Quora_DataChallenge.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "benhw4HSH16e"
   },
   "source": [
    "## Improving User Engagement on Quora Mobile App\n",
    "\n",
    "Suppose you are a Data Scientist on the Mobile team at Quora. The team has just introduced a new UI design to the Quora app. The goal of the new design is to increase user engagement (measured by minutes spent on site). The team ran an A/B test to evaluate the change. Using the data, help the team understand the impact of the UI change better.\n",
    "\n",
    "Tables provided are as follows:\n",
    "    1. t1_user_active_min.csv\n",
    "    2. t2_user_variant.csv\n",
    "    3. t3_user_active_min_pre.csv\n",
    "    4. t4_user_attributes.csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "SL3BEOnoH16g"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from datetime import datetime as dt\n",
    "from datetime import timedelta\n",
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "InteractiveShell.ast_node_interactivity = \"all\"\n",
    "%matplotlib inline\n",
    "pd.set_option('float_format', '{:f}'.format)\n",
    "\n",
    "from scipy import stats\n",
    "from scipy.stats import ttest_ind_from_stats\n",
    "from statsmodels.stats.power import TTestIndPower"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "8-MjGb9VH16p"
   },
   "outputs": [],
   "source": [
    "t1_usr_active_min = pd.read_csv(\"t1_user_active_min.csv\")\n",
    "t2_usr_variant = pd.read_csv(\"t2_user_variant.csv\")\n",
    "t3_usr_active_min_pre = pd.read_csv(\"t3_user_active_min_pre.csv\")\n",
    "t4_usr_attributes = pd.read_csv(\"t4_user_attributes.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 467
    },
    "id": "IGJqDheNH16z",
    "outputId": "23f63225-e3ef-442f-93e8-d68bfc2e0475"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "t1_user_active_min\n"
     ]
    },
    {
     "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>uid</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-22</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>2019-03-11</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uid          dt  active_mins\n",
       "0    0  2019-02-22     5.000000\n",
       "1    0  2019-03-11     5.000000"
      ]
     },
     "execution_count": 10,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "t2_user_variant\n"
     ]
    },
    {
     "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>dt</th>\n",
       "      <th>signup_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-11-07</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uid  variant_number          dt signup_date\n",
       "0    0               0  2019-02-06  2018-09-24\n",
       "1    1               0  2019-02-06  2016-11-07"
      ]
     },
     "execution_count": 10,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "t3_user_active_min_pre\n"
     ]
    },
    {
     "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>uid</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>2018-11-08</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uid          dt  active_mins\n",
       "0    0  2018-09-24     3.000000\n",
       "1    0  2018-11-08     4.000000"
      ]
     },
     "execution_count": 10,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "t4_user_attributes\n"
     ]
    },
    {
     "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>uid</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>reader</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uid gender   user_type\n",
       "0    0   male  non_reader\n",
       "1    1   male      reader"
      ]
     },
     "execution_count": 10,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"t1_user_active_min\")\n",
    "t1_usr_active_min.head(2)\n",
    "print(\"t2_user_variant\")\n",
    "t2_usr_variant.head(2)\n",
    "print(\"t3_user_active_min_pre\")\n",
    "t3_usr_active_min_pre.head(2)\n",
    "print(\"t4_user_attributes\")\n",
    "t4_usr_attributes.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "XyB9gGIqH166",
    "outputId": "22811a79-8101-4e8e-81d7-a1090a86e1e6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pre Experiments date range: 2018-08-10  -  2019-02-06\n",
      "Duration of Experiment: 2019-02-06  -  2019-07-05\n",
      "Experiment Duration(Days): 150\n"
     ]
    }
   ],
   "source": [
    "print(\"Pre Experiments date range:\", t3_usr_active_min_pre.dt.min(),\" - \",t2_usr_variant.dt.max())\n",
    "print(\"Duration of Experiment:\",t1_usr_active_min.dt.min(),\" - \", t1_usr_active_min.dt.max())\n",
    "\n",
    "delta = dt.strptime(t1_usr_active_min.dt.max(),\"%Y-%m-%d\")-dt.strptime(t1_usr_active_min.dt.min(),\"%Y-%m-%d\")\n",
    "print(\"Experiment Duration(Days):\",delta.days+1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "n4dMiSyAH17A"
   },
   "source": [
    "### Exploring User Distribtion for control & treatment group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "_ovmUwDPH17C",
    "outputId": "4aff20e5-aa36-4401-c8b1-88b76f918e56"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    40000\n",
       "1    10000\n",
       "Name: variant_number, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train - Test ratio: 80% - 20%\n"
     ]
    }
   ],
   "source": [
    "t2_usr_variant['variant_number'].value_counts()\n",
    "print(\"Train - Test ratio: 80% - 20%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "bAO0I8PIH17I",
    "outputId": "4cfcd5a8-76ba-4c04-cecc-5420aa4c847e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "50000"
      ]
     },
     "execution_count": 13,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "46633"
      ]
     },
     "execution_count": 13,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "49697"
      ]
     },
     "execution_count": 13,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Checking no. of unique users\n",
    "t2_usr_variant.uid.nunique()\n",
    "t1_usr_active_min.uid.nunique()\n",
    "t3_usr_active_min_pre.uid.nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "g5rRu8HxH17N"
   },
   "source": [
    "### Preparing a master table with user\n",
    "\n",
    "Preparing now a master table with these following important user informations: User Daily Activity (uid, active_min), \n",
    "User Signup, User Variant Number, User characteristics (gender, user_types). Motive to prepare this dataset is to have table with complete user details and use this for hypothesis testing and further insights.\n",
    "\n",
    "<b>I have specifically done  a left join of t2_usr_varient table with t1_usr_active_min so as to preserve the activity (or non-activity) of all the 50000 users who are are part of the control & treatment group.</b>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 129
    },
    "id": "g9owIMgfH17N",
    "outputId": "476be48a-2365-4f00-a5d5-0b96ae128284"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1069769, 8)"
      ]
     },
     "execution_count": 14,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "50000"
      ]
     },
     "execution_count": 14,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>dt_x</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>dt_y</th>\n",
       "      <th>active_mins</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-02-22</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-03-11</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uid  variant_number        dt_x  ... active_mins gender   user_type\n",
       "0    0               0  2019-02-06  ...    5.000000   male  non_reader\n",
       "1    0               0  2019-02-06  ...    5.000000   male  non_reader\n",
       "\n",
       "[2 rows x 8 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "usr_postexp_df = pd.merge(t2_usr_variant,t1_usr_active_min, on='uid', how='left')\n",
    "usr_postexp_df = pd.merge(usr_postexp_df,t4_usr_attributes, on='uid', how='left')\n",
    "usr_postexp_df.shape\n",
    "usr_postexp_df.uid.nunique()\n",
    "usr_postexp_df.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "M4eN8DpKH17S"
   },
   "outputs": [],
   "source": [
    "usr_postexp_df.rename(columns={'dt_x':'experiement_dt','dt_y':'dt'}, inplace=True )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "H-Zdi_KPH17X"
   },
   "source": [
    "### We know that the expriment was live for 150 days, so I am checking in the following codes if the user were active daily or not. My motive is to have an apple to apple comparison for same no of days, the mean active mins for control and treatment group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 458
    },
    "id": "0zh2I0OEH17Y",
    "outputId": "24fd874c-1081-4eb4-8552-5fed0414e9d8"
   },
   "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>tot_days</th>\n",
       "      <th>tot_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>43.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>95</td>\n",
       "      <td>15205.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>17.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>24</td>\n",
       "      <td>77.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>20</td>\n",
       "      <td>39.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uid  variant_number  tot_days     tot_mins\n",
       "0    0               0        13    43.000000\n",
       "1    1               0        95 15205.000000\n",
       "2    2               0         7    17.000000\n",
       "3    3               0        24    77.000000\n",
       "4    4               0        20    39.000000"
      ]
     },
     "execution_count": 16,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "array([ 13,  95,   7,  24,  20,  14,   0,   9,  31,   6,   2,  12,  18,\n",
       "        89,  34,  10,  35,  33,  37,  81,  56,  17,  16,  21,  50,  30,\n",
       "         5,  25,   1,  15,   8,  62,  11,  54,   3,   4,  55,  59,  27,\n",
       "        40,  19,  58,  66,  64,  71,  29,  57,  41,  23, 103,  60,  80,\n",
       "        76,  38,  32,  84,  92,  39,  70,  74,  51,  45,  77,  22,  63,\n",
       "        75,  28,  72,  43,  53,  67,  26,  93,  88,  42,  36,  68,  91,\n",
       "        83,  49,  48,  47,  86,  97,  99,  65,  69,  82,  44, 102,  61,\n",
       "        98, 101,  46, 100,  85,  94,  78,  96,  52,  87,  79,  90,  73,\n",
       "       106, 104, 107, 105, 108, 110, 109])"
      ]
     },
     "execution_count": 16,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "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>variant_number</th>\n",
       "      <th>tot_days</th>\n",
       "      <th>tot_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>22.173925</td>\n",
       "      <td>783.719625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>17.944500</td>\n",
       "      <td>722.094000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   variant_number  tot_days   tot_mins\n",
       "0               0 22.173925 783.719625\n",
       "1               1 17.944500 722.094000"
      ]
     },
     "execution_count": 16,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = usr_postexp_df.groupby(['uid','variant_number'], as_index=False).agg({'dt':'count','active_mins': 'sum'})\n",
    "temp.rename(columns={'active_mins': 'tot_mins','dt': 'tot_days'},inplace=True)\n",
    "temp.head()\n",
    "temp.tot_days.unique()\n",
    "temp.groupby(['variant_number'], as_index=False).agg({'tot_days':'mean','tot_mins':'mean'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "wKmVp66eH17b"
   },
   "source": [
    "### Thus, from the above two aggregates function we see that no user (either 0/1 variant) was present for all single day of the experiment. Also the mean days control & treatment group users were active is 22 and 18 days respectively. Thus we now need to populate the above table for each day of the experiement and compare the means of the two groups\n",
    "\n",
    "### Also from the last table, we see that there is a difference between the average tot_mins spend by users in control and treatment groups, but we need to be check if there are many outliers (like15202.0) misrepresenting this number.\n",
    "\n",
    "## Outlier Treatment\n",
    "\n",
    "From the `usr_postexp_df` df, only column 'active_mins' has outliers in it. Since a day has maximum 1440 mins, therefore I am removing all the entries with values greater than 1500 mins (taken a buffer of 60 mins more)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 318
    },
    "id": "v4I__vxfH17d",
    "outputId": "18b09d8a-116c-47d5-f942-97f294b23160"
   },
   "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>active_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1069769.000000</td>\n",
       "      <td>1069769.000000</td>\n",
       "      <td>1066402.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>24226.815362</td>\n",
       "      <td>0.168482</td>\n",
       "      <td>36.168091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>14175.415856</td>\n",
       "      <td>0.374294</td>\n",
       "      <td>1270.483515</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>12023.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>23945.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>36186.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>17.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>49999.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>99999.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 uid  variant_number    active_mins\n",
       "count 1069769.000000  1069769.000000 1066402.000000\n",
       "mean    24226.815362        0.168482      36.168091\n",
       "std     14175.415856        0.374294    1270.483515\n",
       "min         0.000000        0.000000       1.000000\n",
       "25%     12023.000000        0.000000       2.000000\n",
       "50%     23945.000000        0.000000       5.000000\n",
       "75%     36186.000000        0.000000      17.000000\n",
       "max     49999.000000        1.000000   99999.000000"
      ]
     },
     "execution_count": 17,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "(172, 8)"
      ]
     },
     "execution_count": 17,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "(172, 8)"
      ]
     },
     "execution_count": 17,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "usr_postexp_df.describe()\n",
    "#plt.hist(usr_postexp_df.active_mins)\n",
    "usr_postexp_df[usr_postexp_df.active_mins>=1500].shape\n",
    "usr_postexp_df[usr_postexp_df.active_mins==99999].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "sGHuXTSCH17i",
    "outputId": "4dfb7b11-8e99-4c1a-d99e-27c7203b7277"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1069769, 8)"
      ]
     },
     "execution_count": 18,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "172"
      ]
     },
     "execution_count": 18,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "usr_postexp_df.shape\n",
    "temp = usr_postexp_df['active_mins'] >=1500\n",
    "usr_postexp_df2 = usr_postexp_df[~temp]\n",
    "usr_postexp_df.shape[0]-usr_postexp_df2.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "JGFLMyjhH17l"
   },
   "source": [
    "Thus we have removed the outlier values: 172 entries with active_mins >=1500 mins"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 363
    },
    "id": "96vgcmHNH17m",
    "outputId": "f9e6b457-db3a-4f10-a0a4-38467f5338c5"
   },
   "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>experiement_dt</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-02-22</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-03-11</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-03-18</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-03-22</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-04-03</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uid  variant_number experiement_dt  ... active_mins gender   user_type\n",
       "0    0               0     2019-02-06  ...    5.000000   male  non_reader\n",
       "1    0               0     2019-02-06  ...    5.000000   male  non_reader\n",
       "2    0               0     2019-02-06  ...    3.000000   male  non_reader\n",
       "3    0               0     2019-02-06  ...    4.000000   male  non_reader\n",
       "4    0               0     2019-02-06  ...    9.000000   male  non_reader\n",
       "\n",
       "[5 rows x 8 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "uid                  0\n",
       "variant_number       0\n",
       "experiement_dt       0\n",
       "signup_date          0\n",
       "dt                3367\n",
       "active_mins       3367\n",
       "gender               0\n",
       "user_type            0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "usr_postexp_df2.head()\n",
    "usr_postexp_df2.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "QArhIaduH17t"
   },
   "source": [
    "From the above NA value table, we see that there are around 3367 users for whom there is no activity even for a single day."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 336
    },
    "id": "BXGNlVizH17t",
    "outputId": "631e4de1-1004-481b-fda6-0513463c9967"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3367, 8)"
      ]
     },
     "execution_count": 20,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>experiement_dt</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>193</th>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2019-01-18</td>\n",
       "      <td>NaN</td>\n",
       "      <td>nan</td>\n",
       "      <td>male</td>\n",
       "      <td>new_user</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>258</th>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2017-04-20</td>\n",
       "      <td>NaN</td>\n",
       "      <td>nan</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>797</th>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2019-01-25</td>\n",
       "      <td>NaN</td>\n",
       "      <td>nan</td>\n",
       "      <td>male</td>\n",
       "      <td>new_user</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1612</th>\n",
       "      <td>78</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2014-08-11</td>\n",
       "      <td>NaN</td>\n",
       "      <td>nan</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1905</th>\n",
       "      <td>94</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2019-02-03</td>\n",
       "      <td>NaN</td>\n",
       "      <td>nan</td>\n",
       "      <td>female</td>\n",
       "      <td>new_user</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      uid  variant_number experiement_dt  ... active_mins  gender   user_type\n",
       "193     8               0     2019-02-06  ...         nan    male    new_user\n",
       "258    12               0     2019-02-06  ...         nan    male  non_reader\n",
       "797    35               0     2019-02-06  ...         nan    male    new_user\n",
       "1612   78               0     2019-02-06  ...         nan    male  non_reader\n",
       "1905   94               0     2019-02-06  ...         nan  female    new_user\n",
       "\n",
       "[5 rows x 8 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "3367"
      ]
     },
     "execution_count": 20,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "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>variant_number</th>\n",
       "      <th>uid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2575</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>792</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   variant_number   uid\n",
       "0               0  2575\n",
       "1               1   792"
      ]
     },
     "execution_count": 20,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = usr_postexp_df2[usr_postexp_df2.active_mins.isna()]\n",
    "temp.shape\n",
    "temp.head()\n",
    "temp.uid.nunique()\n",
    "temp.groupby(['variant_number'],as_index=False).agg({'uid':'nunique'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "v8XtujfOH17y"
   },
   "source": [
    "### Thus we see that a lot of users (3k+) from control group and also treatment groups have shown no activity during the experiment period. \n",
    "### Currently, I am removing these users since for these users the active_mins will be 0 for all days and since they were not active even for single day, so they didn't see the newly designed UI and hence have no role to play."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 224
    },
    "id": "zyiQNtoeH17z",
    "outputId": "4f70c4a9-b27f-4e15-fd84-c4bb29d207df"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1066230, 8)"
      ]
     },
     "execution_count": 21,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>experiement_dt</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-02-22</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-03-11</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-03-18</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-03-22</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2019-04-03</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uid  variant_number experiement_dt  ... active_mins gender   user_type\n",
       "0    0               0     2019-02-06  ...    5.000000   male  non_reader\n",
       "1    0               0     2019-02-06  ...    5.000000   male  non_reader\n",
       "2    0               0     2019-02-06  ...    3.000000   male  non_reader\n",
       "3    0               0     2019-02-06  ...    4.000000   male  non_reader\n",
       "4    0               0     2019-02-06  ...    9.000000   male  non_reader\n",
       "\n",
       "[5 rows x 8 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "usr_postexp_dfnew = usr_postexp_df2[~usr_postexp_df2.active_mins.isna()]\n",
    "usr_postexp_dfnew.shape\n",
    "usr_postexp_dfnew.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7ec4IsDSH173"
   },
   "source": [
    "Let's look at the variant number 0/1 wise the mean active_mins and it's std deviation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 143
    },
    "id": "0Y-EEgtbH175",
    "outputId": "6eda902d-ec8c-43ad-9163-584885fbad95"
   },
   "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>mean</th>\n",
       "      <th>median</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>variant_number</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>19.337660</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>44.797631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>23.526294</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>54.191356</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    mean   median       std\n",
       "variant_number                             \n",
       "0              19.337660 5.000000 44.797631\n",
       "1              23.526294 7.000000 54.191356"
      ]
     },
     "execution_count": 22,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "usr_postexp_dfnew.groupby(['variant_number'],as_index=False).active_mins.agg({'mean','std','median'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "yIrXODWVH179"
   },
   "source": [
    "#### Populating the above df for each user * 150 days of experiment. This will give us an apple-to-apply comparison, ie the daily acitve _min for each user for every single day when the experiment was live\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "pz_lv7YXH17-",
    "outputId": "9a64f59d-cccc-4f73-8497-ce2c25e59a36"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Before a& After reshaping the data at daily level: (1066230, 8) (6994950, 8)\n"
     ]
    }
   ],
   "source": [
    "daily_usr_act = usr_postexp_dfnew.pivot_table(index=['variant_number','uid','experiement_dt','signup_date',\n",
    "                                                      'gender','user_type'],columns = 'dt',values='active_mins',fill_value=0).reset_index()\n",
    "daily_usr_act = daily_usr_act.melt(id_vars=['variant_number','uid','experiement_dt','signup_date',\n",
    "                                                      'gender','user_type'], var_name='dt',value_name='active_mins')\n",
    "print(\"Before a& After reshaping the data at daily level:\",usr_postexp_dfnew.shape, daily_usr_act.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "G-Fq-m8PH18D"
   },
   "source": [
    "### Let's now see the impact of new UI on average active_mins for control and treatment group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 143
    },
    "id": "pomz6dlIH18E",
    "outputId": "34b3488a-5518-4ea9-c955-4478b2dae2f0"
   },
   "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>median</th>\n",
       "      <th>var</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>variant_number</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>366.763169</td>\n",
       "      <td>3.054808</td>\n",
       "      <td>19.151062</td>\n",
       "      <td>5613750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>444.027311</td>\n",
       "      <td>3.056017</td>\n",
       "      <td>21.071956</td>\n",
       "      <td>1381200</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                median        var     mean       std    count\n",
       "variant_number                                               \n",
       "0                    0 366.763169 3.054808 19.151062  5613750\n",
       "1                    0 444.027311 3.056017 21.071956  1381200"
      ]
     },
     "execution_count": 24,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_usr_act.groupby(['variant_number'],as_index=False).active_mins.agg({'mean','std','median','var','count'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "UQoqw4gZH18P"
   },
   "outputs": [],
   "source": [
    "mean_0 = daily_usr_act[daily_usr_act['variant_number']==0].active_mins.mean()\n",
    "mean_1 = daily_usr_act[daily_usr_act['variant_number']==1].active_mins.mean()\n",
    "\n",
    "std_0 = daily_usr_act[daily_usr_act['variant_number']==0].active_mins.std()\n",
    "std_1 = daily_usr_act[daily_usr_act['variant_number']==1].active_mins.std()\n",
    "\n",
    "nobs0 = daily_usr_act[daily_usr_act['variant_number']==0].active_mins.count()\n",
    "nobs1 = daily_usr_act[daily_usr_act['variant_number']==1].active_mins.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "CrH3PTfEH18T"
   },
   "source": [
    "### Performing t-test to understand if there was any impact of new UI on user's dUIaily engagement\n",
    "\n",
    "<b> <u>Null Hypothesis Ho:</u> New UI had no effect on daily user enagegment i.e. there is no difference in control and variant users' average active mins\n",
    "<u>Alternate Hypothesis:</u> New UI had an effect on daily user engagment, .e. there is a difference in control and variant users' average active mins</b>\n",
    "\n",
    "Here using the Welch t-test because both variane and sample sizes of two groups are different."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "AYnRrT6KH18V",
    "outputId": "a9b88292-1913-40cc-d8a3-6dbddb31edf9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ttest_indResult(statistic=0.06145955015857426, pvalue=0.9509932339796476)"
      ]
     },
     "execution_count": 26,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ttest_ind_from_stats(mean1=mean_1,std1=std_1,nobs1=nobs1,\n",
    "                     mean2=mean_0,std2=std_0,nobs2=nobs0, equal_var=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "IDihHr6VH18Y"
   },
   "source": [
    "<b> The p-value from the above t-test is very high and thus we fail to reject the null-hypothesis. </b>\n",
    "#### Preparing Confidence Interval"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "DPhvXJv4H18Y"
   },
   "outputs": [],
   "source": [
    "def calc_CI(mean_1,std_1,nobs1,mean_0,std_0,nobs0,CI_interval):\n",
    "    se = np.sqrt((std_1 * std_1) / nobs1 + (std_0 * std_0) / nobs0)\n",
    "    mean_diff = mean_1-mean_0\n",
    "    qt = stats.norm.isf((1-CI_interval)/2)\n",
    "    lower = mean_diff - qt*se\n",
    "    upper = mean_diff + qt*se\n",
    "    return (lower,upper)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ZnFU8C2eH18d",
    "outputId": "4e6972fc-db5a-447f-b1c4-db2a184aa7ee"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-0.03733892403590165, 0.039756441143676366)"
      ]
     },
     "execution_count": 28,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CI = calc_CI(mean_1,std_1,nobs1,mean_0,std_0,nobs0,0.95)\n",
    "CI"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0hcuAhj3H18h"
   },
   "source": [
    "## Pre and Post Experiment Data Analysis\n",
    "\n",
    "You decide to dive deeper into the data, so you gather a table of active minutes by user from before the experiment began. You should now use table 3 (t3_user_active_min_pre.csv) along with tables 1 and 2 for this question.\n",
    "\n",
    "Using the statistical method of your choice and the pre-experiment data, update your 95% confidence interval of the overall average treatment effect.\n",
    "\n",
    "Please enter your confidence interval below, after rounding your answer to second place after the decimal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "dmNff5NeH18i",
    "outputId": "aeb8d655-32de-408e-815c-7a6edb7d88e0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pre Experiments date range: 2018-08-10  -  2019-02-06\n",
      "Duration of Experiment: 2019-02-06  -  2019-07-05\n",
      "Experiment Duration(Days): 150\n",
      "Pre Experiment Duration(Days): 180\n"
     ]
    }
   ],
   "source": [
    "print(\"Pre Experiments date range:\", t3_usr_active_min_pre.dt.min(),\" - \",t2_usr_variant.dt.max())\n",
    "print(\"Duration of Experiment:\",t1_usr_active_min.dt.min(),\" - \", t1_usr_active_min.dt.max())\n",
    "\n",
    "delta = dt.strptime(t1_usr_active_min.dt.max(),\"%Y-%m-%d\")-dt.strptime(t1_usr_active_min.dt.min(),\"%Y-%m-%d\")\n",
    "predelta = dt.strptime(t3_usr_active_min_pre.dt.max(),\"%Y-%m-%d\")-dt.strptime(t3_usr_active_min_pre.dt.min(),\"%Y-%m-%d\") \n",
    "print(\"Experiment Duration(Days):\",delta.days+1)\n",
    "print(\"Pre Experiment Duration(Days):\",predelta.days+1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fTXyB9A1H18m"
   },
   "source": [
    "#### In the following lines of code, I reiterating steps & function used above to prepare data ready for t-test and CI calculation.\n",
    "#### Steps taken:\n",
    "\n",
    "- Preparing Master Table: pre-experiement 0/1 variant group's average activity mins\n",
    "- Outlier removal\n",
    "- Performing T-test for pre and post experiment data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 161
    },
    "id": "fkcI-d-IH18m",
    "outputId": "db0ac17d-73f9-4c7a-f986-64881be3b7d4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1190396, 8)"
      ]
     },
     "execution_count": 30,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "50000"
      ]
     },
     "execution_count": 30,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>dt_x</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>dt_y</th>\n",
       "      <th>active_mins</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2018-11-08</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2018-11-24</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uid  variant_number        dt_x  ... active_mins gender   user_type\n",
       "0    0               0  2019-02-06  ...    3.000000   male  non_reader\n",
       "1    0               0  2019-02-06  ...    4.000000   male  non_reader\n",
       "2    0               0  2019-02-06  ...    3.000000   male  non_reader\n",
       "\n",
       "[3 rows x 8 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "usr_preexp_df = pd.merge(t2_usr_variant,t3_usr_active_min_pre,on=\"uid\",how=\"left\")\n",
    "preexp_usr_df = pd.merge(usr_preexp_df,t4_usr_attributes,on=\"uid\",how=\"left\")\n",
    "preexp_usr_df.shape\n",
    "preexp_usr_df.uid.nunique()\n",
    "preexp_usr_df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "rjdi-qh7H18q"
   },
   "outputs": [],
   "source": [
    "preexp_usr_df.rename(columns={'dt_x':'experiement_dt','dt_y':'dt'}, inplace=True )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "FeD3OQfqH18t"
   },
   "source": [
    "### Outlier Treatment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 318
    },
    "id": "hbLNePYxH18t",
    "outputId": "9101fe05-8b4e-43b3-b70a-cf7aae6f9280"
   },
   "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>active_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1190396.000000</td>\n",
       "      <td>1190396.000000</td>\n",
       "      <td>1190093.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>24206.442924</td>\n",
       "      <td>0.168585</td>\n",
       "      <td>32.203151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>14180.354303</td>\n",
       "      <td>0.374385</td>\n",
       "      <td>1181.530546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>11975.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>23933.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>36197.250000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>49999.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>99999.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 uid  variant_number    active_mins\n",
       "count 1190396.000000  1190396.000000 1190093.000000\n",
       "mean    24206.442924        0.168585      32.203151\n",
       "std     14180.354303        0.374385    1181.530546\n",
       "min         0.000000        0.000000       1.000000\n",
       "25%     11975.000000        0.000000       2.000000\n",
       "50%     23933.000000        0.000000       4.000000\n",
       "75%     36197.250000        0.000000      14.000000\n",
       "max     49999.000000        1.000000   99999.000000"
      ]
     },
     "execution_count": 32,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "(166, 8)"
      ]
     },
     "execution_count": 32,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preexp_usr_df.describe()\n",
    "#plt.hist(usr_postexp_df.active_mins)\n",
    "preexp_usr_df[preexp_usr_df.active_mins>=1500].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "fkHCryFUH18y",
    "outputId": "92367e9e-618e-4e3b-871f-91b24e2a4c7a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1190396, 8)"
      ]
     },
     "execution_count": 33,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "166"
      ]
     },
     "execution_count": 33,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preexp_usr_df.shape\n",
    "temp = preexp_usr_df['active_mins'] >=1500\n",
    "preexp_usr_df2 = preexp_usr_df[~temp]\n",
    "preexp_usr_df.shape[0]-preexp_usr_df2.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "wYWzn1CfH182"
   },
   "source": [
    "Thus we have removed the outlier values: 166 entries with active_mins >=1500 mins"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 363
    },
    "id": "lYNPvaX5H183",
    "outputId": "ecdfbcb1-286c-4d1c-8c33-6eb96e527f4c"
   },
   "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>experiement_dt</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2018-11-08</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2018-11-24</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2018-11-28</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-09-24</td>\n",
       "      <td>2018-12-02</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uid  variant_number experiement_dt  ... active_mins gender   user_type\n",
       "0    0               0     2019-02-06  ...    3.000000   male  non_reader\n",
       "1    0               0     2019-02-06  ...    4.000000   male  non_reader\n",
       "2    0               0     2019-02-06  ...    3.000000   male  non_reader\n",
       "3    0               0     2019-02-06  ...    6.000000   male  non_reader\n",
       "4    0               0     2019-02-06  ...    6.000000   male  non_reader\n",
       "\n",
       "[5 rows x 8 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "uid                 0\n",
       "variant_number      0\n",
       "experiement_dt      0\n",
       "signup_date         0\n",
       "dt                303\n",
       "active_mins       303\n",
       "gender              0\n",
       "user_type           0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preexp_usr_df2.head()\n",
    "preexp_usr_df2.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zM33s1hbH186"
   },
   "source": [
    "### Here we see that around 303 users have no activity during the pre-experiment period. It would be interesting to see if they had any activity during the experiment days and if so, did they have signup enough days before the launch of experiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 244
    },
    "id": "vZOC08vPH187",
    "outputId": "de870c46-3bc3-4edc-b99f-ae0ec6fa220c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(642, 15)"
      ]
     },
     "execution_count": 35,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "155"
      ]
     },
     "execution_count": 35,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "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>uid</th>\n",
       "      <th>variant_number_x</th>\n",
       "      <th>experiement_dt_x</th>\n",
       "      <th>signup_date_x</th>\n",
       "      <th>dt_x</th>\n",
       "      <th>active_mins_x</th>\n",
       "      <th>gender_x</th>\n",
       "      <th>user_type_x</th>\n",
       "      <th>variant_number_y</th>\n",
       "      <th>experiement_dt_y</th>\n",
       "      <th>signup_date_y</th>\n",
       "      <th>dt_y</th>\n",
       "      <th>active_mins_y</th>\n",
       "      <th>gender_y</th>\n",
       "      <th>user_type_y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>148</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-09-16</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-09-16</td>\n",
       "      <td>NaN</td>\n",
       "      <td>nan</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>148</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-09-16</td>\n",
       "      <td>2019-03-09</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-09-16</td>\n",
       "      <td>NaN</td>\n",
       "      <td>nan</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>148</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-09-16</td>\n",
       "      <td>2019-03-24</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-09-16</td>\n",
       "      <td>NaN</td>\n",
       "      <td>nan</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>148</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-09-16</td>\n",
       "      <td>2019-05-19</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-09-16</td>\n",
       "      <td>NaN</td>\n",
       "      <td>nan</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>148</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-09-16</td>\n",
       "      <td>2019-05-24</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-09-16</td>\n",
       "      <td>NaN</td>\n",
       "      <td>nan</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uid  variant_number_x experiement_dt_x  ... active_mins_y gender_y  user_type_y\n",
       "0  148                 0       2019-02-06  ...           nan   female   non_reader\n",
       "1  148                 0       2019-02-06  ...           nan   female   non_reader\n",
       "2  148                 0       2019-02-06  ...           nan   female   non_reader\n",
       "3  148                 0       2019-02-06  ...           nan   female   non_reader\n",
       "4  148                 0       2019-02-06  ...           nan   female   non_reader\n",
       "\n",
       "[5 rows x 15 columns]"
      ]
     },
     "execution_count": 35,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = preexp_usr_df2[preexp_usr_df2.dt.isna()]\n",
    "check_ifactivity = pd.merge(usr_postexp_dfnew,temp, on = \"uid\",how=\"inner\")\n",
    "check_ifactivity.shape\n",
    "check_ifactivity.uid.nunique()\n",
    "check_ifactivity.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fY00aOo6H189"
   },
   "source": [
    "### Out of the 303 non-active users duing pre-experiment phase, 155 had activity during the experiment. Let's now check if they had signed up enough days before the experiment or during the pre-experiment phase."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "id": "iRwD_39aH18-",
    "outputId": "077ab612-56b7-47cd-98f8-c6a21fe90ea8"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "string"
      },
      "text/plain": [
       "'2013-05-16'"
      ]
     },
     "execution_count": 36,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "string"
      },
      "text/plain": [
       "'2019-02-04'"
      ]
     },
     "execution_count": 36,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Duration of Experiment: 2019-02-06 - 2019-07-05\n"
     ]
    }
   ],
   "source": [
    "check_ifactivity.signup_date_x.min()\n",
    "check_ifactivity.signup_date_x.max()\n",
    "print(\"Duration of Experiment: 2019-02-06 - 2019-07-05\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bSrsHvjrH19A"
   },
   "source": [
    "### Thus we see that pre-experiment phase contains user have signed up 2 days before the experiment began. Therefore for a fair comparison, we should only keep those users who had signedup 150 days before the experiment began. This will give us a fair monitoring period: 150 days before the experiment and 150 days after the experiment.\n",
    "\n",
    "### Experiment Begin Date: 2019-02-06\n",
    "### 150 days before Experiment Start Date: 2018-09-09\n",
    "\n",
    "<b> Thus we will now eliminate all the users who have signup after the above date. Only for the filtered set of users, we will now perform pre-post expeirment</b>\n",
    "\n",
    "<b> Since the experiment is for 150 days starting from 2019-02-06 whereas we have pre-experiment data for 180 days, for equal comparison we will only consider the last 150 days activity data for pre-experiment phase</b>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "3CXTjH78H19B",
    "outputId": "c6c378ab-fffb-43ab-9298-9e34d1f772e1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Removing users signed up after 2018-09-09 from preexp df:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "50000"
      ]
     },
     "execution_count": 37,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100947\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "36337"
      ]
     },
     "execution_count": 37,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Removing users signed up after 2018-09-09 from postexp df:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "46633"
      ]
     },
     "execution_count": 37,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "152041\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "35296"
      ]
     },
     "execution_count": 37,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Removing user activity older than 150 days before the start of experiment\n"
     ]
    }
   ],
   "source": [
    "print(\"Removing users signed up after 2018-09-09 from preexp df:\")\n",
    "bef = preexp_usr_df2.shape[0]\n",
    "preexp_usr_df2.uid.nunique()\n",
    "preexp_usr_df2 = preexp_usr_df2[preexp_usr_df2.signup_date<='2018-09-09']\n",
    "aft = preexp_usr_df2.shape[0]\n",
    "print(bef-aft)\n",
    "preexp_usr_df2.uid.nunique()\n",
    "\n",
    "\n",
    "print(\"Removing users signed up after 2018-09-09 from postexp df:\")\n",
    "bef = usr_postexp_dfnew.shape[0]\n",
    "usr_postexp_dfnew.uid.nunique()\n",
    "usr_postexp_final = usr_postexp_dfnew[usr_postexp_dfnew.signup_date<='2018-09-09']\n",
    "aft = usr_postexp_final.shape[0]\n",
    "print(bef-aft)\n",
    "usr_postexp_final.uid.nunique()\n",
    "\n",
    "print(\"Removing user activity older than 150 days before the start of experiment\")\n",
    "preexp_usr_final = preexp_usr_df2[preexp_usr_df2.dt>='2018-09-09']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "XhSxeIp3H19D"
   },
   "source": [
    "### Thus, we see that we have eliminated around 13k+ users from pre-experiment dataframe who had signuped during our monitoring period. We have also done the same for our post experiment (or during experiment) dataframe. From the postexp df, we have removed around 11k around users.\n",
    "### This should now help us in comparing the pre & post experiment activity for our users better\n",
    "\n",
    "### Now, I am trying to have only those users in my t-test ready data who were present during both the pre-experiment and during experiment phase. \n",
    "\n",
    "I had to debate a lot with myself about whether to include a)those user who were present in both pre & post experiment period or b)all those users who were present in either duration. After much thinking, I believe in first stage of analysis, I should focus on users present in both phases of experiment. Maybe in 2nd level of analysis, I can study those users who were present in pre-experiment period but not during experiment period and vice-versa.\n",
    "\n",
    "I take this approach because since I have already filtered out users who had signed up 150 days before experiment, therefore there absolute absence in either the pre or during experiment phase is likely to be due to a cause unrelated to new UI design launched"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Yw4Fj1TiH19D",
    "outputId": "4bad1dde-3d0f-4ce1-9338-c51350251ac4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(923767, 8)"
      ]
     },
     "execution_count": 38,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "(923767, 8)"
      ]
     },
     "execution_count": 38,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Removing those users who had no activity registered during the pre-experiment phase\n",
    "temp = preexp_usr_final[preexp_usr_final.dt.isna()]\n",
    "temp = temp.drop_duplicates(subset=['uid'],keep='first')\n",
    "preexp_usr_final.shape\n",
    "preexp_usr_final = preexp_usr_final[~preexp_usr_final.uid.isin(temp.uid)]\n",
    "preexp_usr_final.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "GlfMdQNlH19H",
    "outputId": "a28f9690-f94a-4987-8293-cc6a2928f2e8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(35263, 15)"
      ]
     },
     "execution_count": 39,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Thus we see that we have 36k users for our study.\n"
     ]
    }
   ],
   "source": [
    "## Preparing user_ids\n",
    "uids = preexp_usr_final.drop_duplicates(subset=['uid'],keep='first')\n",
    "uids2 = usr_postexp_final.drop_duplicates(subset=['uid'],keep='first')\n",
    "uids = pd.merge(uids,uids2, on='uid',how='inner')\n",
    "uids.shape\n",
    "print(\"Thus we see that we have 36k users for our study.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "W0kihlSRH19J"
   },
   "source": [
    "### Filtering out preexp and postexperiment data for above users only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 225
    },
    "id": "nolwZJTzH19K",
    "outputId": "4bc82fbd-cb40-48e2-ec84-0b7387266270"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(916683, 8) (914047, 8)\n"
     ]
    },
    {
     "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>experiement_dt</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-11-07</td>\n",
       "      <td>2019-02-07</td>\n",
       "      <td>79.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-11-07</td>\n",
       "      <td>2019-02-09</td>\n",
       "      <td>211.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>reader</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    uid  variant_number experiement_dt  ... active_mins gender  user_type\n",
       "13    1               0     2019-02-06  ...   79.000000   male     reader\n",
       "14    1               0     2019-02-06  ...  211.000000   male     reader\n",
       "\n",
       "[2 rows x 8 columns]"
      ]
     },
     "execution_count": 40,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "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>uid</th>\n",
       "      <th>variant_number</th>\n",
       "      <th>experiement_dt</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-11-07</td>\n",
       "      <td>2018-09-09</td>\n",
       "      <td>371.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>reader</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-11-07</td>\n",
       "      <td>2018-09-10</td>\n",
       "      <td>70.000000</td>\n",
       "      <td>male</td>\n",
       "      <td>reader</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    uid  variant_number experiement_dt  ... active_mins gender  user_type\n",
       "44    1               0     2019-02-06  ...  371.000000   male     reader\n",
       "45    1               0     2019-02-06  ...   70.000000   male     reader\n",
       "\n",
       "[2 rows x 8 columns]"
      ]
     },
     "execution_count": 40,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lmtd_usr_postexp_df = usr_postexp_final[usr_postexp_final.uid.isin(uids.uid)]\n",
    "lmtd_preexp_usr_df = preexp_usr_final[preexp_usr_final.uid.isin(uids.uid)]\n",
    "print(lmtd_preexp_usr_df.shape,lmtd_usr_postexp_df.shape)\n",
    "lmtd_usr_postexp_df.head(2)\n",
    "lmtd_preexp_usr_df.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WvVghog1H19N"
   },
   "source": [
    "#### Populating the above two df: preexp_usr_df2 & usr_postexp_dfnew2  for each user * 150 days of experiment. This will give us an apple-to-apply comparison, ie the daily acitve _min for each user for every single day when the experiment was live"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "W0206TelH19O",
    "outputId": "e299235c-4ca2-41b1-c13e-91a60bb90888"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Post Exp Data: Before a& After reshaping the data at daily level: (914047, 8) (5289450, 8)\n"
     ]
    }
   ],
   "source": [
    "daily_usr_act_prepost = lmtd_usr_postexp_df.pivot_table(index=['variant_number','uid','experiement_dt','signup_date',\n",
    "                                                      'gender','user_type'],columns = 'dt',values='active_mins',fill_value=0).reset_index()\n",
    "daily_usr_act_prepost = daily_usr_act_prepost.melt(id_vars=['variant_number','uid','experiement_dt','signup_date',\n",
    "                                                      'gender','user_type'], var_name='dt',value_name='active_mins')\n",
    "print(\"Post Exp Data: Before a& After reshaping the data at daily level:\",lmtd_usr_postexp_df.shape, daily_usr_act_prepost.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "2QNSmaUvH19P",
    "outputId": "eef6005c-63a2-49c8-958b-261025eb64e8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pre Exp Data:Before a& After reshaping the data at daily level: (916683, 8) (5289450, 8)\n"
     ]
    }
   ],
   "source": [
    "daily_usr_act_pre = lmtd_preexp_usr_df.pivot_table(index=['variant_number','uid','experiement_dt','signup_date',\n",
    "                                                      'gender','user_type'],columns = 'dt',values='active_mins',fill_value=0).reset_index()\n",
    "daily_usr_act_pre = daily_usr_act_pre.melt(id_vars=['variant_number','uid','experiement_dt','signup_date',\n",
    "                                                      'gender','user_type'], var_name='dt',value_name='active_mins')\n",
    "print(\"Pre Exp Data:Before a& After reshaping the data at daily level:\",lmtd_preexp_usr_df.shape, daily_usr_act_pre.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 449
    },
    "id": "oyusryYtH19S",
    "outputId": "ac9fb1ca-b8ee-4cd4-a0ee-d0f1a76f5124"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Before pre & post experiment analysis:\n"
     ]
    },
    {
     "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>median</th>\n",
       "      <th>var</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>variant_number</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>366.763169</td>\n",
       "      <td>3.054808</td>\n",
       "      <td>19.151062</td>\n",
       "      <td>5613750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>444.027311</td>\n",
       "      <td>3.056017</td>\n",
       "      <td>21.071956</td>\n",
       "      <td>1381200</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                median        var     mean       std    count\n",
       "variant_number                                               \n",
       "0                    0 366.763169 3.054808 19.151062  5613750\n",
       "1                    0 444.027311 3.056017 21.071956  1381200"
      ]
     },
     "execution_count": 43,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "After keeping users present in both phase + signup_dt<=2018-09-09 + dt>='2018-09-09'\n"
     ]
    },
    {
     "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>median</th>\n",
       "      <th>var</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>variant_number</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>436.127385</td>\n",
       "      <td>3.597240</td>\n",
       "      <td>20.883663</td>\n",
       "      <td>4267500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>527.369301</td>\n",
       "      <td>3.664980</td>\n",
       "      <td>22.964523</td>\n",
       "      <td>1021950</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                median        var     mean       std    count\n",
       "variant_number                                               \n",
       "0                    0 436.127385 3.597240 20.883663  4267500\n",
       "1                    0 527.369301 3.664980 22.964523  1021950"
      ]
     },
     "execution_count": 43,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pre-experiment user activity for 0/1 variant number\n"
     ]
    },
    {
     "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>median</th>\n",
       "      <th>var</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>variant_number</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>444.105545</td>\n",
       "      <td>3.567514</td>\n",
       "      <td>21.073812</td>\n",
       "      <td>4267500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>185.709906</td>\n",
       "      <td>2.120730</td>\n",
       "      <td>13.627542</td>\n",
       "      <td>1021950</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                median        var     mean       std    count\n",
       "variant_number                                               \n",
       "0                    0 444.105545 3.567514 21.073812  4267500\n",
       "1                    0 185.709906 2.120730 13.627542  1021950"
      ]
     },
     "execution_count": 43,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#daily_usr_act.head()\n",
    "print(\"Before pre & post experiment analysis:\")\n",
    "daily_usr_act.groupby(['variant_number'],as_index=False).active_mins.agg({'mean','std','median','var','count'})\n",
    "print(\"After keeping users present in both phase + signup_dt<=2018-09-09 + dt>='2018-09-09'\")\n",
    "daily_usr_act_prepost.groupby(['variant_number'],as_index=False).active_mins.agg({'mean','std','median','var','count'})\n",
    "print(\"Pre-experiment user activity for 0/1 variant number\")\n",
    "daily_usr_act_pre.groupby(['variant_number'],as_index=False).active_mins.agg({'mean','std','median','var','count'})\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KOcrW317H19U"
   },
   "source": [
    "### From the above grpah we see that compared with the pre-experiment activity, for the treatment group the average active_mins have improved significantly whereas for the control group, the average active_mins is similar. Let's see if this same result is validated from Welch test too!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "z1l5qmy1H19V"
   },
   "outputs": [],
   "source": [
    "mean_0_pp = daily_usr_act_prepost[daily_usr_act_prepost['variant_number']==0].active_mins.mean()\n",
    "mean_1_pp = daily_usr_act_prepost[daily_usr_act_prepost['variant_number']==1].active_mins.mean()\n",
    "\n",
    "std_0_pp = daily_usr_act_prepost[daily_usr_act_prepost['variant_number']==0].active_mins.std()\n",
    "std_1_pp = daily_usr_act_prepost[daily_usr_act_prepost['variant_number']==1].active_mins.std()\n",
    "\n",
    "nobs0_pp = daily_usr_act_prepost[daily_usr_act_prepost['variant_number']==0].active_mins.count()\n",
    "nobs1_pp = daily_usr_act_prepost[daily_usr_act_prepost['variant_number']==1].active_mins.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "UuTU-GcTH19W",
    "outputId": "9455cdb8-483e-4d4d-d24d-df0d5ce45f7d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ttest_indResult(statistic=2.7243687718351812, pvalue=0.0064425297082406855)"
      ]
     },
     "execution_count": 45,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ttest_ind_from_stats(mean1=mean_1_pp,std1=std_1_pp,nobs1=nobs1_pp,\n",
    "                     mean2=mean_0_pp,std2=std_0_pp,nobs2=nobs0_pp, equal_var=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "C-RpJWtJH19Y"
   },
   "source": [
    "### The p-value is now significant taking the alpha=0.05. We now have enough evidence to reject our null hypothesis. It seems that the new UI design has an impact over user enagagment.\n",
    "\n",
    "### Calculating Confidence Interval"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "wS1eennFH19Z",
    "outputId": "80a4b4c1-82fe-406a-c170-d8640541f1cb"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.01900648458041384, 0.11647323484016717)"
      ]
     },
     "execution_count": 46,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CI = calc_CI(mean_1_pp,std_1_pp,nobs1_pp,mean_0_pp,std_0_pp,nobs0_pp,0.95)\n",
    "CI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "pLTL3qsWH19a",
    "outputId": "f5688015-b09b-4c5d-f689-9ab2bd6648e2"
   },
   "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>variant_number</th>\n",
       "      <th>uid</th>\n",
       "      <th>experiement_dt</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-11-07</td>\n",
       "      <td>male</td>\n",
       "      <td>reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-03-04</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2017-03-09</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-06-25</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2017-01-22</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   variant_number  uid experiement_dt  ...   user_type          dt active_mins\n",
       "0               0    1     2019-02-06  ...      reader  2019-02-06           0\n",
       "1               0    3     2019-02-06  ...  non_reader  2019-02-06           0\n",
       "2               0    4     2019-02-06  ...  non_reader  2019-02-06           0\n",
       "3               0    5     2019-02-06  ...  non_reader  2019-02-06           0\n",
       "4               0    6     2019-02-06  ...  non_reader  2019-02-06           0\n",
       "\n",
       "[5 rows x 8 columns]"
      ]
     },
     "execution_count": 47,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_usr_act_prepost.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "IUCCwzd4H19d"
   },
   "source": [
    "### Pre-Post Analysis with Covariates\n",
    "\n",
    "Now we will be exploring the impact of new UI across two main covariates: gender and user_types. Here, we are interested to see if there had been significant change for any one group or not."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 395
    },
    "id": "RiRKRap1H19e",
    "outputId": "1640d5aa-fd4e-4b1d-e18f-9b14741969ec"
   },
   "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>variant_number</th>\n",
       "      <th>uid</th>\n",
       "      <th>experiement_dt</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-11-07</td>\n",
       "      <td>male</td>\n",
       "      <td>reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-03-04</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2017-03-09</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-06-25</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2017-01-22</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   variant_number  uid experiement_dt  ...   user_type          dt active_mins\n",
       "0               0    1     2019-02-06  ...      reader  2019-02-06           0\n",
       "1               0    3     2019-02-06  ...  non_reader  2019-02-06           0\n",
       "2               0    4     2019-02-06  ...  non_reader  2019-02-06           0\n",
       "3               0    5     2019-02-06  ...  non_reader  2019-02-06           0\n",
       "4               0    6     2019-02-06  ...  non_reader  2019-02-06           0\n",
       "\n",
       "[5 rows x 8 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "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>variant_number</th>\n",
       "      <th>uid</th>\n",
       "      <th>experiement_dt</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2016-11-07</td>\n",
       "      <td>male</td>\n",
       "      <td>reader</td>\n",
       "      <td>2018-09-09</td>\n",
       "      <td>371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-03-04</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2018-09-09</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2017-03-09</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2018-09-09</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2018-06-25</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2018-09-09</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2017-01-22</td>\n",
       "      <td>female</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2018-09-09</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   variant_number  uid experiement_dt  ...   user_type          dt active_mins\n",
       "0               0    1     2019-02-06  ...      reader  2018-09-09         371\n",
       "1               0    3     2019-02-06  ...  non_reader  2018-09-09           0\n",
       "2               0    4     2019-02-06  ...  non_reader  2018-09-09           1\n",
       "3               0    5     2019-02-06  ...  non_reader  2018-09-09           0\n",
       "4               0    6     2019-02-06  ...  non_reader  2018-09-09           0\n",
       "\n",
       "[5 rows x 8 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_usr_act_prepost.head()\n",
    "daily_usr_act_pre.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 112
    },
    "id": "F6BCCTrCH19n",
    "outputId": "f11e2a20-1e41-4d9b-8b17-331cc0348812"
   },
   "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>variant_number</th>\n",
       "      <th>uid</th>\n",
       "      <th>experiement_dt</th>\n",
       "      <th>signup_date</th>\n",
       "      <th>gender</th>\n",
       "      <th>user_type</th>\n",
       "      <th>dt</th>\n",
       "      <th>active_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>40001</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2017-04-29</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>40002</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>2015-08-22</td>\n",
       "      <td>male</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>2019-02-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   variant_number    uid experiement_dt  ...   user_type          dt active_mins\n",
       "0               1  40001     2019-02-06  ...  non_reader  2019-02-06           0\n",
       "1               1  40002     2019-02-06  ...  non_reader  2019-02-06           0\n",
       "\n",
       "[2 rows x 8 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Merging both the pre & post dataset to have two main df with vairant number's activity pre & post new UI launch\n",
    "temp1 = daily_usr_act_prepost[daily_usr_act_prepost.variant_number==1]\n",
    "temp2 = daily_usr_act_pre[daily_usr_act_pre.variant_number==1]\n",
    "daily_usr_1 = pd.concat([temp1,temp2],ignore_index=True)\n",
    "\n",
    "daily_usr_1.head(2)\n",
    "del temp1\n",
    "del temp2\n",
    "\n",
    "check10 = daily_usr_act_prepost[daily_usr_act_prepost.variant_number==0]\n",
    "check20 = daily_usr_act_pre[daily_usr_act_pre.variant_number==0]\n",
    "daily_usr_0 = pd.concat([check10,check20],ignore_index=True)\n",
    "\n",
    "del check10\n",
    "del check20"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KmPNlM-FH19h"
   },
   "source": [
    "### Checking if the distribution of users across gender and user_types is balanced or not!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 255
    },
    "id": "h7YL3s94H19h",
    "outputId": "cfe589e0-4a06-438c-f9ad-5f7f9a7ffa3a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "50000 50000 50000\n"
     ]
    },
    {
     "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>variant_number</th>\n",
       "      <th>gender</th>\n",
       "      <th>uid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>female</td>\n",
       "      <td>11607</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>male</td>\n",
       "      <td>22237</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>6156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>2870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>5520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>unknown</td>\n",
       "      <td>1610</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   variant_number   gender    uid\n",
       "0               0   female  11607\n",
       "1               0     male  22237\n",
       "2               0  unknown   6156\n",
       "3               1   female   2870\n",
       "4               1     male   5520\n",
       "5               1  unknown   1610"
      ]
     },
     "execution_count": 50,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = pd.merge(t2_usr_variant,t4_usr_attributes,on=\"uid\",how=\"left\")\n",
    "print(t2_usr_variant.shape[0], t4_usr_attributes.shape[0], temp.shape[0])\n",
    "temp.groupby(['variant_number','gender'],as_index=False).agg({'uid':'count'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "B9JW5KY0H19j"
   },
   "source": [
    "### Thus we see that though the distribution of male-female users are same in both variant group the male users are more prominant than the female users"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nlVaPBj0H19k"
   },
   "source": [
    "### Gender-wise active_mins spent pre & post the new UI launch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 556
    },
    "id": "IzYmJ24rH19k",
    "outputId": "d969e612-68d7-4c96-b253-d7ece3639d99"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Before the Experiment:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>median</th>\n",
       "      <th>std</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>variant_number</th>\n",
       "      <th>gender</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">0</th>\n",
       "      <th>female</th>\n",
       "      <td>2.821831</td>\n",
       "      <td>0</td>\n",
       "      <td>18.242764</td>\n",
       "      <td>1176300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>male</th>\n",
       "      <td>4.106219</td>\n",
       "      <td>0</td>\n",
       "      <td>22.868391</td>\n",
       "      <td>2491800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unknown</th>\n",
       "      <td>2.791405</td>\n",
       "      <td>0</td>\n",
       "      <td>18.215894</td>\n",
       "      <td>599400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">1</th>\n",
       "      <th>female</th>\n",
       "      <td>1.652412</td>\n",
       "      <td>0</td>\n",
       "      <td>10.456504</td>\n",
       "      <td>274650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>male</th>\n",
       "      <td>2.369502</td>\n",
       "      <td>0</td>\n",
       "      <td>14.972533</td>\n",
       "      <td>597750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unknown</th>\n",
       "      <td>1.986459</td>\n",
       "      <td>0</td>\n",
       "      <td>13.097676</td>\n",
       "      <td>149550</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           mean  median       std    count\n",
       "variant_number gender                                     \n",
       "0              female  2.821831       0 18.242764  1176300\n",
       "               male    4.106219       0 22.868391  2491800\n",
       "               unknown 2.791405       0 18.215894   599400\n",
       "1              female  1.652412       0 10.456504   274650\n",
       "               male    2.369502       0 14.972533   597750\n",
       "               unknown 1.986459       0 13.097676   149550"
      ]
     },
     "execution_count": 51,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "After the Experiment:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>median</th>\n",
       "      <th>std</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>variant_number</th>\n",
       "      <th>gender</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">0</th>\n",
       "      <th>female</th>\n",
       "      <td>2.775077</td>\n",
       "      <td>0</td>\n",
       "      <td>17.458284</td>\n",
       "      <td>1176300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>male</th>\n",
       "      <td>4.247247</td>\n",
       "      <td>0</td>\n",
       "      <td>23.219596</td>\n",
       "      <td>2491800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unknown</th>\n",
       "      <td>2.508522</td>\n",
       "      <td>0</td>\n",
       "      <td>16.165456</td>\n",
       "      <td>599400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">1</th>\n",
       "      <th>female</th>\n",
       "      <td>2.733810</td>\n",
       "      <td>0</td>\n",
       "      <td>17.665349</td>\n",
       "      <td>274650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>male</th>\n",
       "      <td>4.202844</td>\n",
       "      <td>0</td>\n",
       "      <td>25.561195</td>\n",
       "      <td>597750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unknown</th>\n",
       "      <td>3.225242</td>\n",
       "      <td>0</td>\n",
       "      <td>20.400976</td>\n",
       "      <td>149550</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           mean  median       std    count\n",
       "variant_number gender                                     \n",
       "0              female  2.775077       0 17.458284  1176300\n",
       "               male    4.247247       0 23.219596  2491800\n",
       "               unknown 2.508522       0 16.165456   599400\n",
       "1              female  2.733810       0 17.665349   274650\n",
       "               male    4.202844       0 25.561195   597750\n",
       "               unknown 3.225242       0 20.400976   149550"
      ]
     },
     "execution_count": 51,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Before the Experiment:\")\n",
    "daily_usr_act_pre.groupby(['variant_number','gender'],as_index=False).active_mins.agg({'mean','std','median','count'})\n",
    "\n",
    "print(\"After the Experiment:\")\n",
    "daily_usr_act_prepost.groupby(['variant_number','gender'],as_index=False).active_mins.agg({'mean','std','median','count'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 626
    },
    "id": "Q8zOnOBSH19w",
    "outputId": "763e6d54-b1c2-4990-b9ab-89dc3842c4b4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1296x576 with 0 Axes>"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4bab29f4e0>"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.0, 1.0, 'Average time spent by gender - Variant Group')"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4bab29f4e0>"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x7f4baad395f8>"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(2019-02-06, 1, 'experiment')"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4baa656eb8>"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.0, 1.0, 'Average time spent by gender - Control Group')"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4baa656eb8>"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x7f4baa8e2828>"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(2019-02-06, 2, 'experiment')"
      ]
     },
     "execution_count": 57,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(18,8))\n",
    "plt.subplot(2,1,1)\n",
    "plt.title('Average time spent by gender - Variant Group',loc='left', fontsize=15)\n",
    "data=daily_usr_1.groupby(['gender','dt'],as_index=False).active_mins.mean()\n",
    "sns.lineplot(data=data,x='dt',y='active_mins',hue='gender')\n",
    "plt.axvline(x = daily_usr_1.experiement_dt.max(), color = 'red')\n",
    "plt.text(x = daily_usr_1.experiement_dt.max(),y=1,s ='experiment', fontsize =15)\n",
    "plt.subplot(2,1,2)\n",
    "plt.title('Average time spent by gender - Control Group',loc='left',fontsize=15)\n",
    "data=daily_usr_0.groupby(['gender','dt'],as_index=False).active_mins.mean()\n",
    "sns.lineplot(data=data,x='dt',y='active_mins',hue='gender')\n",
    "plt.axvline(x = daily_usr_0.experiement_dt.max(), color = 'red')\n",
    "plt.text(x = daily_usr_0.experiement_dt.max(),y=2,s ='experiment', fontsize =15)\n",
    "plt.savefig(\"Time Series Analysis for Average time spent by gender\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "HyxPUrNZUbMO"
   },
   "source": [
    "Thus we see from above graph that for variant group=1 or Treatment group there is significant improvement in average acitve minutes. Also, comparing within the genders, we see male members have a higher activity in comparison their female counterparts. In control group there is not much change in acitivity."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "CKhHTdQNUPo9"
   },
   "source": [
    "# Checking if the distribution of users across user_types is balanced or not!¶"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 318
    },
    "id": "3RVrK_B7H195",
    "outputId": "6f4cb864-3da3-4931-ea29-1fe3c268bcb7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "50000 50000 50000\n"
     ]
    },
    {
     "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>variant_number</th>\n",
       "      <th>user_type</th>\n",
       "      <th>uid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>contributor</td>\n",
       "      <td>915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>new_user</td>\n",
       "      <td>3653</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>28699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>reader</td>\n",
       "      <td>6733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>contributor</td>\n",
       "      <td>129</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>new_user</td>\n",
       "      <td>1235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1</td>\n",
       "      <td>non_reader</td>\n",
       "      <td>7367</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1</td>\n",
       "      <td>reader</td>\n",
       "      <td>1269</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   variant_number    user_type    uid\n",
       "0               0  contributor    915\n",
       "1               0     new_user   3653\n",
       "2               0   non_reader  28699\n",
       "3               0       reader   6733\n",
       "4               1  contributor    129\n",
       "5               1     new_user   1235\n",
       "6               1   non_reader   7367\n",
       "7               1       reader   1269"
      ]
     },
     "execution_count": 53,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = pd.merge(t2_usr_variant,t4_usr_attributes,on=\"uid\",how=\"left\")\n",
    "print(t2_usr_variant.shape[0], t4_usr_attributes.shape[0], temp.shape[0])\n",
    "temp.groupby(['variant_number','user_type'],as_index=False).agg({'uid':'count'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "v54imRguU2dA"
   },
   "source": [
    "### Thus we see that though the distribution of reader, non-reader & contributor users are same in both variant group there is a high proportion of non-reader users in the control & treatment groups. Given that the aim was to increase user engagement amongst all types of users, therefore the distribution should have been balanced. But having said so, I am also wondering if there was some critical changes introduced in new UI meant for converting non-reader users to other types. This information will be probably more useful. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 556
    },
    "id": "idWQ8imfH1-A",
    "outputId": "2319af06-84e0-45bf-898f-b371fb59ac4c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Before the Experiment:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>median</th>\n",
       "      <th>std</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>variant_number</th>\n",
       "      <th>user_type</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">0</th>\n",
       "      <th>contributor</th>\n",
       "      <td>31.377330</td>\n",
       "      <td>0</td>\n",
       "      <td>78.160294</td>\n",
       "      <td>118350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non_reader</th>\n",
       "      <td>0.679109</td>\n",
       "      <td>0</td>\n",
       "      <td>4.643202</td>\n",
       "      <td>3279300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reader</th>\n",
       "      <td>10.672939</td>\n",
       "      <td>0</td>\n",
       "      <td>32.850768</td>\n",
       "      <td>869850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">1</th>\n",
       "      <th>contributor</th>\n",
       "      <td>20.423425</td>\n",
       "      <td>0</td>\n",
       "      <td>59.951643</td>\n",
       "      <td>16350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non_reader</th>\n",
       "      <td>0.632386</td>\n",
       "      <td>0</td>\n",
       "      <td>3.987967</td>\n",
       "      <td>838950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reader</th>\n",
       "      <td>7.817684</td>\n",
       "      <td>0</td>\n",
       "      <td>25.093694</td>\n",
       "      <td>166650</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                mean  median       std    count\n",
       "variant_number user_type                                       \n",
       "0              contributor 31.377330       0 78.160294   118350\n",
       "               non_reader   0.679109       0  4.643202  3279300\n",
       "               reader      10.672939       0 32.850768   869850\n",
       "1              contributor 20.423425       0 59.951643    16350\n",
       "               non_reader   0.632386       0  3.987967   838950\n",
       "               reader       7.817684       0 25.093694   166650"
      ]
     },
     "execution_count": 54,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "After the Experiment:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>median</th>\n",
       "      <th>std</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>variant_number</th>\n",
       "      <th>user_type</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">0</th>\n",
       "      <th>contributor</th>\n",
       "      <td>30.591390</td>\n",
       "      <td>0</td>\n",
       "      <td>75.813302</td>\n",
       "      <td>118350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non_reader</th>\n",
       "      <td>0.746917</td>\n",
       "      <td>0</td>\n",
       "      <td>5.362558</td>\n",
       "      <td>3279300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reader</th>\n",
       "      <td>10.670074</td>\n",
       "      <td>0</td>\n",
       "      <td>32.702177</td>\n",
       "      <td>869850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">1</th>\n",
       "      <th>contributor</th>\n",
       "      <td>31.027951</td>\n",
       "      <td>0</td>\n",
       "      <td>87.315865</td>\n",
       "      <td>16350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non_reader</th>\n",
       "      <td>1.164220</td>\n",
       "      <td>0</td>\n",
       "      <td>8.378647</td>\n",
       "      <td>838950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reader</th>\n",
       "      <td>13.569739</td>\n",
       "      <td>0</td>\n",
       "      <td>43.927194</td>\n",
       "      <td>166650</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                mean  median       std    count\n",
       "variant_number user_type                                       \n",
       "0              contributor 30.591390       0 75.813302   118350\n",
       "               non_reader   0.746917       0  5.362558  3279300\n",
       "               reader      10.670074       0 32.702177   869850\n",
       "1              contributor 31.027951       0 87.315865    16350\n",
       "               non_reader   1.164220       0  8.378647   838950\n",
       "               reader      13.569739       0 43.927194   166650"
      ]
     },
     "execution_count": 54,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Before the Experiment:\")\n",
    "daily_usr_act_pre.groupby(['variant_number','user_type'],as_index=False).active_mins.agg({'mean','std','median','count'})\n",
    "\n",
    "print(\"After the Experiment:\")\n",
    "daily_usr_act_prepost.groupby(['variant_number','user_type'],as_index=False).active_mins.agg({'mean','std','median','count'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 628
    },
    "id": "SS_KZ3GkH1-D",
    "outputId": "e23ff4d6-c86b-472e-c822-ab8183a9d08c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1296x576 with 0 Axes>"
      ]
     },
     "execution_count": 56,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4bad5bf0f0>"
      ]
     },
     "execution_count": 56,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.0, 1.0, 'Average time spent by user_type - Variant Group')"
      ]
     },
     "execution_count": 56,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4bad5bf0f0>"
      ]
     },
     "execution_count": 56,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x7f4bab4bb6a0>"
      ]
     },
     "execution_count": 56,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(2019-02-06, 1, 'experiment')"
      ]
     },
     "execution_count": 56,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4baacff208>"
      ]
     },
     "execution_count": 56,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.0, 1.0, 'Average time spent by user_type - Control Group')"
      ]
     },
     "execution_count": 56,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4baacff208>"
      ]
     },
     "execution_count": 56,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x7f4bab33e3c8>"
      ]
     },
     "execution_count": 56,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(2019-02-06, 2, 'experiment')"
      ]
     },
     "execution_count": 56,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(18,8))\n",
    "plt.subplot(2,1,1)\n",
    "plt.title('Average time spent by user_type - Variant Group',loc='left', fontsize=15)\n",
    "data=daily_usr_1.groupby(['user_type','dt'],as_index=False).active_mins.mean()\n",
    "sns.lineplot(data=data,x='dt',y='active_mins',hue='user_type')\n",
    "#plt.xticks(np.arange(daily_usr_1.dt.min(),daily_usr_1.dt.max(),30))\n",
    "plt.axvline(x = daily_usr_1.experiement_dt.max(), color = 'red')\n",
    "plt.text(x = daily_usr_1.experiement_dt.max(),y=1,s ='experiment', fontsize =15)\n",
    "plt.subplot(2,1,2)\n",
    "plt.title('Average time spent by user_type - Control Group',loc='left',fontsize=15)\n",
    "data=daily_usr_0.groupby(['user_type','dt'],as_index=False).active_mins.mean()\n",
    "sns.lineplot(data=data,x='dt',y='active_mins',hue='user_type')\n",
    "plt.axvline(x = daily_usr_0.experiement_dt.max(), color = 'red')\n",
    "plt.text(x = daily_usr_0.experiement_dt.max(),y=2,s ='experiment', fontsize =15)\n",
    "plt.savefig(\"Time Series Analysis for Average time spent by user_type\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "qLkgumtuZalg"
   },
   "source": [
    "## Performing Co-variates test\n",
    "\n",
    "### Gender"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "WIJP4zETZXap",
    "outputId": "356d0757-3f88-43d5-e58c-ff061dc672b3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "male\n",
      "T-results for user type- male\n",
      "Ttest_indResult(statistic=-40.89238052193022, pvalue=0.0)\n",
      "female\n",
      "T-results for user type- female\n",
      "Ttest_indResult(statistic=-26.555714805327135, pvalue=2.506335129311254e-155)\n",
      "unknown\n",
      "T-results for user type- unknown\n",
      "Ttest_indResult(statistic=-1.2571698339802369, pvalue=0.2086927807651504)\n"
     ]
    }
   ],
   "source": [
    "gender = ['male','female','unknown']\n",
    "for user in gender:\n",
    "    print(user)\n",
    "    mean_0_pp = daily_usr_0[daily_usr_0['gender']==user].active_mins.mean()\n",
    "    mean_1_pp = daily_usr_1[daily_usr_1['gender']==user].active_mins.mean()\n",
    "\n",
    "    std_0_pp = daily_usr_0[daily_usr_0['gender']==user].active_mins.std()\n",
    "    std_1_pp = daily_usr_1[daily_usr_1['gender']==user].active_mins.std()\n",
    "\n",
    "    nobs0_pp = daily_usr_0[daily_usr_0['gender']==user].active_mins.count()\n",
    "    nobs1_pp = daily_usr_1[daily_usr_1['gender']==user].active_mins.count()\n",
    "    \n",
    "    print('T-results for user type-',user)\n",
    "    print(ttest_ind_from_stats(mean1=mean_1_pp,std1=std_1_pp,nobs1=nobs1_pp,\n",
    "                     mean2=mean_0_pp,std2=std_0_pp,nobs2=nobs0_pp, equal_var=False))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "s8FOBkgHcvRi"
   },
   "source": [
    "### User_Type\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "IjIJwt0MZXuA",
    "outputId": "520270af-8bc6-4f31-b383-b2c6649d3efc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "contributor\n",
      "T-results for user type- contributor\n",
      "Ttest_indResult(statistic=-11.834916226870043, pvalue=2.8930249870937706e-32)\n",
      "reader\n",
      "T-results for user type- reader\n",
      "Ttest_indResult(statistic=0.3316864679767433, pvalue=0.7401261755642135)\n",
      "non_reader\n",
      "T-results for user type- non_reader\n",
      "Ttest_indResult(statistic=34.09324557971776, pvalue=1.0829136072296474e-254)\n"
     ]
    }
   ],
   "source": [
    "user_type = ['non_reader','contributor','reader']\n",
    "for user in user_type:\n",
    "    print(user)\n",
    "\n",
    "    mean_0_pp = daily_usr_0[daily_usr_0['user_type']==user].active_mins.mean()\n",
    "    mean_1_pp = daily_usr_1[daily_usr_1['user_type']==user].active_mins.mean()\n",
    "\n",
    "    std_0_pp = daily_usr_0[daily_usr_0['user_type']==user].active_mins.std()\n",
    "    std_1_pp = daily_usr_1[daily_usr_1['user_type']==user].active_mins.std()\n",
    "\n",
    "    nobs0_pp = daily_usr_0[daily_usr_0['user_type']==user].active_mins.count()\n",
    "    nobs1_pp = daily_usr_1[daily_usr_1['user_type']==user].active_mins.count()\n",
    "    \n",
    "    print('T-results for user type-',user)\n",
    "    print(ttest_ind_from_stats(mean1=mean_1_pp,std1=std_1_pp,nobs1=nobs1_pp,\n",
    "                     mean2=mean_0_pp,std2=std_0_pp,nobs2=nobs0_pp, equal_var=False))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "HdE6sNXYnjwH"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "xWJSF-KaZXzi"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "JHc_hYeEH1-F"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "RMoZxeRZH1-I",
    "scrolled": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "cmX9e3tPH1-L",
    "scrolled": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Quora_DataChallenge.ipynb",
   "provenance": [],
   "toc_visible": true
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}