRt3YbNg3Zpw+Dd1JIktUNmPqVsuAc82LMmSZLUopkzZ3L33Xc/sd7T08Nuu+02rOcwWZMkSWrRC1/4Qm6//XZ++ctf8thjj7F06VKOOeaYYT2Hl0ElSZJatN122/GpT32KI488kq1bt3LyySez//77D+85hvVoGvNWrllfqd78hW0ORJI0Jg12D3M7pps6+uijOfroo4f1mI28DCpJklRjJmuSJEk1ZrImSZJUYyZrkiRJNWayJkmSVGOOBtXgqs5KIEmShp09a5IkSUNw8sknM23aNJ773Oe25fj2rKk7KvfWHdjWMCRJo8wgf1+2f2wzbL9D9eNVmMv6TW96E6eddhonnnhi9eM2wZ41SZKkITj00EOZMmVK245vsiZJklRjJmuSJEk1ZrImSZJUYyZrkiRJNWayJkmSNAQnnHAC8+fP59Zbb2XmzJksWbJkWI/vozskSdLoMcijNh7bsIEdJk0a1lNecMEFw3q8vuxZkyRJqrGOJWsRsWNEXBURv4iImyLig2X5HhFxZUTcHhHfjIjty/IdyvXV5fbZnYpVkiSpLjrZs7YZeElmvgCYAxwVEfOAjwLnZubewIPAKWX9U4AHM3Mv4NyyniRJ0pjSsXvWMjOBjeXq+HJJ4CXAX5Tl5wMfAD4LHFu+B7gI+FRERHkcjXAr16yvVG/C5N8wb92yCjU/NrSAJEkjVmYSEd0OY0BDTV06es9aRIyLiOuAdcAy4A7goczcUlbpAXYv3+8O3A1Qbn8YmNrJeCVJUr3tuOOOrF+/fsgJUbtkJuvXr2fHHXds+RjRjQ8XETsD3wHeD3ypvNRJRMwCLsvM50XETcCRmdlTbrsDODgz1/c51iJgEcD06dMPWrp0aUsxbdy4kYkTJ7b6kUa3Db+uXHXT5i2DV2rClnE7sd3WRwY/7w7PqHS8aZOamLxX/fJnpZ5sl/qxTTojIpgwYQLjxo2rVL8bvXBbt25l06ZNT0koFy5cuCoz5w62f1ce3ZGZD0XECmAesHNEbFf2ns0E1pbVeoBZQE9EbAdMBh7o51iLgcUAc+fOzQULFrQU04oVK2h135Hq3GW3Vap3+sSVlY+5cl21y5tVPTB5DlMevm7Qerf94aJKxztuwT5DDWnMG4s/KyOB7VI/tkk9jcR26eRo0GeUPWpExNOBlwK3AMuB15bVTgIuKd9fWq5Tbv+R96tJkqSxppM9azOA8yNiHEWSeGFmfi8ibgaWRsQ/AdcCvY/9XQJ8NSJWU/SoHd/BWCVJkmqhk6NBrwcO6Kd8DXBwP+WPAq/rQGiSJEm15QwGkiRJNWayJkmSVGMma5IkSTVmsiZJklRjXXnOmkaWqlNDSZKk4WfPmiRJUo2ZrEmSJNWYyZokSVKNmaxJkiTVmAMMNLYsP7tavYVntjcOSZIqsmdNkiSpxkzWJEmSasxkTZIkqcZM1iRJkmrMZE2SJKnGHA2qMaXq1FnzF7Y5EEmSKrJnTZIkqcZM1iRJkmrMZE2SJKnGvGdtDJt31+JuhyBJkgZhz5okSVKNmaxJkiTVmMmaJElSjXnPmkYF77+TJI1W9qxJkiTVmMmaJElSjZmsSZIk1ZjJmiRJUo2ZrEmSJNWYyZokSVKNdSxZi4hZEbE8Im6JiJsi4m1l+Qci4p6IuK5cjm7Y58yIWB0Rt0bEkZ2KVZIkqS46+Zy1LcA7M/OaiJgErIqIZeW2czPzY42VI2I/4Hhgf2A34D8jYp/M3NrBmCVJkrqqY8laZt4L3Fu+3xARtwC7b2OXY4GlmbkZ+GVErAYOBla2PVhp+dnV6i08s71xSJLGvK7MYBARs4EDgCuBFwOnRcSJwNUUvW8PUiRyP2/YrYdtJ3fSsFm5Zn2levMXtjkQSdKYF5nZ2RNGTAR+DJyVmd+OiOnA/UACHwZmZObJEfFpYGVmfq3cbwlwWWZe3Od4i4BFANOnTz9o6dKlLcW1ceNGJk6c2OrHGpE23d/T7RAGtWXcTmy39ZFuhzGgCbvO7HYIHTcWf1ZGAtulfmyTeqpTuyxcuHBVZs4drF5He9YiYjxwMfD1zPw2QGbe17D9C8D3ytUeYFbD7jOBtX2PmZmLgcUAc+fOzQULFrQU24oVK2h135Fq5ZJ3dTuEQT0weQ5THr6u22EMaP5r39DtEDpuLP6sjAS2S/3YJvU0Etulk6NBA1gC3JKZn2gon9FQ7TXAjeX7S4HjI2KHiNgD2Bu4qlPxSpIk1UEne9ZeDLwRuCEiertK3gucEBFzKC6D3gm8FSAzb4qIC4GbKUaSnupIUEmSNNZ0cjToT4HoZ9Nl29jnLOCstgUlSZJUc85gIEmSVGMma5IkSTVmsiZJklRjJmuSJEk1ZrImSZJUYyZrkiRJNWayJkmSVGNdmchdGovOXXZbpXqnH7FPmyORJI0k9qxJkiTVmMmaJElSjZmsSZIk1ZjJmiRJUo2ZrEmSJNWYyZokSVKNmaxJkiTVmM9ZG4WqPs9rXpvjkCRJQ2eyJnXIvLsWV6z5sbbGIUkaWbwMKkmSVGP2rI1C1XtwJElS3dmzJkmSVGNNJWsRMaVdgUiSJOmpmu1ZWxsRSyPiiLZEI0mSpCdp9p611wBvBr4bEfcB5wFfzsxfDXtk0giwcsm7uh2CJGmUa6pnLTO/n5nHAbsBnwBeDdwRET+MiD+PiO3bEaQkSdJY1dIAg8x8IDP/NTMPAN4OHApcQHGZ9J8iYqfhDFKSJGmsaunRHRExDTiR4pLobOAiYAlFj9sZwMHAy4YnREmSpLGrqWQtIo4BTgZeDtwKfB74amY+2FDnOuDa4QxSkiRprGq2Z+3rwDeBQzPzygHqrAE+OqSoJEmSBDSfrM3IzI3bqpCZvwX+ofWQJEmS1KvZAQYvi4hX9S2MiFdFxGuGKSZJkiSVmk3WPgQ83k/5o8CHt7VjRMyKiOURcUtE3BQRbyvLp0TEsoi4vXzdpSyPiPhkRKyOiOsj4sAmY5UkSRrxmr0M+mzgf/opvx3Yc5B9twDvzMxrImISsCoilgFvAq7IzHMi4gyK0aTvoRjEsHe5vAj4bPkqjW7Lz65Wb+GZ7Y1DklQLzfasPUSRsPW1N7BhWztm5r2ZeU35fgNwC7A7cCxwflntfIoH7VKWfyULPwd2jogZTcYrSZI0ojWbrF0KnBsRTyRsEbEX8PFyWyURMRs4ALgSmJ6Z90KR0AHTymq7A3c37NZTlkmSJI0ZkZnVK0dMBi4HDqJIngBmAtcAR2bmQxWOMRH4MXBWZn47Ih7KzJ0btj+YmbtExH8AZ2fmT8vyK4B3Z+aqPsdbBCwCmD59+kFLly6t/Hkabdy4kYkTJ7a0b91sur9n8EojxJZxO7Hd1ke6HUZHTdih4t0Jk57Z3kAGMJp+VkYT26V+bJN6qlO7LFy4cFVmzh2sXlP3rGXmwxExHzgKmAMERaJ2eVbI+iJiPHAx8PXM/HZZfF9EzMjMe8vLnOvK8h5gVsPuM4G1/cS0GFgMMHfu3FywYEEzH+kJK1asoNV962Y0TS7+wOQ5THn4um6H0VHz95xareKC49sbyABG08/KaGK71I9tUk8jsV2anm6qTMq+Xy6VRURQTEl1S2Z+omHTpcBJwDnl6yUN5adFxFKKgQUP914ulSRJGiuaTtYiYi7wEop7y550z1tmvmMbu74YeCNwQzklFcB7KZK0CyPiFOAu4HXltsuAo4HVwCMU85BKkiSNKc3ODXo6xWCCOykuSTZe+tzmZdDy3rMYYPPh/dRP4NRm4pMkSRptmu1ZOx14R2b+SzuCkSRJ0pM1++iOyTTxiA5JkiQNTbPJ2oXAy9oRiCRJkp6q2cugdwAfjoh5wA30mSc0Mz85XIFJkiSp+WTtVIpJ2w/nqYMCEjBZkyRJGkbNPhR31uC1JEmSNFyavWftCRExtXzQrSRJktqk2eesjQc+CPwNMBHYB1gTEWcDv8rMzw1/iJL6tfzsavUWntneOCRJbdXsPWv/APwZcArwlYbyVcDfASZr0hCtXLO+Ur3Kc4hKkka0Zi+Dvh54a2ZeDPyuofwGYN9hi0qSJElA88nabhRTTfU1jhbmGZUkSdK2NZus3Qwc0k/564Brhx6OJEmSGjXbG/Yh4MsRsRtFovenEbEvcCLwquEOTpIkaaxrqmctMy+huG/tGIpLn2cBzwNenZk/HP7wJEmSxram7zPLzMuAy9oQiyRJkvpo+aG4kiRJar9mH4r7IMUcoP3KzClDjkiSJElPaPYy6Lv6rI8HDgBeDVR8nLokSZKqanYi9yX9lUfE1cBhwxKRBnTustsq1ZvX5jgkSVLnDNc9a1cAxw7TsSRJklQarmTtdUC1CQ0lSZJUWbMDDK7lyQMMAngm8AzgtGGMS/2Yd9fibocgSZI6rNkBBt/rs/474DfA8sy8aXhCkiRJUq9mBxj8Q7sCkSRJ0lM1PYOBpHpYuababaLzF7Y5EElSWzV7z9rjbOOhuI0yc/uWIpIkSdITmu1ZeyfwfuC7wMqybD7wKuADFPevSZIkaZg0m6y9BHhfZn6+oWxxRPwVcHRmHjN8oUmSJKnZ56y9lOIBuH39J3D40MORJElSo2aTtfXAn/ZT/hrg/qGHI0mSpEbNXgb9APDFiDiM39+zNg84Cli0rR0j4jzglcC6zHxuWfYB4C/5/b1u783My8ptZwKnAFuBv83My5uMVRLA8rOr1Vt4ZnvjkCS1pKmetcz8EnAIsBE4DvhzYBNwWGaeN8juX6ZI6vo6NzPnlEtvorYfcDywf7nPZyJiXDOxSpIkjQZNP2ctM38G/KyF/X4SEbMrVj8WWJqZm4FfRsRq4GB+35snSZI0JjQ9kXtEPCMi3h4Rn4yIqWXZvIh4VosxnBYR10fEeRGxS1m2O3B3Q52eskySJGlMicxKz7gtKkccQDEa9B5gX+A5mbkmIj4I7JWZrx9k/9nA9xruWZtOMTAhgQ8DMzLz5Ij4NLAyM79W1lsCXJaZF/dzzEWU98tNnz79oKVLl1b+PI02btzIxIkTW9q3Uzbd39PtEDpuy7id2G7rI90OY0SbsEPFDvRJz6xUbST8rIxFtkv92Cb1VKd2Wbhw4arMnDtYvWYvg34c+Exm/n1EbGgo/wHQdJaUmff1vo+IL/D7ieJ7gFkNVWcCawc4xmJgMcDcuXNzwYIFzYYBwIoVK2h1305ZueRd3Q6h4x6YPIcpD1/X7TBGtPl7Tq1WccHxlaqNhJ+Vsch2qR/bpJ5GYrs0exn0IOBL/ZSvBaY3e/KImNGw+hrgxvL9pcDxEbFDROwB7A1c1ezxJUmSRrpme9YeBf6gn/J9GWSqqYi4AFgA7BoRPcA/AgsiYg7FZdA7gbcCZOZNEXEhcDOwBTg1M7c2GaukJpy77LZK9Q4Y3+ZAJElP0myy9l3g/RFxXLmeEfGHwDnAt7e1Y2ae0E/xkm3UPws4q8n4JEmSRpVmL4O+E3gmsA54OvBjYDXwCPC+4Q1NkiRJTfWsZebDEfHHwBHAgRTJ3jXA5dnMsFJJkiRVUjlZi4jxwArg5Mz8IfDDdgUlqfPm3bW4Ur3Nz35lmyORJDWqfBk0Mx+nGJX5u/aFI0mSpEbN3rP2VYrJ1SVJktQBzY4G3R54S0QcAVxNMYn7EzLzHcMVmCRJkppP1uYA15fv9+uzzQEGUg2tXLO+2yFIkoagUrIWEc8HbszMQ9ocjyRJkhpUvWftWmDX3pWI+I8+U0VJkiSpDaoma9Fn/VCKh+JKkiSpjZodDSpJkqQOqjrAIHnqAAIHFEhj0LoNmytN+n76Eft0IBpJGv2qJmsBfC0iNpfrOwJfiIhHGitl5jHDGZwkSdJYVzVZO7/P+teGOxBJkiQ9VaVkLTPf3O5AJEmS9FQOMJAkSaqxZmcwUDssP7vbEUiSpJqyZ02SJKnGTNYkSZJqzGRNkiSpxkzWJEmSasxkTZIkqcZM1iRJkmrMR3fUwMo167sdgiRJqimTNUndV/VZgwvPbG8cklRDXgaVJEmqMXvWJDVlwubfMG/dsgo1P9b2WCRpLLBnTZIkqcbsWZPUHs55K0nDwmRNUls0M8p5/p5T2xiJJI1sHbsMGhHnRcS6iLixoWxKRCyLiNvL113K8oiIT0bE6oi4PiIO7FSckiRJddLJe9a+DBzVp+wM4IrM3Bu4olwHeDmwd7ksAj7boRglSZJqpWOXQTPzJxExu0/xscCC8v35wArgPWX5VzIzgZ9HxM4RMSMz7+1MtJI6qeol0/kL2xyIJNVQt0eDTu9NwMrXaWX57sDdDfV6yjJJkqQxpa4DDKKfsuy3YsQiikulTJ8+nRUrVrR0wo0bN7a871BtmjynK+cdCbaM24kH/H5qpZtt0q2f0ZGgm7/D1D/bpJ5GYrt0O1m7r/fyZkTMANaV5T3ArIZ6M4G1/R0gMxcDiwHmzp2bCxYsaCmQFStW0Oq+Q7Vyybu6ct6R4IHJc5jy8HXdDkMNutkm86fePXglGJPTUnXzd5j6Z5vU00hsl24na5cCJwHnlK+XNJSfFhFLgRcBD3u/mqTKnGtU0ijSsWQtIi6gGEywa0T0AP9IkaRdGBGnAHcBryurXwYcDawGHgHe3Kk4JekpTP4kdVEnR4OeMMCmw/upm8Cp7Y1IkiSp/rp9GVSSuscpsSSNAN1+dIckSZK2wWRNkiSpxrwMKmnEqDzTgRPDSxpF7FmTJEmqMXvWJGkQzl0qqZvsWZMkSaoxkzVJkqQaM1mTJEmqMe9ZkzTqOGpU0mhisiZpzKqa1ElSN3kZVJIkqcZM1iRJkmrMZE2SJKnGTNYkSZJqzGRNkiSpxhwNKknD5Nxlt1Wqd/oR+7Q5EkmjicmaJA2TeXctrljzY22NQ9Lo4mVQSZKkGjNZkyRJqjGTNUmSpBrznjVJ6rCqAxGgicEIG34Ny88evN7CMyufW1I92LMmSZJUYyZrkiRJNeZlUEmqsaqXTA9ocxySuseeNUmSpBozWZMkSaoxL4O2UdXLF/PaHIekkavqrAibp82tVM8psaSRx2RNktR2bXlciTRGeBlUkiSpxmrRsxYRdwIbgK3AlsycGxFTgG8Cs4E7geMy88FuxdiK6pM6SxpL/N0gqRm1SNZKCzPz/ob1M4ArMvOciDijXH9Pd0KTpHrbtHkLK9etH7ziH1Y73sol76pUb/4pH6t2QEktq/Nl0GOB88v35wOv7mIskiRJXVGXnrUEfhgRCXw+MxcD0zPzXoDMvDcipnU1QkkaBbwEK408kZndjoGI2C0z15YJ2TLg/wCXZubODXUezMxd+tl3EbAIYPr06QctXbq0pRg2btzIxIkTW9p3IJvu7xnW441FW8btxHZbH+l2GGpgm9RTt9plwq4zK9Vbt2Fz5WNOm7RDq+HUSjv+rmjo6tQuCxcuXJWZgz53pxY9a5m5tnxdFxHfAQ4G7ouIGWWv2gxg3QD7LgYWA8ydOzcXLFjQUgwrVqyg1X0HUvWeDw3sgclzmPLwdd0OQw1sk3rqVrvMf+0bKtVr5tEdx+XF1SouPLPyMbuhHX9XNHQjsV26nqxFxATgaZm5oXz/MuBDwKXAScA55esl3Yt+CFJBAAAHFElEQVRSklQ7y8+uVq/mSZ00mK4na8B04DsRAUU838jMH0TEfwMXRsQpwF3A67oYoyRJUld0PVnLzDXAC/opXw8c3vmIJEmS6qPOj+6QJEka80zWJEmSasxkTZIkqca6fs+aJGn0a+phvHtOrVRt5ZoK02sB8xdWP7VURyZrkiRR/Vlwpx+xT5sjkZ7MZE2S1LKqCc68Jo5ZtcdsuFXv/XPyenWWyZokaVRrR0IpdZLJmiSpZU4ML7WfyZokSW2wbsPmSr163gOnwZistcAJ2iVpDKs6JykHtjUMjR0ma5IkjRROXj8mmaxJktSEyqNVp7U3Do0dJmuSJI0QVRPFn2/xmXGjicmaJEltMGHzb5i3btmg9c5dtqjyMX28yNhksiZJGtXq/niRbsbnrA0jgxO5S5Ik1Zg9a5IkjVF1n2Kras8fjO7eP5M1SZK0Tc0kTVWM5sSqHUzWJEkaZbp1H1zde+pGKpM1SZLUUcPdUzfamaxJkqRtqvuIWqD67A4xv71xtIHJmiRJqqVmksSVVSs+u6VQuspHd0iSJNWYPWuSJKmjunlZdd2GzZXumavTiFWTNUmSNGZUnQasTiNWvQwqSZJUYyZrkiRJNWayJkmSVGMma5IkSTVmsiZJklRjtU/WIuKoiLg1IlZHxBndjkeSJKmTap2sRcQ44NPAy4H9gBMiYr/uRiVJktQ5tU7WgIOB1Zm5JjMfA5YCx3Y5JkmSpI6pe7K2O3B3w3pPWSZJkjQm1H0Gg+inLJ9UIWIRsKhc3RgRt7Z4rl2B+1vcV+1ju9SPbVJPtkv92Cb1VK1d3vLx9kcCz6pSqe7JWg8wq2F9JrC2sUJmLgaGPMlYRFydmXOHehwNL9ulfmyTerJd6sc2qaeR2C51vwz638DeEbFHRGwPHA9c2uWYJEmSOqbWPWuZuSUiTgMuB8YB52XmTV0OS5IkqWNqnawBZOZlwGUdONWQL6WqLWyX+rFN6sl2qR/bpJ5GXLtEZg5eS5IkSV1R93vWJEmSxjSTNZzSqpMi4ryIWBcRNzaUTYmIZRFxe/m6S1keEfHJsl2uj4gDG/Y5qax/e0Sc1I3PMppExKyIWB4Rt0TETRHxtrLctumSiNgxIq6KiF+UbfLBsnyPiLiy/H6/WQ6+IiJ2KNdXl9tnNxzrzLL81og4sjufaPSIiHERcW1EfK9ct026LCLujIgbIuK6iLi6LBs9v78yc0wvFAMX7gD2BLYHfgHs1+24RusCHAocCNzYUPbPwBnl+zOAj5bvjwa+T/G8vXnAlWX5FGBN+bpL+X6Xbn+2kbwAM4ADy/eTgNsopnizbbrXJgFMLN+PB64sv+sLgePL8s8Bf12+/xvgc+X744Fvlu/3K3+v7QDsUf6+G9ftzzeSF+AdwDeA75Xrtkn32+ROYNc+ZaPm95c9a05p1VGZ+RPggT7FxwLnl+/PB17dUP6VLPwc2DkiZgBHAssy84HMfBBYBhzV/uhHr8y8NzOvKd9vAG6hmC3EtumS8rvdWK6OL5cEXgJcVJb3bZPetroIODwioixfmpmbM/OXwGqK33tqQUTMBF4BfLFcD2yTuho1v79M1pzSqg6mZ+a9UCQNwLSyfKC2sc3aqLxUcwBFT45t00Xl5bbrgHUUfzjuAB7KzC1llcbv94nvvtz+MDAV22S4/QvwbuB35fpUbJM6SOCHEbEqipmNYBT9/qr9ozs6YNAprdQ1A7WNbdYmETERuBh4e2b+b9EJ0H/Vfspsm2GWmVuBORGxM/Ad4I/6q1a+2iZtFhGvBNZl5qqIWNBb3E9V26TzXpyZayNiGrAsIv5nG3VHXLvYs1ZhSiu13X1lFzTl67qyfKC2sc3aICLGUyRqX8/Mb5fFtk0NZOZDwAqK+2t2joje/2g3fr9PfPfl9skUtxzYJsPnxcAxEXEnxS0zL6HoabNNuiwz15av6yj+Y3Mwo+j3l8maU1rVwaVA76ibk4BLGspPLEfuzAMeLruyLwdeFhG7lKN7XlaWqUXlfTRLgFsy8xMNm2ybLomIZ5Q9akTE04GXUtxLuBx4bVmtb5v0ttVrgR9lcdf0pcDx5cjEPYC9gas68ylGl8w8MzNnZuZsir8VP8rM12ObdFVETIiISb3vKX7v3Mho+v3V7REOdVgoRobcRnE/yPu6Hc9oXoALgHuBxyn+F3MKxT0cVwC3l69TyroBfLpslxuAuQ3HOZniptzVwJu7/blG+gL8CUV3//XAdeVytG3T1TZ5PnBt2SY3Au8vy/ek+MO+GvgWsENZvmO5vrrcvmfDsd5XttWtwMu7/dlGwwIs4PejQW2T7rbFnhSja38B3NT7d3w0/f5yBgNJkqQa8zKoJElSjZmsSZIk1ZjJmiRJUo2ZrEmSJNWYyZokSVKNmaxJkiTVmMmaJElSjZmsSZIk1dj/B05iLYw4G/dLAAAAAElFTkSuQmCC\n",
            "text/plain": "<Figure size 720x360 with 1 Axes>"
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": "The hypothes is: rev in the variant part is greater than the control part.\nThe p-value is: 0.2909\nThe result is not significant. The hypothesis can not be nagted!\n\nFor tablet\nThe maximun rev of a user : 59820\n98% users with rev below: 5501\nThe minimun rev of a user : 0\n********** Data Visualization for 98% users ******\n"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<Figure size 720x360 with 1 Axes>"
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": "The hypothes is: rev in the variant part is greater than the control part.\nThe p-value is: 0.3559\nThe result is not significant. The hypothesis can not be nagted!\n\nFor desktop\nThe maximun rev of a user : 48808\n98% users with rev below: 5650\nThe minimun rev of a user : 0\n********** Data Visualization for 98% users ******\n"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<Figure size 720x360 with 1 Axes>"
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": "The hypothes is: rev in the variant part is greater than the control part.\nThe p-value is: 0.4081\nThe result is not significant. The hypothesis can not be nagted!\n\n"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "* ## Significance Analysis for Every Order Value "
    },
    {
      "metadata": {
        "code_folding": [
          0
        ],
        "scrolled": false,
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Check the KPI (every order value) distribution for thr test \n# calculate the rev per transaction (every order value)\ndf['Rev./Trans'] = df.apply(lambda x: x['rev']/x['num_trans'] if x['num_trans']!= 0 else 0, axis = 1)\npercentile = 98\nkpi = 'Rev./Trans'\nbins = 100\n# Significant level \nalpha = 0.1\n# Define hypothesis\nHypothesis = \"less\" #\"less\", \"two-sided\", \"greater\"\n# Step 1 get the general one\nprint ('For all devices')\nlower,pp,upper = visionlise_data(df,percentile,test_names,kpi,bins)\n\ndf_c = df[df[kpi] <= pp] # Here you remove the 1 percent data\nMannwhitney_test_for_values(df_c,Hypothesis,alpha,test_names,kpi)\nprint ()\n\n# Step 2 get the overview for different devices\nfor i in range(len(device_list)):\n    df_tmp = df[df['device']==device_list[i]]\n    print ('For ' + device_list[i] )\n    #plt.subplot(1,3,i+1)\n    lower_tmp,pp_tmp,upper_tmp = visionlise_data(df_tmp,percentile,test_names,kpi,bins)\n    df_c_tmp = df_tmp[df_tmp[kpi] <= pp] # Here you remove the 1 percent data\n    Mannwhitney_test_for_values(df_c_tmp,Hypothesis,alpha,test_names,kpi)\n    print ()",
      "execution_count": 13,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": "For all devices\nThe maximun Rev./Trans of a user : 13999\n98% users with Rev./Trans below: 2306\nThe minimun Rev./Trans of a user : 0\n********** Data Visualization for 98% users ******\n"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<Figure size 720x360 with 1 Axes>"
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": "The hypothes is: Rev./Trans in the variant part is less than the control part.\nThe p-value is: 0.7239\nThe result is not significant. The hypothesis can not be nagted!\n\nFor mobile\nThe maximun Rev./Trans of a user : 13999\n98% users with Rev./Trans below: 2180\nThe minimun Rev./Trans of a user : 0\n********** Data Visualization for 98% users ******\n"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<Figure size 720x360 with 1 Axes>"
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": "The hypothes is: Rev./Trans in the variant part is less than the control part.\nThe p-value is: 0.6773\nThe result is not significant. The hypothesis can not be nagted!\n\nFor tablet\nThe maximun Rev./Trans of a user : 6321\n98% users with Rev./Trans below: 2410\nThe minimun Rev./Trans of a user : 0\n********** Data Visualization for 98% users ******\n"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<Figure size 720x360 with 1 Axes>"
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": "The hypothes is: Rev./Trans in the variant part is less than the control part.\nThe p-value is: 0.7277\nThe result is not significant. The hypothesis can not be nagted!\n\nFor desktop\nThe maximun Rev./Trans of a user : 12348\n98% users with Rev./Trans below: 2367\nThe minimun Rev./Trans of a user : 0\n********** Data Visualization for 98% users ******\n"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<Figure size 720x360 with 1 Axes>"
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": "The hypothes is: Rev./Trans in the variant part is less than the control part.\nThe p-value is: 0.5819\nThe result is not significant. The hypothesis can not be nagted!\n\n"
        }
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3",
      "language": "python"
    },
    "language_info": {
      "name": "python",
      "version": "3.6.4",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "latex_envs": {
      "eqNumInitial": 1,
      "eqLabelWithNumbers": true,
      "current_citInitial": 1,
      "cite_by": "apalike",
      "bibliofile": "biblio.bib",
      "LaTeX_envs_menu_present": true,
      "labels_anchors": false,
      "latex_user_defs": false,
      "user_envs_cfg": false,
      "report_style_numbering": false,
      "autoclose": false,
      "autocomplete": true,
      "hotkeys": {
        "equation": "Ctrl-E",
        "itemize": "Ctrl-I"
      }
    },
    "gist": {
      "id": "b9796cb531aa43564ebba5386b121a17",
      "data": {
        "description": "Get AB test data and do a simple AB test Analysis on the result",
        "public": true
      }
    },
    "_draft": {
      "nbviewer_url": "https://gist.github.com/b9796cb531aa43564ebba5386b121a17"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 2
}