Analyze_ab_test_results_notebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analyze A/B Test Results\n",
    "\n",
    "## Table of Contents\n",
    "- [Introduction](#intro)\n",
    "- [Part I - Probability](#probability)\n",
    "- [Part II - A/B Test](#ab_test)\n",
    "- [Part III - Regression](#regression)\n",
    "- [Conclusions](#conclusion)\n",
    "\n",
    "<a id='intro'></a>\n",
    "### Introduction\n",
    "\n",
    "A/B tests are very commonly performed by data analysts and data scientists.  It is important that you get some practice working with the difficulties of these \n",
    "\n",
    "For this project, you will be working to understand the results of an A/B test run by an e-commerce website.  Your goal is to work through this notebook to help the company understand if they should implement the new page, keep the old page, or perhaps run the experiment longer to make their decision.\n",
    "\n",
    "**As you work through this notebook, follow along in the classroom and answer the corresponding quiz questions associated with each question.** The labels for each classroom concept are provided for each question.  This will assure you are on the right track as you work through the project, and you can feel more confident in your final submission meeting the criteria.  As a final check, assure you meet all the criteria on the [RUBRIC](https://review.udacity.com/#!/projects/37e27304-ad47-4eb0-a1ab-8c12f60e43d0/rubric).\n",
    "\n",
    "<a id='probability'></a>\n",
    "#### Part I - Probability\n",
    "\n",
    "To get started, let's import our libraries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 340,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import random\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "#We are setting the seed to assure you get the same answers on quizzes as we set up\n",
    "random.seed(42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`1.` Now, read in the `ab_data.csv` data. Store it in `df`.  **Use your dataframe to answer the questions in Quiz 1 of the classroom.**\n",
    "\n",
    "a. Read in the dataset and take a look at the top few rows here:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 341,
   "metadata": {},
   "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>user_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>group</th>\n",
       "      <th>landing_page</th>\n",
       "      <th>converted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>851104</td>\n",
       "      <td>2017-01-21 22:11:48.556739</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>804228</td>\n",
       "      <td>2017-01-12 08:01:45.159739</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>661590</td>\n",
       "      <td>2017-01-11 16:55:06.154213</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>853541</td>\n",
       "      <td>2017-01-08 18:28:03.143765</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>864975</td>\n",
       "      <td>2017-01-21 01:52:26.210827</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id                   timestamp      group landing_page  converted\n",
       "0   851104  2017-01-21 22:11:48.556739    control     old_page          0\n",
       "1   804228  2017-01-12 08:01:45.159739    control     old_page          0\n",
       "2   661590  2017-01-11 16:55:06.154213  treatment     new_page          0\n",
       "3   853541  2017-01-08 18:28:03.143765  treatment     new_page          0\n",
       "4   864975  2017-01-21 01:52:26.210827    control     old_page          1"
      ]
     },
     "execution_count": 341,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('ab_data.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "b. Use the cell below to find the number of rows in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 342,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(294478, 5)"
      ]
     },
     "execution_count": 342,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "c. The number of unique users in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 343,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "290584"
      ]
     },
     "execution_count": 343,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['user_id'].nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "d. The proportion of users converted."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 344,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.12126269856564711"
      ]
     },
     "execution_count": 344,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.query('converted == 1').count()[0]/df['user_id'].nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "e. The number of times the `new_page` and `treatment` don't match."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 345,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3893"
      ]
     },
     "execution_count": 345,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[((df['group'] == 'treatment') == (df['landing_page'] == 'new_page')) == False].count()[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "f. Do any of the rows have missing values?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 346,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "user_id         False\n",
       "timestamp       False\n",
       "group           False\n",
       "landing_page    False\n",
       "converted       False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 346,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().any()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`2.` For the rows where **treatment** does not match with **new_page** or **control** does not match with **old_page**, we cannot be sure if this row truly received the new or old page.  Use **Quiz 2** in the classroom to figure out how we should handle these rows.  \n",
    "\n",
    "a. Now use the answer to the quiz to create a new dataset that meets the specifications from the quiz.  Store your new dataframe in **df2**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 347,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2 = df.drop(df[((df['group'] == 'treatment') == (df['landing_page'] == 'new_page')) == False].index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 348,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 348,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Double Check all of the correct rows were removed - this should be 0\n",
    "df2[((df2['group'] == 'treatment') == (df2['landing_page'] == 'new_page')) == False].shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`3.` Use **df2** and the cells below to answer questions for **Quiz3** in the classroom."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "a. How many unique **user_id**s are in **df2**?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 349,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 290585 entries, 0 to 294477\n",
      "Data columns (total 5 columns):\n",
      "user_id         290585 non-null int64\n",
      "timestamp       290585 non-null object\n",
      "group           290585 non-null object\n",
      "landing_page    290585 non-null object\n",
      "converted       290585 non-null int64\n",
      "dtypes: int64(2), object(3)\n",
      "memory usage: 13.3+ MB\n"
     ]
    }
   ],
   "source": [
    "df2.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 350,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "290584"
      ]
     },
     "execution_count": 350,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['user_id'].nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "b. There is one **user_id** repeated in **df2**.  What is it?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 351,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 351,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['user_id'].duplicated().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "c. What is the row information for the repeat **user_id**? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 352,
   "metadata": {},
   "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>user_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>group</th>\n",
       "      <th>landing_page</th>\n",
       "      <th>converted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1899</th>\n",
       "      <td>773192</td>\n",
       "      <td>2017-01-09 05:37:58.781806</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2893</th>\n",
       "      <td>773192</td>\n",
       "      <td>2017-01-14 02:55:59.590927</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      user_id                   timestamp      group landing_page  converted\n",
       "1899   773192  2017-01-09 05:37:58.781806  treatment     new_page          0\n",
       "2893   773192  2017-01-14 02:55:59.590927  treatment     new_page          0"
      ]
     },
     "execution_count": 352,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2[df2.duplicated(['user_id'], keep=False)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "d. Remove **one** of the rows with a duplicate **user_id**, but keep your dataframe as **df2**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 353,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 290584 entries, 0 to 294477\n",
      "Data columns (total 5 columns):\n",
      "user_id         290584 non-null int64\n",
      "timestamp       290584 non-null object\n",
      "group           290584 non-null object\n",
      "landing_page    290584 non-null object\n",
      "converted       290584 non-null int64\n",
      "dtypes: int64(2), object(3)\n",
      "memory usage: 13.3+ MB\n"
     ]
    }
   ],
   "source": [
    "df2 = df2.drop(df2.query('timestamp == \"2017-01-09 05:37:58.781806\"').index)\n",
    "df2.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`4.` Use **df2** in the cells below to answer the quiz questions related to **Quiz 4** in the classroom.\n",
    "\n",
    "a. What is the probability of an individual converting regardless of the page they receive?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 354,
   "metadata": {},
   "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>user_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>group</th>\n",
       "      <th>landing_page</th>\n",
       "      <th>converted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>851104</td>\n",
       "      <td>2017-01-21 22:11:48.556739</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>804228</td>\n",
       "      <td>2017-01-12 08:01:45.159739</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>661590</td>\n",
       "      <td>2017-01-11 16:55:06.154213</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>853541</td>\n",
       "      <td>2017-01-08 18:28:03.143765</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>864975</td>\n",
       "      <td>2017-01-21 01:52:26.210827</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id                   timestamp      group landing_page  converted\n",
       "0   851104  2017-01-21 22:11:48.556739    control     old_page          0\n",
       "1   804228  2017-01-12 08:01:45.159739    control     old_page          0\n",
       "2   661590  2017-01-11 16:55:06.154213  treatment     new_page          0\n",
       "3   853541  2017-01-08 18:28:03.143765  treatment     new_page          0\n",
       "4   864975  2017-01-21 01:52:26.210827    control     old_page          1"
      ]
     },
     "execution_count": 354,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 355,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.11959708724499628"
      ]
     },
     "execution_count": 355,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['converted'].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "b. Given that an individual was in the `control` group, what is the probability they converted?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 356,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.1203863045004612"
      ]
     },
     "execution_count": 356,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "control_converted_mean = df2.query('group == \"control\"')['converted'].mean()\n",
    "control_converted_mean"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "c. Given that an individual was in the `treatment` group, what is the probability they converted?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 357,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.11880806551510564"
      ]
     },
     "execution_count": 357,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "treatment_converted_mean = df2.query('group == \"treatment\"')['converted'].mean()\n",
    "treatment_converted_mean"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "d. What is the probability that an individual received the new page?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 358,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.50006194422266881"
      ]
     },
     "execution_count": 358,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2[df2['landing_page'] == \"new_page\"].nunique()[0]/df2.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "e. Consider your results from parts (a) through (d) above, and explain below whether you think there is sufficient evidence to conclude that the new treatment page leads to more conversions."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">**Although, the mean for old page getting converted (0.12) is slightly higher than the new page getting converted (0.11), there isn't any sufficient evidence. One reasoning for such a result could be, users declining to accept change. However, that bias haven't been taken into consideration. Therefore, NO conclusion could be made like the new treatment leads to more conversions.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='ab_test'></a>\n",
    "### Part II - A/B Test\n",
    "\n",
    "Notice that because of the time stamp associated with each event, you could technically run a hypothesis test continuously as each observation was observed.  \n",
    "\n",
    "However, then the hard question is do you stop as soon as one page is considered significantly better than another or does it need to happen consistently for a certain amount of time?  How long do you run to render a decision that neither page is better than another?  \n",
    "\n",
    "These questions are the difficult parts associated with A/B tests in general.  \n",
    "\n",
    "\n",
    "`1.` For now, consider you need to make the decision just based on all the data provided.  If you want to assume that the old page is better unless the new page proves to be definitely better at a Type I error rate of 5%, what should your null and alternative hypotheses be?  You can state your hypothesis in terms of words or in terms of **$p_{old}$** and **$p_{new}$**, which are the converted rates for the old and new pages."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">**Hypothesis Part II:**\n",
    "$$H_0: p_{new} - p_{old} \\leq 0$$\n",
    "\n",
    "$$H_1: p_{new} - p_{old} \\gt 0$$\n",
    "\n",
    ">**Null Hypotheis (H-0): The old page is better or equally effective as the new page.**\n",
    "\n",
    ">**Alternative Hypothesis (H-1): The new page is better than the old page.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`2.` Assume under the null hypothesis, $p_{new}$ and $p_{old}$ both have \"true\" success rates equal to the **converted** success rate regardless of page - that is $p_{new}$ and $p_{old}$ are equal. Furthermore, assume they are equal to the **converted** rate in **ab_data.csv** regardless of the page. <br><br>\n",
    "Use a sample size for each page equal to the ones in **ab_data.csv**.  <br><br>\n",
    "Perform the sampling distribution for the difference in **converted** between the two pages over 10,000 iterations of calculating an estimate from the null.  <br><br>\n",
    "Use the cells below to provide the necessary parts of this simulation.  If this doesn't make complete sense right now, don't worry - you are going to work through the problems below to complete this problem.  You can use **Quiz 5** in the classroom to make sure you are on the right track.<br><br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "a. What is the **conversion rate** for $p_{new}$ under the null? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 359,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.11959708724499628"
      ]
     },
     "execution_count": 359,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p_new = df2.converted.mean()\n",
    "p_new"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "b. What is the **conversion rate** for $p_{old}$ under the null? <br><br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 360,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.11959708724499628"
      ]
     },
     "execution_count": 360,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p_old = df2.converted.mean()\n",
    "p_old"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "c. What is $n_{new}$, the number of individuals in the treatment group?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 361,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "145310"
      ]
     },
     "execution_count": 361,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n_new = df2.query('group == \"treatment\"').count()[0]\n",
    "n_new"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "d. What is $n_{old}$, the number of individuals in the control group?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 362,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "145274"
      ]
     },
     "execution_count": 362,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n_old = df2.query('group == \"control\"').count()[0]\n",
    "n_old"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "e. Simulate $n_{new}$ transactions with a conversion rate of $p_{new}$ under the null.  Store these $n_{new}$ 1's and 0's in **new_page_converted**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 363,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17382"
      ]
     },
     "execution_count": 363,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_page_converted = np.random.choice([1,0],size = n_new, replace=True, p = (p_new, (1-p_new)))\n",
    "new_page_converted.sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "f. Simulate $n_{old}$ transactions with a conversion rate of $p_{old}$ under the null.  Store these $n_{old}$ 1's and 0's in **old_page_converted**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 364,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17257"
      ]
     },
     "execution_count": 364,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "old_page_converted = np.random.choice([1,0],size = n_old, replace=True, p = (p_old,1-p_old))\n",
    "old_page_converted.sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "g. Find $p_{new}$ - $p_{old}$ for your simulated values from part (e) and (f)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 365,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.00083080025049298012"
      ]
     },
     "execution_count": 365,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_page_converted.mean() - old_page_converted.mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "h. Create 10,000 $p_{new}$ - $p_{old}$ values using the same simulation process you used in parts (a) through (g) above. Store all 10,000 values in a NumPy array called **p_diffs**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 366,
   "metadata": {},
   "outputs": [],
   "source": [
    "p_diffs = []\n",
    "for _ in range(10000):\n",
    "    new_page_converted = np.random.choice([1,0],size = n_new, replace=True, p = (p_new,1-p_new))\n",
    "    old_page_converted = np.random.choice([1,0],size = n_old, replace=True, p = (p_old,1-p_old))\n",
    "    diff = new_page_converted.mean() - old_page_converted.mean()\n",
    "    p_diffs.append(diff)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "i. Plot a histogram of the **p_diffs**.  Does this plot look like what you expected?  Use the matching problem in the classroom to assure you fully understand what was computed here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 367,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f77e65ed748>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "p_diffs = np.array(p_diffs)\n",
    "plt.hist(p_diffs);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "j. What proportion of the **p_diffs** are greater than the actual difference observed in **ab_data.csv**?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 368,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.0015782389853555567"
      ]
     },
     "execution_count": 368,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#For the original dataset\n",
    "actual_difference = treatment_converted_mean - control_converted_mean\n",
    "actual_difference"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 369,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f77e65fc550>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Calculating for Null Hypothesis\n",
    "\n",
    "null_vals = np.random.normal(0, p_diffs.std(), p_diffs.size)\n",
    "plt.hist(null_vals);\n",
    "plt.axvline(x=actual_difference, color='red');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 370,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.90200000000000002"
      ]
     },
     "execution_count": 370,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(null_vals > actual_difference).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "k. Please explain using the vocabulary you've learned in this course what you just computed in part **j.**  What is this value called in scientific studies?  What does this value mean in terms of whether or not there is a difference between the new and old pages?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> **As the p-value is 0.90, we can conclude that the difference in conversion rate between new page and old page is not significant. We FAIL to REJECT the null hypothesis, thereby stating to agree with the null hypothesis that the old page may be better than or equal to the new page.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "l. We could also use a built-in to achieve similar results.  Though using the built-in might be easier to code, the above portions are a walkthrough of the ideas that are critical to correctly thinking about statistical significance. Fill in the below to calculate the number of conversions for each page, as well as the number of individuals who received each page. Let `n_old` and `n_new` refer the the number of rows associated with the old page and new pages, respectively."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 371,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Old pages that got converted:  17739\n",
      "New pages that got converted:  17498\n",
      "\n",
      "Number of old pages:  145274\n",
      "Number of new pages:  145310\n"
     ]
    }
   ],
   "source": [
    "import statsmodels.api as sm\n",
    "\n",
    "convert_old = df.query('landing_page == \"old_page\"')['converted'].sum()\n",
    "print('Old pages that got converted: ',convert_old)\n",
    "convert_new = df.query('landing_page == \"new_page\"')['converted'].sum()\n",
    "print('New pages that got converted: ',convert_new)\n",
    "\n",
    "print('')\n",
    "n_old = df2.query('landing_page == \"old_page\"').count()[0]\n",
    "print('Number of old pages: ',n_old)\n",
    "n_new = df2.query('landing_page == \"new_page\"').count()[0]\n",
    "print('Number of new pages: ',n_new)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "m. Now use `stats.proportions_ztest` to compute your test statistic and p-value.  [Here](https://docs.w3cub.com/statsmodels/generated/statsmodels.stats.proportion.proportions_ztest/) is a helpful link on using the built in."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 372,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Z score  1.39439055129\n",
      "P value  0.91840015009\n"
     ]
    }
   ],
   "source": [
    "from statsmodels.stats.proportion import proportions_ztest\n",
    "\n",
    "z_score, p_value = sm.stats.proportions_ztest([convert_old, convert_new],[n_old, n_new], alternative='smaller') \n",
    "print('Z score ',z_score)\n",
    "print('P value ',p_value)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "n. What do the z-score and p-value you computed in the previous question mean for the conversion rates of the old and new pages?  Do they agree with the findings in parts **j.** and **k.**?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> **The computed z-score and p-value suggest that the p-value (0.91) is very large (i.e. it is greater than 0.05) and therefore it FAILS to REJECT the null hypothesis. Thus we can say that new page is not better than old page. In other words, the conversion rate for new page is less than or equal to the old page.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='regression'></a>\n",
    "### Part III - A regression approach\n",
    "\n",
    "`1.` In this final part, you will see that the result you achieved in the A/B test in Part II above can also be achieved by performing regression.<br><br> \n",
    "\n",
    "a. Since each row is either a conversion or no conversion, what type of regression should you be performing in this case?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">**For this dataset, since we are categorizing the data based on the conversion, Logistic Regression should be performed.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "b. The goal is to use **statsmodels** to fit the regression model you specified in part **a.** to see if there is a significant difference in conversion based on which page a customer receives. However, you first need to create in df2 a column for the intercept, and create a dummy variable column for which page each user received.  Add an **intercept** column, as well as an **ab_page** column, which is 1 when an individual receives the **treatment** and 0 if **control**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 373,
   "metadata": {},
   "outputs": [],
   "source": [
    "import statsmodels.api as sm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 374,
   "metadata": {
    "scrolled": true
   },
   "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>user_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>group</th>\n",
       "      <th>landing_page</th>\n",
       "      <th>converted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>851104</td>\n",
       "      <td>2017-01-21 22:11:48.556739</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>804228</td>\n",
       "      <td>2017-01-12 08:01:45.159739</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>661590</td>\n",
       "      <td>2017-01-11 16:55:06.154213</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>853541</td>\n",
       "      <td>2017-01-08 18:28:03.143765</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>864975</td>\n",
       "      <td>2017-01-21 01:52:26.210827</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id                   timestamp      group landing_page  converted\n",
       "0   851104  2017-01-21 22:11:48.556739    control     old_page          0\n",
       "1   804228  2017-01-12 08:01:45.159739    control     old_page          0\n",
       "2   661590  2017-01-11 16:55:06.154213  treatment     new_page          0\n",
       "3   853541  2017-01-08 18:28:03.143765  treatment     new_page          0\n",
       "4   864975  2017-01-21 01:52:26.210827    control     old_page          1"
      ]
     },
     "execution_count": 374,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3 = df2\n",
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 375,
   "metadata": {},
   "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>user_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>group</th>\n",
       "      <th>landing_page</th>\n",
       "      <th>converted</th>\n",
       "      <th>intercept</th>\n",
       "      <th>treatment</th>\n",
       "      <th>control</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>851104</td>\n",
       "      <td>2017-01-21 22:11:48.556739</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>804228</td>\n",
       "      <td>2017-01-12 08:01:45.159739</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>661590</td>\n",
       "      <td>2017-01-11 16:55:06.154213</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>853541</td>\n",
       "      <td>2017-01-08 18:28:03.143765</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>864975</td>\n",
       "      <td>2017-01-21 01:52:26.210827</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id                   timestamp      group landing_page  converted  \\\n",
       "0   851104  2017-01-21 22:11:48.556739    control     old_page          0   \n",
       "1   804228  2017-01-12 08:01:45.159739    control     old_page          0   \n",
       "2   661590  2017-01-11 16:55:06.154213  treatment     new_page          0   \n",
       "3   853541  2017-01-08 18:28:03.143765  treatment     new_page          0   \n",
       "4   864975  2017-01-21 01:52:26.210827    control     old_page          1   \n",
       "\n",
       "   intercept  treatment  control  \n",
       "0          1          1        0  \n",
       "1          1          1        0  \n",
       "2          1          0        1  \n",
       "3          1          0        1  \n",
       "4          1          1        0  "
      ]
     },
     "execution_count": 375,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3['intercept'] = 1\n",
    "df3[['treatment','control']] = pd.get_dummies(df3['group'])\n",
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 376,
   "metadata": {},
   "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>user_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>group</th>\n",
       "      <th>landing_page</th>\n",
       "      <th>converted</th>\n",
       "      <th>intercept</th>\n",
       "      <th>ab_page</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>851104</td>\n",
       "      <td>2017-01-21 22:11:48.556739</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>804228</td>\n",
       "      <td>2017-01-12 08:01:45.159739</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>661590</td>\n",
       "      <td>2017-01-11 16:55:06.154213</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>853541</td>\n",
       "      <td>2017-01-08 18:28:03.143765</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>864975</td>\n",
       "      <td>2017-01-21 01:52:26.210827</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id                   timestamp      group landing_page  converted  \\\n",
       "0   851104  2017-01-21 22:11:48.556739    control     old_page          0   \n",
       "1   804228  2017-01-12 08:01:45.159739    control     old_page          0   \n",
       "2   661590  2017-01-11 16:55:06.154213  treatment     new_page          0   \n",
       "3   853541  2017-01-08 18:28:03.143765  treatment     new_page          0   \n",
       "4   864975  2017-01-21 01:52:26.210827    control     old_page          1   \n",
       "\n",
       "   intercept  ab_page  \n",
       "0          1        0  \n",
       "1          1        0  \n",
       "2          1        1  \n",
       "3          1        1  \n",
       "4          1        0  "
      ]
     },
     "execution_count": 376,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.drop(columns={'treatment'}, inplace=True)\n",
    "df3.rename(columns={'control':'ab_page'}, inplace=True)\n",
    "df3.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "c. Use **statsmodels** to instantiate your regression model on the two columns you created in part b., then fit the model using the two columns you created in part **b.** to predict whether or not an individual converts. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 377,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.366118\n",
      "         Iterations 6\n"
     ]
    }
   ],
   "source": [
    "log_mod = sm.Logit(df3['converted'], df3[['intercept','ab_page']])\n",
    "ans = log_mod.fit()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "d. Provide the summary of your model below, and use it as necessary to answer the following questions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 378,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td>Model:</td>              <td>Logit</td>       <td>No. Iterations:</td>    <td>6.0000</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Dependent Variable:</td>     <td>converted</td>    <td>Pseudo R-squared:</td>    <td>0.000</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "         <td>Date:</td>        <td>2020-06-11 04:32</td>       <td>AIC:</td>        <td>212780.3502</td>\n",
       "</tr>\n",
       "<tr>\n",
       "   <td>No. Observations:</td>       <td>290584</td>            <td>BIC:</td>        <td>212801.5095</td>\n",
       "</tr>\n",
       "<tr>\n",
       "       <td>Df Model:</td>              <td>1</td>         <td>Log-Likelihood:</td>  <td>-1.0639e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Df Residuals:</td>         <td>290582</td>          <td>LL-Null:</td>      <td>-1.0639e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "      <td>Converged:</td>           <td>1.0000</td>           <td>Scale:</td>         <td>1.0000</td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "      <td></td>       <th>Coef.</th>  <th>Std.Err.</th>     <th>z</th>      <th>P>|z|</th> <th>[0.025</th>  <th>0.975]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>intercept</th> <td>-1.9888</td>  <td>0.0081</td>  <td>-246.6690</td> <td>0.0000</td> <td>-2.0046</td> <td>-1.9730</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ab_page</th>   <td>-0.0150</td>  <td>0.0114</td>   <td>-1.3109</td>  <td>0.1899</td> <td>-0.0374</td> <td>0.0074</td> \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary2.Summary'>\n",
       "\"\"\"\n",
       "                          Results: Logit\n",
       "==================================================================\n",
       "Model:              Logit            No. Iterations:   6.0000     \n",
       "Dependent Variable: converted        Pseudo R-squared: 0.000      \n",
       "Date:               2020-06-11 04:32 AIC:              212780.3502\n",
       "No. Observations:   290584           BIC:              212801.5095\n",
       "Df Model:           1                Log-Likelihood:   -1.0639e+05\n",
       "Df Residuals:       290582           LL-Null:          -1.0639e+05\n",
       "Converged:          1.0000           Scale:            1.0000     \n",
       "-------------------------------------------------------------------\n",
       "              Coef.   Std.Err.      z      P>|z|    [0.025   0.975]\n",
       "-------------------------------------------------------------------\n",
       "intercept    -1.9888    0.0081  -246.6690  0.0000  -2.0046  -1.9730\n",
       "ab_page      -0.0150    0.0114    -1.3109  0.1899  -0.0374   0.0074\n",
       "==================================================================\n",
       "\n",
       "\"\"\""
      ]
     },
     "execution_count": 378,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ans.summary2()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "e. What is the p-value associated with **ab_page**? Why does it differ from the value you found in **Part II**? "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">**Hypothesis Part III:**\n",
    "$$H_0: p_{new} - p_{old} = 0$$\n",
    "$$H_1: p_{new} - p_{old} \\neq 0$$\n",
    "\n",
    ">**In this section, the hypothesis test is presented in a different way. Rather than the conversion rate for new page being greater than old page, here we are checking whether the difference in conversion rate between new page and old page is zero or not. That's why the p value is differs from the value found in Part II. Also, the obtained p value supports, that we fail to reject the null hypothesis. That means null hypothesis is true and there exsists no change in the conversion rate between new and old pages.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "f. Now, you are considering other things that might influence whether or not an individual converts.  Discuss why it is a good idea to consider other factors to add into your regression model.  Are there any disadvantages to adding additional terms into your regression model?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> **Yes, it is a good idea to add more factors in the regression model, since it can help us to assess the impact of various factors on the conversion rate and to predict better than before.**\n",
    "\n",
    ">**Some factors like age of the visitor, field of interest, background (eduaction/profession), average time spent on the websites, visual preferences (dark mode/light mode) might help to decide between keeping the old page and launching the new page.**\n",
    "\n",
    ">**One major drawback could be, even after adding new factors, they might not impact the conversion rate. Also, the dataset size would increase which would result into more cleaning and wrangling.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "g. Now along with testing if the conversion rate changes for different pages, also add an effect based on which country a user lives in. You will need to read in the **countries.csv** dataset and merge together your datasets on the appropriate rows.  [Here](https://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.join.html) are the docs for joining tables. \n",
    "\n",
    "Does it appear that country had an impact on conversion?  Don't forget to create dummy variables for these country columns - **Hint: You will need two columns for the three dummy variables.** Provide the statistical output as well as a written response to answer this question."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 379,
   "metadata": {},
   "outputs": [],
   "source": [
    "df4 = pd.read_csv('countries.csv')\n",
    "df5 = df4.set_index('user_id').join(df3.set_index('user_id'), how='inner')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 380,
   "metadata": {},
   "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>country</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>group</th>\n",
       "      <th>landing_page</th>\n",
       "      <th>converted</th>\n",
       "      <th>intercept</th>\n",
       "      <th>ab_page</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>user_id</th>\n",
       "      <th></th>\n",
       "      <th></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>834778</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-14 23:08:43.304998</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>928468</th>\n",
       "      <td>US</td>\n",
       "      <td>2017-01-23 14:44:16.387854</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>822059</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-16 14:04:14.719771</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>711597</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-22 03:14:24.763511</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>710616</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-16 13:14:44.000513</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        country                   timestamp      group landing_page  \\\n",
       "user_id                                                               \n",
       "834778       UK  2017-01-14 23:08:43.304998    control     old_page   \n",
       "928468       US  2017-01-23 14:44:16.387854  treatment     new_page   \n",
       "822059       UK  2017-01-16 14:04:14.719771  treatment     new_page   \n",
       "711597       UK  2017-01-22 03:14:24.763511    control     old_page   \n",
       "710616       UK  2017-01-16 13:14:44.000513  treatment     new_page   \n",
       "\n",
       "         converted  intercept  ab_page  \n",
       "user_id                                 \n",
       "834778           0          1        0  \n",
       "928468           0          1        1  \n",
       "822059           1          1        1  \n",
       "711597           0          1        0  \n",
       "710616           0          1        1  "
      ]
     },
     "execution_count": 380,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df5.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 381,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "US    203619\n",
       "UK     72466\n",
       "CA     14499\n",
       "Name: country, dtype: int64"
      ]
     },
     "execution_count": 381,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df5.country.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 382,
   "metadata": {},
   "outputs": [],
   "source": [
    "df5[['CA', 'UK', 'US']] = pd.get_dummies(df5['country'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 383,
   "metadata": {},
   "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>country</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>group</th>\n",
       "      <th>landing_page</th>\n",
       "      <th>converted</th>\n",
       "      <th>intercept</th>\n",
       "      <th>ab_page</th>\n",
       "      <th>CA</th>\n",
       "      <th>UK</th>\n",
       "      <th>US</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>user_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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>834778</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-14 23:08:43.304998</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>928468</th>\n",
       "      <td>US</td>\n",
       "      <td>2017-01-23 14:44:16.387854</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>822059</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-16 14:04:14.719771</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>711597</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-22 03:14:24.763511</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>710616</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-16 13:14:44.000513</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        country                   timestamp      group landing_page  \\\n",
       "user_id                                                               \n",
       "834778       UK  2017-01-14 23:08:43.304998    control     old_page   \n",
       "928468       US  2017-01-23 14:44:16.387854  treatment     new_page   \n",
       "822059       UK  2017-01-16 14:04:14.719771  treatment     new_page   \n",
       "711597       UK  2017-01-22 03:14:24.763511    control     old_page   \n",
       "710616       UK  2017-01-16 13:14:44.000513  treatment     new_page   \n",
       "\n",
       "         converted  intercept  ab_page  CA  UK  US  \n",
       "user_id                                             \n",
       "834778           0          1        0   0   1   0  \n",
       "928468           0          1        1   0   0   1  \n",
       "822059           1          1        1   0   1   0  \n",
       "711597           0          1        0   0   1   0  \n",
       "710616           0          1        1   0   1   0  "
      ]
     },
     "execution_count": 383,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df5.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 384,
   "metadata": {},
   "outputs": [],
   "source": [
    "df5.drop(columns={'CA'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 385,
   "metadata": {
    "scrolled": true
   },
   "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>country</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>group</th>\n",
       "      <th>landing_page</th>\n",
       "      <th>converted</th>\n",
       "      <th>intercept</th>\n",
       "      <th>ab_page</th>\n",
       "      <th>UK</th>\n",
       "      <th>US</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>user_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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>834778</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-14 23:08:43.304998</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>928468</th>\n",
       "      <td>US</td>\n",
       "      <td>2017-01-23 14:44:16.387854</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>822059</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-16 14:04:14.719771</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>711597</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-22 03:14:24.763511</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>710616</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-16 13:14:44.000513</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        country                   timestamp      group landing_page  \\\n",
       "user_id                                                               \n",
       "834778       UK  2017-01-14 23:08:43.304998    control     old_page   \n",
       "928468       US  2017-01-23 14:44:16.387854  treatment     new_page   \n",
       "822059       UK  2017-01-16 14:04:14.719771  treatment     new_page   \n",
       "711597       UK  2017-01-22 03:14:24.763511    control     old_page   \n",
       "710616       UK  2017-01-16 13:14:44.000513  treatment     new_page   \n",
       "\n",
       "         converted  intercept  ab_page  UK  US  \n",
       "user_id                                         \n",
       "834778           0          1        0   1   0  \n",
       "928468           0          1        1   0   1  \n",
       "822059           1          1        1   1   0  \n",
       "711597           0          1        0   1   0  \n",
       "710616           0          1        1   1   0  "
      ]
     },
     "execution_count": 385,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df5.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 386,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.366116\n",
      "         Iterations 6\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td>Model:</td>              <td>Logit</td>       <td>No. Iterations:</td>    <td>6.0000</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Dependent Variable:</td>     <td>converted</td>    <td>Pseudo R-squared:</td>    <td>0.000</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "         <td>Date:</td>        <td>2020-06-11 04:32</td>       <td>AIC:</td>        <td>212780.8333</td>\n",
       "</tr>\n",
       "<tr>\n",
       "   <td>No. Observations:</td>       <td>290584</td>            <td>BIC:</td>        <td>212812.5723</td>\n",
       "</tr>\n",
       "<tr>\n",
       "       <td>Df Model:</td>              <td>2</td>         <td>Log-Likelihood:</td>  <td>-1.0639e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Df Residuals:</td>         <td>290581</td>          <td>LL-Null:</td>      <td>-1.0639e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "      <td>Converged:</td>           <td>1.0000</td>           <td>Scale:</td>         <td>1.0000</td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "      <td></td>       <th>Coef.</th>  <th>Std.Err.</th>     <th>z</th>     <th>P>|z|</th> <th>[0.025</th>  <th>0.975]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>intercept</th> <td>-2.0375</td>  <td>0.0260</td>  <td>-78.3639</td> <td>0.0000</td> <td>-2.0885</td> <td>-1.9866</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>US</th>        <td>0.0408</td>   <td>0.0269</td>   <td>1.5178</td>  <td>0.1291</td> <td>-0.0119</td> <td>0.0935</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>UK</th>        <td>0.0507</td>   <td>0.0284</td>   <td>1.7863</td>  <td>0.0740</td> <td>-0.0049</td> <td>0.1064</td> \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary2.Summary'>\n",
       "\"\"\"\n",
       "                          Results: Logit\n",
       "==================================================================\n",
       "Model:              Logit            No. Iterations:   6.0000     \n",
       "Dependent Variable: converted        Pseudo R-squared: 0.000      \n",
       "Date:               2020-06-11 04:32 AIC:              212780.8333\n",
       "No. Observations:   290584           BIC:              212812.5723\n",
       "Df Model:           2                Log-Likelihood:   -1.0639e+05\n",
       "Df Residuals:       290581           LL-Null:          -1.0639e+05\n",
       "Converged:          1.0000           Scale:            1.0000     \n",
       "-------------------------------------------------------------------\n",
       "               Coef.   Std.Err.     z      P>|z|    [0.025   0.975]\n",
       "-------------------------------------------------------------------\n",
       "intercept     -2.0375    0.0260  -78.3639  0.0000  -2.0885  -1.9866\n",
       "US             0.0408    0.0269    1.5178  0.1291  -0.0119   0.0935\n",
       "UK             0.0507    0.0284    1.7863  0.0740  -0.0049   0.1064\n",
       "==================================================================\n",
       "\n",
       "\"\"\""
      ]
     },
     "execution_count": 386,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import statsmodels.api as sm\n",
    "\n",
    "log_mod2 = sm.Logit(df5['converted'],df5[['intercept','US','UK']])\n",
    "ans2 = log_mod2.fit()\n",
    "ans2.summary2()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">**The country attribute doesn't seem to have any impact on the conversion rate. In the above summary, we can observe that from the p values for both US and UK. They are not statistically significant. Additionally, their confidence interval also contains zero.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "h. Though you have now looked at the individual factors of country and page on conversion, we would now like to look at an interaction between page and country to see if there significant effects on conversion.  Create the necessary additional columns, and fit the new model.  \n",
    "\n",
    "Provide the summary results, and your conclusions based on the results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 387,
   "metadata": {},
   "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>country</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>group</th>\n",
       "      <th>landing_page</th>\n",
       "      <th>converted</th>\n",
       "      <th>intercept</th>\n",
       "      <th>ab_page</th>\n",
       "      <th>UK</th>\n",
       "      <th>US</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>user_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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>834778</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-14 23:08:43.304998</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>928468</th>\n",
       "      <td>US</td>\n",
       "      <td>2017-01-23 14:44:16.387854</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>822059</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-16 14:04:14.719771</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>711597</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-22 03:14:24.763511</td>\n",
       "      <td>control</td>\n",
       "      <td>old_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>710616</th>\n",
       "      <td>UK</td>\n",
       "      <td>2017-01-16 13:14:44.000513</td>\n",
       "      <td>treatment</td>\n",
       "      <td>new_page</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        country                   timestamp      group landing_page  \\\n",
       "user_id                                                               \n",
       "834778       UK  2017-01-14 23:08:43.304998    control     old_page   \n",
       "928468       US  2017-01-23 14:44:16.387854  treatment     new_page   \n",
       "822059       UK  2017-01-16 14:04:14.719771  treatment     new_page   \n",
       "711597       UK  2017-01-22 03:14:24.763511    control     old_page   \n",
       "710616       UK  2017-01-16 13:14:44.000513  treatment     new_page   \n",
       "\n",
       "         converted  intercept  ab_page  UK  US  \n",
       "user_id                                         \n",
       "834778           0          1        0   1   0  \n",
       "928468           0          1        1   0   1  \n",
       "822059           1          1        1   1   0  \n",
       "711597           0          1        0   1   0  \n",
       "710616           0          1        1   1   0  "
      ]
     },
     "execution_count": 387,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df5.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 388,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.366113\n",
      "         Iterations 6\n"
     ]
    }
   ],
   "source": [
    "import statsmodels.api as sm\n",
    "\n",
    "log_mod3 = sm.Logit(df5['converted'],df5[['intercept','ab_page','US','UK']])\n",
    "ans3 = log_mod3.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 389,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td>Model:</td>              <td>Logit</td>       <td>No. Iterations:</td>    <td>6.0000</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Dependent Variable:</td>     <td>converted</td>    <td>Pseudo R-squared:</td>    <td>0.000</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "         <td>Date:</td>        <td>2020-06-11 04:32</td>       <td>AIC:</td>        <td>212781.1253</td>\n",
       "</tr>\n",
       "<tr>\n",
       "   <td>No. Observations:</td>       <td>290584</td>            <td>BIC:</td>        <td>212823.4439</td>\n",
       "</tr>\n",
       "<tr>\n",
       "       <td>Df Model:</td>              <td>3</td>         <td>Log-Likelihood:</td>  <td>-1.0639e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Df Residuals:</td>         <td>290580</td>          <td>LL-Null:</td>      <td>-1.0639e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "      <td>Converged:</td>           <td>1.0000</td>           <td>Scale:</td>         <td>1.0000</td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "      <td></td>       <th>Coef.</th>  <th>Std.Err.</th>     <th>z</th>     <th>P>|z|</th> <th>[0.025</th>  <th>0.975]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>intercept</th> <td>-2.0300</td>  <td>0.0266</td>  <td>-76.2488</td> <td>0.0000</td> <td>-2.0822</td> <td>-1.9778</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ab_page</th>   <td>-0.0149</td>  <td>0.0114</td>   <td>-1.3069</td> <td>0.1912</td> <td>-0.0374</td> <td>0.0075</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>US</th>        <td>0.0408</td>   <td>0.0269</td>   <td>1.5161</td>  <td>0.1295</td> <td>-0.0119</td> <td>0.0934</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>UK</th>        <td>0.0506</td>   <td>0.0284</td>   <td>1.7835</td>  <td>0.0745</td> <td>-0.0050</td> <td>0.1063</td> \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary2.Summary'>\n",
       "\"\"\"\n",
       "                          Results: Logit\n",
       "==================================================================\n",
       "Model:              Logit            No. Iterations:   6.0000     \n",
       "Dependent Variable: converted        Pseudo R-squared: 0.000      \n",
       "Date:               2020-06-11 04:32 AIC:              212781.1253\n",
       "No. Observations:   290584           BIC:              212823.4439\n",
       "Df Model:           3                Log-Likelihood:   -1.0639e+05\n",
       "Df Residuals:       290580           LL-Null:          -1.0639e+05\n",
       "Converged:          1.0000           Scale:            1.0000     \n",
       "-------------------------------------------------------------------\n",
       "               Coef.   Std.Err.     z      P>|z|    [0.025   0.975]\n",
       "-------------------------------------------------------------------\n",
       "intercept     -2.0300    0.0266  -76.2488  0.0000  -2.0822  -1.9778\n",
       "ab_page       -0.0149    0.0114   -1.3069  0.1912  -0.0374   0.0075\n",
       "US             0.0408    0.0269    1.5161  0.1295  -0.0119   0.0934\n",
       "UK             0.0506    0.0284    1.7835  0.0745  -0.0050   0.1063\n",
       "==================================================================\n",
       "\n",
       "\"\"\""
      ]
     },
     "execution_count": 389,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ans3.summary2()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 390,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "intercept    0.131332\n",
       "ab_page      0.985168\n",
       "US           1.041599\n",
       "UK           1.051944\n",
       "dtype: float64"
      ]
     },
     "execution_count": 390,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.exp(ans3.params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 391,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.00010001000100010001"
      ]
     },
     "execution_count": 391,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1/_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 392,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "group\n",
       "control      0.120399\n",
       "treatment    0.118920\n",
       "Name: converted, dtype: float64"
      ]
     },
     "execution_count": 392,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('group').mean()['converted']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">**Even after including both the factors, page and country, we can observe from the above summary that these factors do not have p value less than 0.05 and thus are not statistically significant. They do not have any influence on the conversions.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='conclusion'></a>\n",
    "## Conclusions\n",
    "\n",
    ">**1) From the above analysis, we can conclude that the old page is better than or equally effective as the new page. In all the scenarios, we failed to reject this null hypothesis.**\n",
    "\n",
    ">**2) Inorder to launch the new page, the company needs to continue with the study for a longer duration and also need to take into account more factors which actually the conversions.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 393,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 393,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from subprocess import call\n",
    "call(['python', '-m', 'nbconvert', 'Analyze_ab_test_results_notebook.ipynb'])"
   ]
  }
 ],
 "metadata": {
  "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}