11. User Referral Program.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.simplefilter('ignore')\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from scipy.stats import ttest_ind\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>date</th>\n",
       "      <th>country</th>\n",
       "      <th>money_spent</th>\n",
       "      <th>is_referral</th>\n",
       "      <th>device_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>FR</td>\n",
       "      <td>65</td>\n",
       "      <td>0</td>\n",
       "      <td>EVDCJTZMVMJDG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>CA</td>\n",
       "      <td>54</td>\n",
       "      <td>0</td>\n",
       "      <td>WUBZFTVKXGQQX</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>FR</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>CBAPCJRTFNUJG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>UK</td>\n",
       "      <td>73</td>\n",
       "      <td>0</td>\n",
       "      <td>PRGXJZAJKMXRH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>MX</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>PRGXJZAJKMXRH</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id       date country  money_spent  is_referral      device_id\n",
       "0        2 2015-10-03      FR           65            0  EVDCJTZMVMJDG\n",
       "1        3 2015-10-03      CA           54            0  WUBZFTVKXGQQX\n",
       "2        6 2015-10-03      FR           35            0  CBAPCJRTFNUJG\n",
       "3        7 2015-10-03      UK           73            0  PRGXJZAJKMXRH\n",
       "4        7 2015-10-03      MX           35            0  PRGXJZAJKMXRH"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('./data/referral.csv', parse_dates=['date'])\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 97341 entries, 0 to 97340\n",
      "Data columns (total 6 columns):\n",
      "user_id        97341 non-null int64\n",
      "date           97341 non-null datetime64[ns]\n",
      "country        97341 non-null object\n",
      "money_spent    97341 non-null int64\n",
      "is_referral    97341 non-null int64\n",
      "device_id      97341 non-null object\n",
      "dtypes: datetime64[ns](1), int64(3), object(2)\n",
      "memory usage: 4.5+ MB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unique country:\t 9\n",
      "Unique user id:\t 18809\n",
      "Unique device:\t 17887\n"
     ]
    }
   ],
   "source": [
    "print('Unique country:\\t', len(data['country'].unique()))\n",
    "print('Unique user id:\\t', len(data['user_id'].unique()))\n",
    "print('Unique device:\\t', len(data['device_id'].unique()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# extract date information\n",
    "data['dayofweek'] = data['date'].apply(lambda x: x.dayofweek)\n",
    "start_date = pd.to_datetime('2015-10-31')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>date</th>\n",
       "      <th>country</th>\n",
       "      <th>money_spent</th>\n",
       "      <th>is_referral</th>\n",
       "      <th>device_id</th>\n",
       "      <th>dayofweek</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>FR</td>\n",
       "      <td>65</td>\n",
       "      <td>0</td>\n",
       "      <td>EVDCJTZMVMJDG</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>CA</td>\n",
       "      <td>54</td>\n",
       "      <td>0</td>\n",
       "      <td>WUBZFTVKXGQQX</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>FR</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>CBAPCJRTFNUJG</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>UK</td>\n",
       "      <td>73</td>\n",
       "      <td>0</td>\n",
       "      <td>PRGXJZAJKMXRH</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>MX</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>PRGXJZAJKMXRH</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id       date country  money_spent  is_referral      device_id  \\\n",
       "0        2 2015-10-03      FR           65            0  EVDCJTZMVMJDG   \n",
       "1        3 2015-10-03      CA           54            0  WUBZFTVKXGQQX   \n",
       "2        6 2015-10-03      FR           35            0  CBAPCJRTFNUJG   \n",
       "3        7 2015-10-03      UK           73            0  PRGXJZAJKMXRH   \n",
       "4        7 2015-10-03      MX           35            0  PRGXJZAJKMXRH   \n",
       "\n",
       "   dayofweek  \n",
       "0          5  \n",
       "1          5  \n",
       "2          5  \n",
       "3          5  \n",
       "4          5  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x864 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# daily money spent visualization\n",
    "fig, ax = plt.subplots(nrows=2, ncols=1, figsize=(18, 12))\n",
    "sns.countplot(x='date', data=data, ax=ax[0])\n",
    "ax[0].set_ylabel('Daily Transaction Number', fontsize=12)\n",
    "\n",
    "sns.barplot(x='date', y='money_spent', data=data, ax=ax[1])\n",
    "ax[1].set_ylabel('Daily Transaction Amount', fontsize=12)\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x864 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# daily money spent visualization\n",
    "fig, ax = plt.subplots(nrows=2, ncols=1, figsize=(18, 12))\n",
    "sns.countplot(x='country', hue='is_referral', data=data, ax=ax[0])\n",
    "ax[0].set_ylabel('Country-wise Transaction Number', fontsize=12)\n",
    "\n",
    "sns.barplot(x='country', y='money_spent', hue='is_referral', data=data, ax=ax[1])\n",
    "ax[1].set_ylabel('Country-wise Transaction Amount', fontsize=12)\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x864 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# daily money spent visualization (Monday is 0 and Sunday is 6)\n",
    "fig, ax = plt.subplots(nrows=2, ncols=1, figsize=(18, 12))\n",
    "sns.countplot(x='dayofweek', hue='is_referral', data=data, ax=ax[0])\n",
    "ax[0].set_ylabel('Day-wise Transaction Number', fontsize=12)\n",
    "\n",
    "sns.barplot(x='dayofweek', y='money_spent', hue='is_referral', data=data, ax=ax[1])\n",
    "ax[1].set_ylabel('Day-wise Transaction Amount', fontsize=12)\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>date</th>\n",
       "      <th>country</th>\n",
       "      <th>money_spent</th>\n",
       "      <th>is_referral</th>\n",
       "      <th>device_id</th>\n",
       "      <th>dayofweek</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>FR</td>\n",
       "      <td>65</td>\n",
       "      <td>0</td>\n",
       "      <td>EVDCJTZMVMJDG</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>CA</td>\n",
       "      <td>54</td>\n",
       "      <td>0</td>\n",
       "      <td>WUBZFTVKXGQQX</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>FR</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>CBAPCJRTFNUJG</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>UK</td>\n",
       "      <td>73</td>\n",
       "      <td>0</td>\n",
       "      <td>PRGXJZAJKMXRH</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>2015-10-03</td>\n",
       "      <td>MX</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>PRGXJZAJKMXRH</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id       date country  money_spent  is_referral      device_id  \\\n",
       "0        2 2015-10-03      FR           65            0  EVDCJTZMVMJDG   \n",
       "1        3 2015-10-03      CA           54            0  WUBZFTVKXGQQX   \n",
       "2        6 2015-10-03      FR           35            0  CBAPCJRTFNUJG   \n",
       "3        7 2015-10-03      UK           73            0  PRGXJZAJKMXRH   \n",
       "4        7 2015-10-03      MX           35            0  PRGXJZAJKMXRH   \n",
       "\n",
       "   dayofweek  \n",
       "0          5  \n",
       "1          5  \n",
       "2          5  \n",
       "3          5  \n",
       "4          5  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Daily Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def basic_stat(df):\n",
    "    \"\"\" function to calculate statistics \"\"\"\n",
    "    no_purchase = len(df)\n",
    "    money_spent = df['money_spent'].sum()\n",
    "    unique_user = len(df['user_id'].unique())\n",
    "    \n",
    "    return pd.Series([no_purchase, money_spent, unique_user], \n",
    "                     index=['no_purchase', 'money_spent', 'unique_user'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def daily_stat(data):\n",
    "    grouped = data.groupby('date').apply(basic_stat).reset_index()\n",
    "    ref = grouped[grouped['date'] < start_date]\n",
    "    new = grouped[grouped['date'] >= start_date]\n",
    "    \n",
    "    names = ['no_purchase', 'money_spent', 'unique_user']\n",
    "    vals = []\n",
    "    for name in names:\n",
    "        a = ref[name].values\n",
    "        b = new[name].values\n",
    "        # p-value should be one-side test\n",
    "        p_val = ttest_ind(a, b, equal_var=False)[1] / 2\n",
    "        vals.append(p_val)\n",
    "        \n",
    "    return pd.Series(vals, index=names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "no_purchase    0.348257\n",
       "money_spent    0.135194\n",
       "unique_user    0.059545\n",
       "dtype: float64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_stat(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Country-wise Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>no_purchase</th>\n",
       "      <th>money_spent</th>\n",
       "      <th>unique_user</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CA</td>\n",
       "      <td>0.233985</td>\n",
       "      <td>0.351704</td>\n",
       "      <td>0.268256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CH</td>\n",
       "      <td>0.003072</td>\n",
       "      <td>0.006941</td>\n",
       "      <td>0.003142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DE</td>\n",
       "      <td>0.011798</td>\n",
       "      <td>0.081459</td>\n",
       "      <td>0.015665</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ES</td>\n",
       "      <td>0.072638</td>\n",
       "      <td>0.037522</td>\n",
       "      <td>0.057954</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>FR</td>\n",
       "      <td>0.058996</td>\n",
       "      <td>0.031843</td>\n",
       "      <td>0.041124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>IT</td>\n",
       "      <td>0.057454</td>\n",
       "      <td>0.025730</td>\n",
       "      <td>0.043911</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>MX</td>\n",
       "      <td>0.032430</td>\n",
       "      <td>0.009670</td>\n",
       "      <td>0.026203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>UK</td>\n",
       "      <td>0.261183</td>\n",
       "      <td>0.048490</td>\n",
       "      <td>0.204398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>US</td>\n",
       "      <td>0.307801</td>\n",
       "      <td>0.248874</td>\n",
       "      <td>0.368860</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  country  no_purchase  money_spent  unique_user\n",
       "0      CA     0.233985     0.351704     0.268256\n",
       "1      CH     0.003072     0.006941     0.003142\n",
       "2      DE     0.011798     0.081459     0.015665\n",
       "3      ES     0.072638     0.037522     0.057954\n",
       "4      FR     0.058996     0.031843     0.041124\n",
       "5      IT     0.057454     0.025730     0.043911\n",
       "6      MX     0.032430     0.009670     0.026203\n",
       "7      UK     0.261183     0.048490     0.204398\n",
       "8      US     0.307801     0.248874     0.368860"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.groupby('country').apply(daily_stat).reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Day-of-week-wise Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>dayofweek</th>\n",
       "      <th>no_purchase</th>\n",
       "      <th>money_spent</th>\n",
       "      <th>unique_user</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0.012484</td>\n",
       "      <td>0.000368</td>\n",
       "      <td>0.000665</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0.156965</td>\n",
       "      <td>0.007701</td>\n",
       "      <td>0.008802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0.009713</td>\n",
       "      <td>0.000301</td>\n",
       "      <td>0.001270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0.016584</td>\n",
       "      <td>0.000675</td>\n",
       "      <td>0.002578</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0.016875</td>\n",
       "      <td>0.000595</td>\n",
       "      <td>0.000683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>0.006828</td>\n",
       "      <td>0.000107</td>\n",
       "      <td>0.000125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>0.011717</td>\n",
       "      <td>0.000133</td>\n",
       "      <td>0.000012</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   dayofweek  no_purchase  money_spent  unique_user\n",
       "0          0     0.012484     0.000368     0.000665\n",
       "1          1     0.156965     0.007701     0.008802\n",
       "2          2     0.009713     0.000301     0.001270\n",
       "3          3     0.016584     0.000675     0.002578\n",
       "4          4     0.016875     0.000595     0.000683\n",
       "5          5     0.006828     0.000107     0.000125\n",
       "6          6     0.011717     0.000133     0.000012"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.groupby('dayofweek').apply(daily_stat).reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Questions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "According to the analysis above, the program doesn't seem have significant impacts to the whole company as a whole.\n",
    "\n",
    "However, based on each country, I find the program has different impact on different country:\n",
    "\n",
    "- the program fails in CH and DE, it significantly decrease the sales in these two countries.\n",
    "- the program succeeds in 'MX','IT','FR','ES', it significantly increase the sales.\n",
    "- the program doesn't seem have any significant effect on UK,CA,US, especially on CA and US."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* first I suggest perform more accurate A/B test (see question 3's answer) and collect more data, to study the impact of the program\n",
    "* since the program has different impact in different country, I suggest studying the reason of such difference. for example, does the program has any cultural conflicts in CH and DE?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This approach isn't an accurate A/B test. \"User Referral\" program isn't the only difference between control group and test group. for example, there may be some special holiday after Oct 31 in some country. or just because the weather get colder after Oct 31, people's requirement on some goods are increased.\n",
    "\n",
    "To get more accurate impact of the program, we need to perform a more careful A/B test. for example:\n",
    "* during the same peroid of time\n",
    "* randomly split the customers into two groups, and let only one group know the User Referral program.\n",
    "* run the experiment some time, then perform the t-test to see whether some 'sale performance index' (e.g., daily spent, daily customers, daily transactions) have significant changes or not."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}