Experimentation and uplift testing - Task 2 .ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "-hJr3MKe7j0W"
   },
   "source": [
    "#Task 2-Experimentation and Uplift testing\n",
    "\n",
    "Julia has asked us to evaluate the performance of a store trial which was performed in stores 77, 86 and 88.\n",
    "\n",
    "This can be broken down by:\n",
    "\n",
    "1. total sales revenue\n",
    "2. total number of customers\n",
    "3. average number of transactions per customer\n",
    "\n",
    "Create a measure to compare different control stores to each of the trial stores to do this write a function to reduce having to re-do the analysis for each trial store. Consider using Pearson correlations or a metric such as a magnitude distance e.g. 1- (Observed distance – minimum distance)/(Maximum distance – minimum distance) as a measure.\n",
    "\n",
    "Once you have selected your control stores, compare each trial and control pair during the trial period. You want to test if total sales are significantly different in the trial period and if so, check if the driver of change is more purchasing customers or more purchases per customers etc.\n",
    "\n",
    "Main areas of Focus are :\n",
    "1. Select control stores – Explore data, define metrics, visualize graphs\n",
    "2. Assessment of the trial – insights/trends by comparing trial stores with control stores\n",
    "3. Collate findings – summarize and provide recommendations\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "oXok4X_bz8el"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 280
    },
    "id": "aqK3ctIA9hQZ",
    "outputId": "b0d3456c-4359-43c5-cb9a-949ae62cb7a1"
   },
   "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>LYLTY_CARD_NBR</th>\n",
       "      <th>DATE</th>\n",
       "      <th>STORE_NBR</th>\n",
       "      <th>TXN_ID</th>\n",
       "      <th>PROD_NBR</th>\n",
       "      <th>PROD_NAME</th>\n",
       "      <th>PROD_QTY</th>\n",
       "      <th>TOT_SALES</th>\n",
       "      <th>PACK_SIZE</th>\n",
       "      <th>BRAND</th>\n",
       "      <th>LIFESTAGE</th>\n",
       "      <th>PREMIUM_CUSTOMER</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1000</td>\n",
       "      <td>2018-10-17</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>Natural Chip        Compny SeaSalt175g</td>\n",
       "      <td>2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>175</td>\n",
       "      <td>NATURAL</td>\n",
       "      <td>YOUNG SINGLES/COUPLES</td>\n",
       "      <td>Premium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1002</td>\n",
       "      <td>2018-09-16</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>58</td>\n",
       "      <td>Red Rock Deli Chikn&amp;Garlic Aioli 150g</td>\n",
       "      <td>1</td>\n",
       "      <td>2.7</td>\n",
       "      <td>150</td>\n",
       "      <td>RRD</td>\n",
       "      <td>YOUNG SINGLES/COUPLES</td>\n",
       "      <td>Mainstream</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1003</td>\n",
       "      <td>2019-03-07</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>52</td>\n",
       "      <td>Grain Waves Sour    Cream&amp;Chives 210G</td>\n",
       "      <td>1</td>\n",
       "      <td>3.6</td>\n",
       "      <td>210</td>\n",
       "      <td>GRNWVES</td>\n",
       "      <td>YOUNG FAMILIES</td>\n",
       "      <td>Budget</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1003</td>\n",
       "      <td>2019-03-08</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>106</td>\n",
       "      <td>Natural ChipCo      Hony Soy Chckn175g</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>175</td>\n",
       "      <td>NATURAL</td>\n",
       "      <td>YOUNG FAMILIES</td>\n",
       "      <td>Budget</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1004</td>\n",
       "      <td>2018-11-02</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>96</td>\n",
       "      <td>WW Original Stacked Chips 160g</td>\n",
       "      <td>1</td>\n",
       "      <td>1.9</td>\n",
       "      <td>160</td>\n",
       "      <td>WOOLWORTHS</td>\n",
       "      <td>OLDER SINGLES/COUPLES</td>\n",
       "      <td>Mainstream</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   LYLTY_CARD_NBR        DATE  ...              LIFESTAGE  PREMIUM_CUSTOMER\n",
       "0            1000  2018-10-17  ...  YOUNG SINGLES/COUPLES           Premium\n",
       "1            1002  2018-09-16  ...  YOUNG SINGLES/COUPLES        Mainstream\n",
       "2            1003  2019-03-07  ...         YOUNG FAMILIES            Budget\n",
       "3            1003  2019-03-08  ...         YOUNG FAMILIES            Budget\n",
       "4            1004  2018-11-02  ...  OLDER SINGLES/COUPLES        Mainstream\n",
       "\n",
       "[5 rows x 12 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "qvi = pd.read_csv(\"/content/QVI_data.csv\")\n",
    "qvi.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7Boe3njjAXJh"
   },
   "source": [
    "**Checking for nulls**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "8n9HdQZD9hNP",
    "outputId": "3abb6a5e-afdd-44f1-b04a-590687d1671e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 264834 entries, 0 to 264833\n",
      "Data columns (total 12 columns):\n",
      " #   Column            Non-Null Count   Dtype  \n",
      "---  ------            --------------   -----  \n",
      " 0   LYLTY_CARD_NBR    264834 non-null  int64  \n",
      " 1   DATE              264834 non-null  object \n",
      " 2   STORE_NBR         264834 non-null  int64  \n",
      " 3   TXN_ID            264834 non-null  int64  \n",
      " 4   PROD_NBR          264834 non-null  int64  \n",
      " 5   PROD_NAME         264834 non-null  object \n",
      " 6   PROD_QTY          264834 non-null  int64  \n",
      " 7   TOT_SALES         264834 non-null  float64\n",
      " 8   PACK_SIZE         264834 non-null  int64  \n",
      " 9   BRAND             264834 non-null  object \n",
      " 10  LIFESTAGE         264834 non-null  object \n",
      " 11  PREMIUM_CUSTOMER  264834 non-null  object \n",
      "dtypes: float64(1), int64(6), object(5)\n",
      "memory usage: 24.2+ MB\n"
     ]
    }
   ],
   "source": [
    "qvi.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "c8StLoLt9hKJ"
   },
   "outputs": [],
   "source": [
    "qvi[\"DATE\"] = pd.to_datetime(qvi[\"DATE\"])\n",
    "qvi[\"YEARMONTH\"] = qvi[\"DATE\"].dt.strftime(\"%Y%m\").astype(\"int\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "U7jVJc6aAz1A"
   },
   "source": [
    "Compile each store's monthly:\n",
    "\n",
    "1. Total sales\n",
    "2. Number of customers,\n",
    "3. Average transactions per customer\n",
    "4. Average chips per customer\n",
    "5. Average price per unit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "BCE_4yoT9hHe"
   },
   "outputs": [],
   "source": [
    "def monthly_store_metrics():\n",
    "    store_yrmo_group = qvi.groupby([\"STORE_NBR\", \"YEARMONTH\"])\n",
    "    total = store_yrmo_group[\"TOT_SALES\"].sum()\n",
    "    num_cust = store_yrmo_group[\"LYLTY_CARD_NBR\"].nunique()\n",
    "    trans_per_cust = store_yrmo_group.size() / num_cust\n",
    "    avg_chips_per_cust = store_yrmo_group[\"PROD_QTY\"].sum() / num_cust\n",
    "    avg_chips_price = total / store_yrmo_group[\"PROD_QTY\"].sum()\n",
    "    aggregates = [total, num_cust, trans_per_cust, avg_chips_per_cust, avg_chips_price]\n",
    "    metrics = pd.concat(aggregates, axis=1)\n",
    "    metrics.columns = [\"TOT_SALES\", \"nCustomers\", \"nTxnPerCust\", \"nChipsPerTxn\", \"avgPricePerUnit\"]\n",
    "    return metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "3r6Z9h4A9hEn",
    "outputId": "5bd0797f-b652-46f2-8630-d1fba0aabc99"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 3169 entries, 0 to 3168\n",
      "Data columns (total 7 columns):\n",
      " #   Column           Non-Null Count  Dtype  \n",
      "---  ------           --------------  -----  \n",
      " 0   STORE_NBR        3169 non-null   int64  \n",
      " 1   YEARMONTH        3169 non-null   int64  \n",
      " 2   TOT_SALES        3169 non-null   float64\n",
      " 3   nCustomers       3169 non-null   int64  \n",
      " 4   nTxnPerCust      3169 non-null   float64\n",
      " 5   nChipsPerTxn     3169 non-null   float64\n",
      " 6   avgPricePerUnit  3169 non-null   float64\n",
      "dtypes: float64(4), int64(3)\n",
      "memory usage: 173.4 KB\n"
     ]
    }
   ],
   "source": [
    "qvi_monthly_metrics = monthly_store_metrics().reset_index()\n",
    "qvi_monthly_metrics.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DUUe-_zxBJ93"
   },
   "source": [
    "**Pre-Trial Observation as this filter only stores with full 12 months observation**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 286
    },
    "id": "r06UZ5339hBz",
    "outputId": "17191181-047c-451b-96bd-7374477bd847"
   },
   "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>STORE_NBR</th>\n",
       "      <th>YEARMONTH</th>\n",
       "      <th>TOT_SALES</th>\n",
       "      <th>nCustomers</th>\n",
       "      <th>nTxnPerCust</th>\n",
       "      <th>nChipsPerTxn</th>\n",
       "      <th>avgPricePerUnit</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>201807</td>\n",
       "      <td>206.9</td>\n",
       "      <td>49</td>\n",
       "      <td>1.061224</td>\n",
       "      <td>1.265306</td>\n",
       "      <td>3.337097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>201808</td>\n",
       "      <td>176.1</td>\n",
       "      <td>42</td>\n",
       "      <td>1.023810</td>\n",
       "      <td>1.285714</td>\n",
       "      <td>3.261111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>201809</td>\n",
       "      <td>278.8</td>\n",
       "      <td>59</td>\n",
       "      <td>1.050847</td>\n",
       "      <td>1.271186</td>\n",
       "      <td>3.717333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>201810</td>\n",
       "      <td>188.1</td>\n",
       "      <td>44</td>\n",
       "      <td>1.022727</td>\n",
       "      <td>1.318182</td>\n",
       "      <td>3.243103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>201811</td>\n",
       "      <td>192.6</td>\n",
       "      <td>46</td>\n",
       "      <td>1.021739</td>\n",
       "      <td>1.239130</td>\n",
       "      <td>3.378947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>201812</td>\n",
       "      <td>189.6</td>\n",
       "      <td>42</td>\n",
       "      <td>1.119048</td>\n",
       "      <td>1.357143</td>\n",
       "      <td>3.326316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1</td>\n",
       "      <td>201901</td>\n",
       "      <td>154.8</td>\n",
       "      <td>35</td>\n",
       "      <td>1.028571</td>\n",
       "      <td>1.200000</td>\n",
       "      <td>3.685714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2</td>\n",
       "      <td>201807</td>\n",
       "      <td>150.8</td>\n",
       "      <td>39</td>\n",
       "      <td>1.051282</td>\n",
       "      <td>1.179487</td>\n",
       "      <td>3.278261</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    STORE_NBR  YEARMONTH  TOT_SALES  ...  nTxnPerCust  nChipsPerTxn  avgPricePerUnit\n",
       "0           1     201807      206.9  ...     1.061224      1.265306         3.337097\n",
       "1           1     201808      176.1  ...     1.023810      1.285714         3.261111\n",
       "2           1     201809      278.8  ...     1.050847      1.271186         3.717333\n",
       "3           1     201810      188.1  ...     1.022727      1.318182         3.243103\n",
       "4           1     201811      192.6  ...     1.021739      1.239130         3.378947\n",
       "5           1     201812      189.6  ...     1.119048      1.357143         3.326316\n",
       "6           1     201901      154.8  ...     1.028571      1.200000         3.685714\n",
       "12          2     201807      150.8  ...     1.051282      1.179487         3.278261\n",
       "\n",
       "[8 rows x 7 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "observ_counts = qvi_monthly_metrics[\"STORE_NBR\"].value_counts()\n",
    "full_observ_index = observ_counts[observ_counts == 12].index\n",
    "full_observ = qvi_monthly_metrics[qvi_monthly_metrics[\"STORE_NBR\"].isin(full_observ_index)]\n",
    "pretrial_full_observ = full_observ[full_observ[\"YEARMONTH\"] < 201902]\n",
    "\n",
    "pretrial_full_observ.head(8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "IXfnW4zq9g_A"
   },
   "outputs": [],
   "source": [
    "def calcCorrTable(metricCol, storeComparison, inputTable=pretrial_full_observ):\n",
    "  control_store_nbrs = inputTable[~inputTable[\"STORE_NBR\"].isin([77, 86, 88])][\"STORE_NBR\"].unique()\n",
    "  corrs = pd.DataFrame(columns = [\"YEARMONTH\", \"Trial_Str\", \"Ctrl_Str\", \"Corr_Score\"])\n",
    "  trial_store = inputTable[inputTable[\"STORE_NBR\"] == storeComparison][metricCol].reset_index()\n",
    "  for control in control_store_nbrs:\n",
    "    concat_df = pd.DataFrame(columns = [\"YEARMONTH\", \"Trial_Str\", \"Ctrl_Str\", \"Corr_Score\"])\n",
    "    control_store = inputTable[inputTable[\"STORE_NBR\"] == control][metricCol].reset_index()\n",
    "    concat_df[\"Corr_Score\"] = trial_store.corrwith(control_store, axis=1)\n",
    "    concat_df[\"Trial_Str\"] = storeComparison\n",
    "    concat_df[\"Ctrl_Str\"] = control\n",
    "    concat_df[\"YEARMONTH\"] = list(inputTable[inputTable[\"STORE_NBR\"] == storeComparison][\"YEARMONTH\"])\n",
    "    corrs = pd.concat([corrs, concat_df])\n",
    "  return corrs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 286
    },
    "id": "SU3jrHJt9g8h",
    "outputId": "b68b3839-d6df-4d78-f474-9306ef63441b"
   },
   "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>YEARMONTH</th>\n",
       "      <th>Trial_Str</th>\n",
       "      <th>Ctrl_Str</th>\n",
       "      <th>Corr_Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>201807</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0.070414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>201808</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0.027276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>201809</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0.002389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>201810</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.020045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>201811</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0.030024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>201812</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0.063946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>201901</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0.001470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>201807</td>\n",
       "      <td>77</td>\n",
       "      <td>2</td>\n",
       "      <td>0.142957</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  YEARMONTH Trial_Str Ctrl_Str  Corr_Score\n",
       "0    201807        77        1    0.070414\n",
       "1    201808        77        1    0.027276\n",
       "2    201809        77        1    0.002389\n",
       "3    201810        77        1   -0.020045\n",
       "4    201811        77        1    0.030024\n",
       "5    201812        77        1    0.063946\n",
       "6    201901        77        1    0.001470\n",
       "0    201807        77        2    0.142957"
      ]
     },
     "execution_count": 12,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr_table = pd.DataFrame()\n",
    "for trial_num in [77, 86, 88]:\n",
    "    corr_table = pd.concat([corr_table, calcCorrTable([\"TOT_SALES\", \"nCustomers\", \"nTxnPerCust\", \"nChipsPerTxn\", \"avgPricePerUnit\"], trial_num)])\n",
    "    \n",
    "corr_table.head(8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "Z7YKfa1C9g5_"
   },
   "outputs": [],
   "source": [
    "def calculateMagnitudeDistance(metricCol, storeComparison, inputTable=pretrial_full_observ):\n",
    "    control_store_nbrs = inputTable[~inputTable[\"STORE_NBR\"].isin([77, 86, 88])][\"STORE_NBR\"].unique()\n",
    "    dists = pd.DataFrame()\n",
    "    trial_store = inputTable[inputTable[\"STORE_NBR\"] == storeComparison][metricCol]\n",
    "    for control in control_store_nbrs:\n",
    "        concat_df  = abs(inputTable[inputTable[\"STORE_NBR\"] == storeComparison].reset_index()[metricCol] - inputTable[inputTable[\"STORE_NBR\"] == control].reset_index()[metricCol])\n",
    "        concat_df[\"YEARMONTH\"] = list(inputTable[inputTable[\"STORE_NBR\"] == storeComparison][\"YEARMONTH\"])\n",
    "        concat_df[\"Trial_Str\"] = storeComparison\n",
    "        concat_df[\"Ctrl_Str\"] = control\n",
    "        dists = pd.concat([dists, concat_df])\n",
    "    for col in metricCol:\n",
    "        dists[col] = 1 - ((dists[col] - dists[col].min()) / (dists[col].max() - dists[col].min()))\n",
    "    dists[\"magnitude\"] = dists[metricCol].mean(axis=1)\n",
    "    return dists"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 404
    },
    "id": "audeWP_B9g3M",
    "outputId": "91439acf-9a64-465a-9d18-35d99eb16334"
   },
   "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>TOT_SALES</th>\n",
       "      <th>nCustomers</th>\n",
       "      <th>nTxnPerCust</th>\n",
       "      <th>nChipsPerTxn</th>\n",
       "      <th>avgPricePerUnit</th>\n",
       "      <th>YEARMONTH</th>\n",
       "      <th>Trial_Str</th>\n",
       "      <th>Ctrl_Str</th>\n",
       "      <th>magnitude</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.935431</td>\n",
       "      <td>0.980769</td>\n",
       "      <td>0.958035</td>\n",
       "      <td>0.739412</td>\n",
       "      <td>0.883569</td>\n",
       "      <td>201807</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0.899443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.942972</td>\n",
       "      <td>0.951923</td>\n",
       "      <td>0.993823</td>\n",
       "      <td>0.802894</td>\n",
       "      <td>0.886328</td>\n",
       "      <td>201808</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0.915588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.961503</td>\n",
       "      <td>0.836538</td>\n",
       "      <td>0.992126</td>\n",
       "      <td>0.730041</td>\n",
       "      <td>0.703027</td>\n",
       "      <td>201809</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0.844647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.988221</td>\n",
       "      <td>0.932692</td>\n",
       "      <td>0.989514</td>\n",
       "      <td>0.940460</td>\n",
       "      <td>0.590528</td>\n",
       "      <td>201810</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0.888283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.962149</td>\n",
       "      <td>0.951923</td>\n",
       "      <td>0.874566</td>\n",
       "      <td>0.730358</td>\n",
       "      <td>0.832481</td>\n",
       "      <td>201811</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0.870296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.207554</td>\n",
       "      <td>0.286822</td>\n",
       "      <td>0.462846</td>\n",
       "      <td>0.779879</td>\n",
       "      <td>0.923887</td>\n",
       "      <td>201809</td>\n",
       "      <td>88</td>\n",
       "      <td>272</td>\n",
       "      <td>0.532198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.346797</td>\n",
       "      <td>0.387597</td>\n",
       "      <td>0.571497</td>\n",
       "      <td>0.796875</td>\n",
       "      <td>0.971133</td>\n",
       "      <td>201810</td>\n",
       "      <td>88</td>\n",
       "      <td>272</td>\n",
       "      <td>0.614780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.286706</td>\n",
       "      <td>0.310078</td>\n",
       "      <td>0.623883</td>\n",
       "      <td>0.813241</td>\n",
       "      <td>0.966999</td>\n",
       "      <td>201811</td>\n",
       "      <td>88</td>\n",
       "      <td>272</td>\n",
       "      <td>0.600181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.347151</td>\n",
       "      <td>0.387597</td>\n",
       "      <td>0.376456</td>\n",
       "      <td>0.699748</td>\n",
       "      <td>0.962198</td>\n",
       "      <td>201812</td>\n",
       "      <td>88</td>\n",
       "      <td>272</td>\n",
       "      <td>0.554630</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.402353</td>\n",
       "      <td>0.449612</td>\n",
       "      <td>0.450378</td>\n",
       "      <td>0.739714</td>\n",
       "      <td>0.971335</td>\n",
       "      <td>201901</td>\n",
       "      <td>88</td>\n",
       "      <td>272</td>\n",
       "      <td>0.602678</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5397 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    TOT_SALES  nCustomers  nTxnPerCust  ...  Trial_Str  Ctrl_Str  magnitude\n",
       "0    0.935431    0.980769     0.958035  ...         77         1   0.899443\n",
       "1    0.942972    0.951923     0.993823  ...         77         1   0.915588\n",
       "2    0.961503    0.836538     0.992126  ...         77         1   0.844647\n",
       "3    0.988221    0.932692     0.989514  ...         77         1   0.888283\n",
       "4    0.962149    0.951923     0.874566  ...         77         1   0.870296\n",
       "..        ...         ...          ...  ...        ...       ...        ...\n",
       "2    0.207554    0.286822     0.462846  ...         88       272   0.532198\n",
       "3    0.346797    0.387597     0.571497  ...         88       272   0.614780\n",
       "4    0.286706    0.310078     0.623883  ...         88       272   0.600181\n",
       "5    0.347151    0.387597     0.376456  ...         88       272   0.554630\n",
       "6    0.402353    0.449612     0.450378  ...         88       272   0.602678\n",
       "\n",
       "[5397 rows x 9 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dist_table = pd.DataFrame()\n",
    "for trial_num in [77, 86, 88]:\n",
    "    dist_table = pd.concat([dist_table, calculateMagnitudeDistance([\"TOT_SALES\", \"nCustomers\", \"nTxnPerCust\", \"nChipsPerTxn\", \"avgPricePerUnit\"], trial_num)])\n",
    "    \n",
    "dist_table.head(8)\n",
    "dist_table"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "v3lZRuueCpFj"
   },
   "source": [
    "We'll select control stores based on how similar monthly total sales in dollar amounts and monthly number of customers are to the trial stores by using correlation and magnitude distance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "Vv6Evktt9g0f"
   },
   "outputs": [],
   "source": [
    "def combine_corr_dist(metricCol, storeComparison, inputTable=pretrial_full_observ):\n",
    "    corrs = calcCorrTable(metricCol, storeComparison, inputTable)\n",
    "    dists = calculateMagnitudeDistance(metricCol, storeComparison, inputTable)\n",
    "    dists = dists.drop(metricCol, axis=1)\n",
    "    combine = pd.merge(corrs, dists, on=[\"YEARMONTH\", \"Trial_Str\", \"Ctrl_Str\"])\n",
    "    return combine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "YmUHwFop9gs5"
   },
   "outputs": [],
   "source": [
    "compare_metrics_table1 = pd.DataFrame()\n",
    "for trial_num in [77, 86, 88]:\n",
    "    compare_metrics_table1 = pd.concat([compare_metrics_table1, combine_corr_dist([\"TOT_SALES\"], trial_num)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "hEK2V9f4CkWp"
   },
   "outputs": [],
   "source": [
    "corr_weight = 0.5\n",
    "dist_weight = 1 - corr_weight"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Nxx4jY57C5H8"
   },
   "source": [
    "Determining the top five highest composite score for each trial based on Total sales"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "IPtYDk5_CkQb",
    "outputId": "7aae1088-8982-46c8-dacc-b0d6931a229f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     Trial_Str  Ctrl_Str  Corr_Score  magnitude  CompScore\n",
      "218         77       233         1.0   0.986477   0.993238\n",
      "239         77       255         1.0   0.979479   0.989739\n",
      "177         77       188         1.0   0.977663   0.988831\n",
      "49          77        53         1.0   0.976678   0.988339\n",
      "120         77       131         1.0   0.976267   0.988134 \n",
      "\n",
      "     Trial_Str  Ctrl_Str  Corr_Score  magnitude  CompScore\n",
      "356         86       109         1.0   0.966783   0.983391\n",
      "401         86       155         1.0   0.965876   0.982938\n",
      "464         86       222         1.0   0.962280   0.981140\n",
      "467         86       225         1.0   0.960512   0.980256\n",
      "471         86       229         1.0   0.951704   0.975852 \n",
      "\n",
      "     Trial_Str  Ctrl_Str  Corr_Score  magnitude  CompScore\n",
      "551         88        40         1.0   0.941165   0.970582\n",
      "538         88        26         1.0   0.904377   0.952189\n",
      "582         88        72         1.0   0.903800   0.951900\n",
      "517         88         4         1.0   0.903466   0.951733\n",
      "568         88        58         1.0   0.891678   0.945839 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "grouped_comparison_table1 = compare_metrics_table1.groupby([\"Trial_Str\", \"Ctrl_Str\"]).mean().reset_index()\n",
    "grouped_comparison_table1[\"CompScore\"] = (corr_weight * grouped_comparison_table1[\"Corr_Score\"]) + (dist_weight * grouped_comparison_table1[\"magnitude\"])\n",
    "for trial_num in compare_metrics_table1[\"Trial_Str\"].unique():\n",
    "    print(grouped_comparison_table1[grouped_comparison_table1[\"Trial_Str\"] == trial_num].sort_values(ascending=False, by=\"CompScore\").head(), '\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "VNxeJb1LCkNj"
   },
   "outputs": [],
   "source": [
    "compare_metrics_table2 = pd.DataFrame()\n",
    "for trial_num in [77, 86, 88]:\n",
    "    compare_metrics_table2 = pd.concat([compare_metrics_table2, combine_corr_dist([\"nCustomers\"], trial_num)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ihvwvmX5DRtN"
   },
   "source": [
    "Determining the top five highest composite score for each trial based on no. of customers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "OcpDDfx4CkKw",
    "outputId": "ea93df2a-914f-463f-8999-97a6bc266a86"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     Trial_Str  Ctrl_Str  Corr_Score  magnitude  CompScore\n",
      "218         77       233         1.0   0.993132   0.996566\n",
      "38          77        41         1.0   0.976648   0.988324\n",
      "101         77       111         1.0   0.968407   0.984203\n",
      "105         77       115         1.0   0.967033   0.983516\n",
      "15          77        17         1.0   0.965659   0.982830 \n",
      "\n",
      "     Trial_Str  Ctrl_Str  Corr_Score  magnitude  CompScore\n",
      "401         86       155         1.0   0.986772   0.993386\n",
      "467         86       225         1.0   0.969577   0.984788\n",
      "356         86       109         1.0   0.969577   0.984788\n",
      "471         86       229         1.0   0.964286   0.982143\n",
      "293         86        39         1.0   0.961640   0.980820 \n",
      "\n",
      "     Trial_Str  Ctrl_Str  Corr_Score  magnitude  CompScore\n",
      "736         88       237         1.0   0.987818   0.993909\n",
      "705         88       203         1.0   0.944629   0.972315\n",
      "551         88        40         1.0   0.942414   0.971207\n",
      "668         88       165         1.0   0.935770   0.967885\n",
      "701         88       199         1.0   0.932447   0.966224 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "grouped_comparison_table2 = compare_metrics_table2.groupby([\"Trial_Str\", \"Ctrl_Str\"]).mean().reset_index()\n",
    "grouped_comparison_table2[\"CompScore\"] = (corr_weight * grouped_comparison_table2[\"Corr_Score\"]) + (dist_weight * grouped_comparison_table2[\"magnitude\"])\n",
    "for trial_num in compare_metrics_table2[\"Trial_Str\"].unique():\n",
    "    print(grouped_comparison_table2[grouped_comparison_table2[\"Trial_Str\"] == trial_num].sort_values(ascending=False, by=\"CompScore\").head(), '\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "1eVqkuZ9CkHz",
    "outputId": "e32f63f9-5198-41eb-cdea-1d97b4dbd293"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial_Str  Ctrl_Str\n",
      "77         233         0.994902\n",
      "           41          0.986020\n",
      "           46          0.984762\n",
      "dtype: float64 \n",
      "\n",
      "Trial_Str  Ctrl_Str\n",
      "86         155         0.988162\n",
      "           109         0.984090\n",
      "           225         0.982522\n",
      "dtype: float64 \n",
      "\n",
      "Trial_Str  Ctrl_Str\n",
      "88         40          0.970895\n",
      "           26          0.958929\n",
      "           72          0.954079\n",
      "dtype: float64 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "for trial_num in compare_metrics_table2[\"Trial_Str\"].unique():\n",
    "    a = grouped_comparison_table1[grouped_comparison_table1[\"Trial_Str\"] == trial_num].sort_values(ascending=False, by=\"CompScore\").set_index([\"Trial_Str\", \"Ctrl_Str\"])[\"CompScore\"]\n",
    "    b = grouped_comparison_table2[grouped_comparison_table2[\"Trial_Str\"] == trial_num].sort_values(ascending=False, by=\"CompScore\").set_index([\"Trial_Str\", \"Ctrl_Str\"])[\"CompScore\"]\n",
    "    print((pd.concat([a,b], axis=1).sum(axis=1)/2).sort_values(ascending=False).head(3), '\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_2MhEdR-Dia-"
   },
   "source": [
    "Similarities based on total sales:\n",
    "1. Trial store 77: Store 233, 255, 188\n",
    "2. Trial store 86: Store 109, 155, 222\n",
    "3. Trial store 88: Store 40, 26, 72\n",
    "\n",
    "Similarities based on No. of Customers:\n",
    "1. Trial store 77: Store 233, 41, 111\n",
    "2. Trial store 86: Store 155, 225, 109\n",
    "3. Trial store 88: Store 237, 203, 40\n",
    "\n",
    "Final SImilarities based on Highest average of both features combined:\n",
    "1. Trial store 77: Store 233\n",
    "2. Trial store 86: Store 155\n",
    "3. Trial store 88: Store 40\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "ApgkSkh-CkFM",
    "outputId": "513894f3-7ccf-4c3f-8562-7cca8506c694"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbkAAAEyCAYAAABnI64zAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de5gcVZ3/8fcnGUgCiQmBIUASmCARAhgRIuK6siheCMplkasuJBhkeQSXBUWiqyuwLILrAqu7sBu5o8tF9CcgmBXF4KIbJCBCSLgEkpAEAgO5AAkCk3x/f9QZrDTTc5/pqerP63n6me5zTld9+0x3fftUna5SRGBmZlZGg2odgJmZWV9xkjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykqtC0ockPd7JttMl3dvXMZWdpJC0S63j6A5JcySdVOs4zGxTdZXkJL2au22U9Fru8WfzbSPifyNi115a7wxJj0l6RdLzku6UNCLVXSPp/N5YTydjGSLpP1McqyTdLmlsRZtjJS2UtE7SU5I+1F/xtUfSJyT9JvVjs6R7JB3aC8s9R9IPeiPGKssfJekqSStT7E9Impmr79fkLumfJD0iqUXSORV1B6TPRv6zMi1XP0fSn3J1nfoi2EYMn80t47XKdebaTU+xrk/9d7mkUanuP3PPeUPSm7nHP+9g/VU/kxV9EZLOrihvSuUNbSz3nIo4XpW0Jld/mKSHJL0s6UVJd0ua0J0+tM6pqyQXEcNbb8AzwCG5sh+2tmvrzdtdkv4KuAA4LiJGAJOAm3px+YO7+JTTgQ8Ak4EdgNXA93LL+xhwEXAiMALYH3i6V4LtAUlHAj8CrgPGAWOAfwQO6Yd1S1JPPiuXAMPJ/vcjgUOBRb0UW3feq4uArwB3VKl/Nv9ZiYhrK+pPy9V164tgRPww91mcWrlOAElfInsvnkXWb/sBOwF3Sdo8Ik7Jtb8AuCm3jKnV1t2Fz+Q0YBVwQhdf3k0V/dealHche/9+Kb2eCcB/ABu6uHzrioioyxuwBPhoun8AsBw4G1gJXN9alms/E3gKeAVYAPx1rm46cG+V9XwZ+GmVupOBN4E3gFeB21P5JGAOsAZ4FDg095xrgMuBO4F1wEfJktWPgWZgMfB37bzuy4Fv5x5/Eng89/h3wIxO9uE7gbuBl4AXgR8Coyr6+MvAw8Basg3J0Fz9WcBzwLPA54AAdmljPSL7UnJWO7EMAr4OLAVeINuYjEx1TWnZ09JyXgT+IdUdlPr/zfQ/+GMqnwP8M/Bb4DVgF+AvgPvTa7kf+Ivc+ucAJ1WJbT5weJW636TY1qX1H5PKP0+WjFYBtwE75J4TwKnAk8DiVPYp4KH0nvkdMLkT/78fAOdUlB1A7n3fxnOqvs4efBbftk7gHak/jq4oH57e55+rKD8H+EEn11f1M5lrsyXZZ/3Y9P6YkqtrfT81tPG8qnEARwIP9Wbf+daJ/3etA6jZC397kmsh+9Y4BBhW+cEDjiJLJoOAY9JGaftUN53qSe5DaSN5LvBBYEhF/TXA+bnHm6WN29eAzYGPpA/brrn2a9OyBgFbAA+QjWo2B3YmG3l9oko8U8g23Duk5/43cGmqG5w+0DNTDMuBfweGVVnWLsDHUp81km2wL63o49+ndY0GFgKnpLqDgOeBPdMG5b+pnuR2S3UT2vl/fi7FvDPZhvAnwPWprnWj9P30v30P8DowKdW/bcNEtjF/BtgDaCAbOa4Gjk+Pj0uPt861r5bkriD7snIiMLGN+k1ed/qfvwjsnfr2e8BvKtrflfp0GPBessT+/vQ/nJb6fki1/krLqZbk3kj/m8Vko9AtK/qlOcX3W+CAXvgsHsDbk9xBZJ/JthLJtcANFWVv+x+2s752P5OpzfFkX8AGA7cD38vVtb6fuprkdgb+lPr0w8Dwnvadbx3f6mp3ZQc2At+MiNcj4rXKyoj4UUQ8GxEbI+Imsm/R+3a00Ij4X+AIsg3WHcBLki5uZzfjfmQb6Qsj4o2IuBv4GdlGtdWtEfHbiNgIvBtojIjzUvunyTbmx1ZZ/pPAMmAF8DLZqPG8VDeGLMkeSbYh2ItsA/r1Kq9tUUTclfqsGbgY+KuKZt9N/baKbGOxVyo/Grg6IuZHxDqyjUM1W6e/z7XT5rPAxRHxdES8CnwVOLZid965EfFaRPwR+CNZsmvPNRHxaES0AB8HnoyI6yOiJSJuAB6jc7tLv0g2yj0NWCBpkaSqu9PSa7kqIh6MiNfTa/mApKZcm29FxKr0Xj0Z+K+IuC8iNkS2e/F1svdSVz1G9j/anizZ7kP2f211NtnGeiwwC7hd0ju7sZ6ObAO8mPq+0nOpvls6+ZmcRrbbcQPZF7BjJW3WyVUcLWlN7vbrtN6nyRL6WOBm4MV0TH54d1+LdcxJ7s+aI+JP1SolnZAOGK9JB5L3pJMftIj4eUQcQvbN+zCykV+1mXg7AMtSAmu1lOyD0WpZ7v5OwA75DxXZKHBMleX/B9noYGuyEdRPgNaD9K3J/XsR8VxEvEi2gTu4rQVJGiPpRkkrJL1MNjKo7JOVufvryRL4W6+z4jVW81L6u307bXaoWMZS/jwC6yiWavLxVS6/dR1j6UBKrBdExD5k/X4z8CNJo6s8ZZN1paT9Eu2/B75U8R4Yn5bTJRGxMiIWpC9zi8mO3X06V39fRLySvthcSzaaq/b+yE++2LGLobwIbFPlmOP2qb7b2vtMShpPNtJqPU5/KzCUbNd+Z9wcEaNytw/n1js3Io6OiEayL5L7A//Qk9di7XOS+7Oq1xyStBPZ6Og0st1To8iOs6hLK8g2HL8iO461Z5X1PguMr5josCPZyKutWJeRHZfJf6hGRESbGx6yb+nXpFHA62S7wvaVtE1ErCbbRZlffnvXYrog1b87It4B/A2d75PnyDbErdrbCD5O9jo/3U6bZ8k29vnltZDtdutItdeYL69cfus6VtAFEfEyWb9tSTbxoC2brEvSlmTJsb33wD9XvAe2SKPNngra304EVf7nsenki2e6uN7/IxuNHpEvTKOeqcCvuri8NlX5TB5P9ppvl7SSbPf/ULLRXa+JiPvJvmTu2VFb6z4nuc7ZkuzD3Awg6UQ6+cZMU4aPlbRVmqW3L9kuvbmpyfNku39a3Uc2yviKpM0kHUC2S+zGKqv4PfCKpLMlDZM0WNKekt5Xpf39wAmSRqbdL18gm9nW+s34auCLkraVtBVwBtnu0raMIJscsFbZzxDOar83NnEzMF3S7pK2AL5ZrWFEBHAm8A1JJ0p6h6RBkv5S0qzU7AbgDEkT0oawdbZdW7u7Kj0PNHUwg/JO4F2SPiOpQdIxwO5U75u3SPqGpPdJ2lzSULIZrmvIknfr+vPvgRuAEyXtJWlIei33RcSSKqv4PnCKpPen99iWkj5ZOSU+F89mKY5BQIOkoa276iR9WNJOaTnjgQvJRjKtP4X4RGrfoOxnN/sDszvqg66KiLVkx8y+J+mgFHMT2ftmOdnksG7pxGdyWlr3Xrnbp4GDJW2dW9SQ1Bett3a3p+n9+nlJ26bHu5HNtJ3b3vOsh/rjwN9AvNHG7MqK+k3KyGbarSLbTXIxcA9pogHtTzzZn+xb54tkE0ieAL6Sq5/In2fF/TSV7ZGWv5a3z+S8htxElVS2A9mGcSXZZIi5ra+tjXi2JtsN80Ja573Avrn6zYDLUt1K4LvkZkRWLGsPskkvr6bX8KWKPluSj4OKg/JkE1xW0sHsylz7g4D/TetrJpsE8clUN4hs8s2yVPcDYKtU10TFRAFyE0VSn9yb+u7Byvrcc/4yvd616e9ftrW8NuL+OtnI/+X0HprDpjMzTyEb2a4hzSZMZU+l9j8DxuXav62fUt/cn5bxHNnPLUZUieeatIz8bXqqO5NsxLg+9eV3W5dDNrnofrL38Rqy99nHeuGzeABVZnQCM1LfvUb2ZeC/Wv+vFe02eW91sL6qn0my45h/IjvOXfm8R8n25rS+nypvH01xtM7Uzd+2JftifHt6Ha+SfT4uAjbraR/6Vv2m9M8zMzMrHe+uNDOz0nKSM7PS0aanDcvfHq11bNa/vLvSzMxKq9fO0dgXttlmm2hqaqp1GGZmhfLAAw+8GNlv8eregE5yTU1NzJs3r9ZhmJkViqT2Tq5QV3xMzszMSstJzszMSstJzszMSstJzszMSstJzszMSstJzszMSstJzszMSstJzszMSqvDJCfpKkkvSJqfKxst6S5JT6a/W6VySfqupEWSHpa0d+4501L7JyX16sUHzczM2tKZM55cA/w7cF2ubCbwq4i4UNLM9Phssiv2Tky39wOXA++XNJrsophTyK679ICk2yK7ErXVq3NGdrH92r6Jw8xKq8ORXET8huzCjXmHAdem+9cCh+fKr4vMXGCUpO2BTwB3RcSqlNjuIrvIo5mZWZ/p7jG5MRHxXLq/EhiT7o8lu5pwq+WprFq5mZlZn+nxxJPIrtXTa9frkXSypHmS5jU3N/fWYs3MrA51N8k9n3ZDkv6+kMpXAONz7calsmrlbxMRsyJiSkRMaWz0lSLMzKz7upvkbgNaZ0hOA27NlZ+QZlnuB6xNuzX/B/i4pK3STMyPpzIzM7M+0+HsSkk3AAcA20haTjZL8kLgZkkzgKXA0an5ncDBwCJgPXAiQESskvRPwP2p3XkRUTmZxczMrFd1mOQi4rgqVQe20TaAU6ss5yrgqi5FZ2Zm1gM+44mZmZWWk5yZmZVWZ854YmZt8RlbzAY8j+TMzKy0nOTMzKy0vLvSLGmaeUeX2i8Z2keBmFmv8UjOzMxKy0nOzMxKy0nOzMxKy0nOzMxKq74nnvh3TmZmpVbfSc6sRLo8O/TCT/ZRJGYDh5Oc9RpPwTezgcbH5MzMrLSc5MzMrLS8u9KsXnnildUBj+TMzKy0SjWS88QHMzPL80jOzMxKy0nOzMxKq1S7K83MCsGTfvqNR3JmZlZaTnJmZlZaTnJmZlZaTnJmZlZannhiZsXjiRvWSR7JmZlZaXkkZ2Y1V/SzFRU9/jLzSM7MzErLSc7MzErLSc7MzErLSc7MzEqrR0lO0hmSHpU0X9INkoZKmiDpPkmLJN0kafPUdkh6vCjVN/XGCzAzM6um20lO0ljg74ApEbEnMBg4FrgIuCQidgFWAzPSU2YAq1P5JamdmZlZn+np7soGYJikBmAL4DngI8Atqf5a4PB0/7D0mFR/oCT1cP1mZmZVdTvJRcQK4DvAM2TJbS3wALAmIlpSs+XA2HR/LLAsPbcltd+6crmSTpY0T9K85ubm7oZnZmbWo92VW5GNziYAOwBbAgf1NKCImBURUyJiSmNjY08XZ2Zmdawnuys/CiyOiOaIeBP4CfBBYFTafQkwDliR7q8AxgOk+pHASz1Yv5mZWbt6kuSeAfaTtEU6tnYgsAD4NXBkajMNuDXdvy09JtXfHRHRg/WbmZm1q9vnroyI+yTdAjwItAB/AGYBdwA3Sjo/lV2ZnnIlcL2kRcAqspmY1hM+E7uZWbt6dILmiPgm8M2K4qeBfdto+yfgqJ6sz8zMrCt8xhMzMystJzkzMystX09uAPE1qczMepdHcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlo9SnKSRkm6RdJjkhZK+oCk0ZLukvRk+rtVaitJ35W0SNLDkvbunZdgZmbWtoYePv/fgNkRcaSkzYEtgK8Bv4qICyXNBGYCZwNTgYnp9n7g8vTXzMz62AMPPLBtQ0PDFcCelGcv3kZgfktLy0n77LPPC2016HaSkzQS2B+YDhARbwBvSDoMOCA1uxaYQ5bkDgOui4gA5qZR4PYR8Vx3YzAzs85paGi4YrvttpvU2Ni4etCgQVHreHrDxo0b1dzcvPvKlSuvAA5tq01PsvkEoBm4WtIfJF0haUtgTC5xrQTGpPtjgWW55y9PZZuQdLKkeZLmNTc39yA8MzPL2bOxsfHlsiQ4gEGDBkVjY+NastFp2216sPwGYG/g8oh4L7CObNfkW9KorUsdGhGzImJKRExpbGzsQXhmZpYzqEwJrlV6TVVzWU+S3HJgeUTclx7fQpb0npe0PUD627qfdAUwPvf8canMzMzqxLnnnrvtLrvsssfEiRP3OOSQQyasX79eGzdu5Itf/OLYpqamPXfeeec9zj///G17a33dPiYXESslLZO0a0Q8DhwILEi3acCF6e+t6Sm3AadJupFswslaH48zM6uNppl37NOby1ty4Scf6KjN4sWLN5s1a9aYxx9/fP7w4cPj4IMP3vmKK64YHREsX758s6eeemr+4MGDWbFiRU8nRb6lpwv6IvDDNLPyaeBEstHhzZJmAEuBo1PbO4GDgUXA+tTWzMzqyIYNG7Ru3bpBQ4YM2fDaa68NGjdu3Jvf/OY3x95www1PDx48GICxY8e29Nb6ejSNNCIeSsfPJkfE4RGxOiJeiogDI2JiRHw0IlalthERp0bEOyPi3RExr3degpmZFcGECRPePPXUU1dOmDBh8rbbbvueESNGbDjiiCNeXrZs2ZDrr79+qz333HPS/vvvP/GRRx4Z0lvrLMtvJczMbIBrbm4efMcdd4xatGjRIytXrnx4/fr1gy677LLRb7zxhoYOHRrz589fOGPGjObp06c39dY6neTMzKxf3H777e/YcccdX99hhx1ahgwZEocffvia3/3ud8PHjBnzxnHHHbca4Pjjj1/zxBNPDOutdTrJmZlZv2hqanrjwQcfHP7KK68M2rhxI3ffffeISZMm/Wnq1KlrZs+ePQLgzjvvHLHTTju93lvr7LUZLGZmZu35yEc+su6QQw5ZPXny5EkNDQ3sscce688888zmdevWDTryyCMnXHbZZWO22GKLjd///veX9NY6neTMzOpQZ6b894VLLrnk2UsuueTZfNmwYcM2zJkzZ1FfrM+7K83MrLSc5MzMrLSc5MzMrLSc5MzMrLSc5MzMrLSc5MzMrLSc5MzMrF8cddRRTaNHj37PxIkT92gtO/PMM3fYdtttJ++2226777bbbrvfdNNNIwEef/zxzYcOHbp3a/lnPvOZHbuzTv9OzsysHp0zslcvtcM5azv83d3nPve5F08//fQXTjzxxAn58lNOOeX588477/nK9uPHj3/9scceW9CTsDySMzOzfjF16tRXGxsbe+0yOp3hJGdmZjV15ZVXbvuud71r96OOOqqpubl5cGv58uXLN580adLu73vf+3adPXv28O4s20nOzMxq5owzznhh6dKljyxcuHDBdttt9+YXvvCF8QA77rjjm4sXL3544cKFCy6++OJl06dP33nVqlVdzllOcmZmVjPjx49vaWhoYPDgwZx22mnNDz300JYAw4YNi+22224DwIc+9KH1O+644+vz588f2tXlO8mZmVnNLF26dLPW+zfeeOOoXXfd9TWAZ599tqGlJTt8t2DBgs2XLFkyZNddd+3yJXg8u9LMzPrFIYccMmHu3LkjVq9e3TBmzJjJM2fOfPaee+4ZsWDBgmEA48aNe+Pqq69eCvCLX/xi+Pnnnz+2oaEhBg0aFJdeeunSMWPGbOjqOp3kzMzqUSem/Pe222+/fXFl2RlnnPFiW22nT5++Zvr06Wt6uk7vrjQzs9JykjMzs9JykjMzs9JykjMzqw8bN27cqFoH0dvSa9pYrd5JzsysPsxvbm4eWaZEt3HjRjU3N48E5ldr49mVZmZ1oKWl5aSVK1desXLlyj0pzwBnIzC/paXlpGoNnOTMzOrAPvvs8wJwaK3j6G9lyeZmZmZv4yRnZmal5SRnZmal5SRnZmal5SRnZmal1eMkJ2mwpD9I+ll6PEHSfZIWSbpJ0uapfEh6vCjVN/V03WZmZu3pjZHc6cDC3OOLgEsiYhdgNTAjlc8AVqfyS1I7MzOzPtOjJCdpHPBJ4Ir0WMBHgFtSk2uBw9P9w9JjUv2Bqb2ZmVmf6OlI7lLgK/z5vGFbA2sioiU9Xg6MTffHAssAUv3a1H4Tkk6WNE/SvObm5h6GZ2Zm9azbSU7Sp4AXIqJXL7wXEbMiYkpETGlsbOzNRZuZWZ3pyWm9PggcKulgYCjwDuDfgFGSGtJobRywIrVfAYwHlktqAEYCL/Vg/WZmZu3q9kguIr4aEeMiogk4Frg7Ij4L/Bo4MjWbBtya7t+WHpPq746I6O76zczMOtIXv5M7GzhT0iKyY25XpvIrga1T+ZnAzD5Yt5mZ2Vt65SoEETEHmJPuPw3s20abPwFH9cb6zMzMOsNnPDEzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9JykjMzs9LqdpKTNF7SryUtkPSopNNT+WhJd0l6Mv3dKpVL0nclLZL0sKS9e+tFmJmZtaUnI7kW4EsRsTuwH3CqpN2BmcCvImIi8Kv0GGAqMDHdTgYu78G6zczMOtTtJBcRz0XEg+n+K8BCYCxwGHBtanYtcHi6fxhwXWTmAqMkbd/tyM3MzDrQK8fkJDUB7wXuA8ZExHOpaiUwJt0fCyzLPW15Kqtc1smS5kma19zc3BvhmZlZnepxkpM0HPgx8PcR8XK+LiICiK4sLyJmRcSUiJjS2NjY0/DMzKyO9SjJSdqMLMH9MCJ+koqfb90Nmf6+kMpXAONzTx+XyszMzPpET2ZXCrgSWBgRF+eqbgOmpfvTgFtz5SekWZb7AWtzuzXNzMx6XUMPnvtB4HjgEUkPpbKvARcCN0uaASwFjk51dwIHA4uA9cCJPVi3mZlZh7qd5CLiXkBVqg9so30Ap3Z3fWZmZl3lM56YmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlpOcmZmVlp9XuSk3SQpMclLZI0s7/Xb2Zm9aNfk5ykwcB/AFOB3YHjJO3enzGYmVn96O+R3L7Aooh4OiLeAG4EDuvnGMzMrE4oIvpvZdKRwEERcVJ6fDzw/og4LdfmZODk9HBX4PE+DGkb4MU+XH5fc/y15fhrp8ixQ9/Hv1NENPbh8gujodYBVIqIWcCs/liXpHkRMaU/1tUXHH9tOf7aKXLsUPz4i6S/d1euAMbnHo9LZWZmZr2uv5Pc/cBESRMkbQ4cC9zWzzGYmVmd6NfdlRHRIuk04H+AwcBVEfFof8ZQoV92i/Yhx19bjr92ihw7FD/+wujXiSdmZmb9yWc8MTOz0nKSMzOz0nKSMzOz0nKSKzBJ29Q6BjOzgcxJriAkTZW0WNK9kt4r6VHgPknLJR1Y6/i6QtJWkt5R6zi6StIYSXun25hax9MbJA2vdQz1zP3f9+omyUlaJekKSQdKUq3j6YZvAQcDZwG/BGZExDuBjwH/UsvAOkPSDpKuk7SW7HRG8yU9I+kcSZvVOr72SNpL0lxgDvDtdLtH0lxJe9c0uJ5bUOsAOiLp3amvl0maJWmrXN3vaxlbLxjw/V90A+60Xn2oGXgIOA+4TtItwA0RMbe2YXXaxohYCCBpfWvcEbFQUhG+rPwAOC8iTpB0BPAh4OvAV8muTHFye0+usWuAv42I+/KFkvYDrgbeU4ugOkvSmdWqgCKMJC4HzgHmAicB90o6NCKeAgb0FyQoRf8XWhE2jr1lXUT8e0R8EPgA2enELpP0tKQLahxbZ6yR9LeSzgJWSzpD0lhJ04BXax1cJ2wdEXMAIuInwP4RsS4ivg7sX9PIOrZlZYIDSF80tqxBPF11AbAVMKLiNpxibANGRMTsiFgTEd8BTgNmpy8ZRfihb9H7v9DqaST31i7KiHiGtNtJ0m7AMTWLqvOmkY18Avg4cBzZmWOWAp+vYVyd1Szpb4BfA0cASwDSruOB/kH/uaQ7gOuAZalsPHACMLtmUXXeg8BPI+KBygpJJ9Ugni6TNDIi1gJExK8lfRr4MTC6tpF1SuH7v8jq5ownki6OiGq7DayPSdoR+A7ZxXIfAs6KiOckbQ0cEBE/rmmAHZA0lezah2NT0Qrgtoi4s3ZRdY6kXYGXIuJtl3aRNCYinq9BWJ0m6TPA05WHFtJ76hsRMaC/5KX+XxURzW3UDfj+L7q6SXJFJ6kBmAEczqYb2p+SnQP0zVrFZmY2UA303US9RlJDOqY1W9LD6fZzSacM9Nl9yfXAXsC5ZLMsD0739yKb1DGg5fr/5wXt/zZJKvSJdh1/35M0UtKFkh5Ls7xfkrQwlY2qdXxlVzcjOUk3AGuAa4HlqXgc2bGu0RExoI/LSXoiIt7V1bqBosj9L6nacR8Bf4yIcf0ZT1c5/tqS9D/A3cC1EbEylW1H9t4/MCI+Xsv4yq6eklzRk8Rc4F+BH0fExlQ2CDgKODMi3l/L+DpS5P6XtIFsgk/+95WRHo+NiM1rElgnOf7akvR4ROza1TrrHfU0u3KVpKNoO0msrmlknXMscBHZzx5Wk33AR5F9Qzy2loF1UpH7/2myb9zPVFZIWtZG+4HG8dfWUklfIRvJPQ/ZhBNgOn+erWt9pG6OyZElgiOB5yU9IelJ4Hmy6ewDPklExJKIOCYiGsl+57dfRDSmssW1jq8TipdrVQsAAAcbSURBVNz/l5L9zqkt3+7PQLrJ8dfWMcDWZGfJWZ2+pM4h+/nD0bUMrB7Uze7KvDRtnYh4qdaxdJakzcmSwYqI+FWaVv0XwEJgVpFmVxax/82smOomyRU9SUj6Idnu5S3IJnAMB34CHEj2f5xWw/A6VOT+z8X+bET8skixg+MfCCTtTLbXYjywAXgC+O+IeLmmgdWBekpyRU8SD0fE5PR7uRXADhGxIZ0x5I8RMbnGIbaryP1f5NjB8deapL8DDgHuIfvpzx/IXsdfA19oPd2d9Y16SnJFTxLzgb3JzpX4DLBTRKySNBT4Q0RMqmmAHShy/xc5dnD8tSbpEWCvFPMWwJ0RcUA6Y8utEfHeGodYavU0u3JQ2u2xJdk3wpHAKmAIBTiTOXAl8BgwGPgH4EeSngb2A26sZWCdVOT+L3Ls4PgHggay3ZRDSFceiIhninwihKKopyRX6CQREZdIuindf1bSdcBHge9HRBGuqVXk/i9y7OD4a+0K4H5J95FdYuoiAEmNZMna+lDd7K6E7MKd8FaSGEWWJJ4pSJKoStLwiBjwl9spcv8XOXZw/LUmaQ9gEjA/Ih6rdTz1pK6SXDVFSRLVSHomInasdRzdVeT+L3Ls4PhrrejxF0E97a5szwJgQCcJlfvqwgO+/9tR5NjB8dda0eMf8OomyZUgSVwA/AvQ0kbdgD9zTZH7v8ixg+OvtaLHX3R1k+QoeJKg+FcXLnL/Fzl2cPy1VvT4C61ujslJ+h3wxSpJYllEjK9BWJ2mgl9duMj9X+TYwfHXWtHjL7p6SnKFThJFV+T+L3Ls4PhrrejxF13dJLmikzQS+CpwOLAt2fW0XgBuBS6MiDU1DM/MbECqm/3BKv4l6G8mu+7aARExOiK2Bj6cym6uaWSdUOT+L3Ls4PhrrejxF13dJDkKniSApoi4KCJWthZExMqIuAjYqYZxdVaR+7/IsYPjr7Wix19odbO7UgW/BL2kXwC/pO2rC38sIj5aw/A6VOT+L3Ls4PhrrejxF109jeSWSvpKSgxAliQknU0xLkFfeXXhVRTr6sJF7v8ixw6Ov9aKHn+h1VOSK3SSiIjVwNXAacD4tNtjUkScDexb2+g6pcj9X+TYwfHXWtHjL7S62V0JIGk3YBwwN3++OEkHRcTs2kXWMWUXXjyV7GrIewGnR8Stqe7BiNi7lvF1RsH7v7Cxg+OvtaLHX2R1M5JLSeJWspHQfEmH5aovqE1UXfJ5YJ+IOBw4APiGpNNTnWoWVScVuf+LHDs4/lorevxFV0+n9WpNEq9KagJukdQUEf9GAZIEMKj1G2BELJF0ANlr2IlixF/k/i9y7OD4a63o8RdaPSW5oieJ5yXtFREPAaQPzKeAq4B31za0Tily/xc5dnD8tVb0+AutbnZXkpJE64P0pvsUsA3FSBInACvzBRHREhEnAPvXJqQuKXL/Fzl2cPy1VvT4C61uJp5IGge05H9Mnav7YET8tgZh1Y0i93+RYwfHX2tFj7/o6ibJmZlZ/amn3ZVmZlZnnOTMzKy0nOSssJS5V9LUXNlRkmZL2iDpodxtZq7NNpLelHRKxfKWSHpE0sOS7kmz31rrQtIPco8bJDVL+lmu7PD03IVpOYfn6q6RtELSkFwMSyS9OxfjKkmL0/1fSmqSNL8ixnMkfbm3+tCs7JzkrLAiO6B8CnCxpKGShpP9uPZU4LWI2Ct3uzD31KOAucBxbSz2wxExmey0S1/Pla8D9pQ0LD3+GLCitVLSe4DvAIdFxCTgUOA7kibnlrEB+FzFa3ikNUbgNuCs9HhAn3DbrCic5KzQImI+cDtwNvCPwHUR8VQHTzsO+BIwNs18a8v/AWMryu4EPplbxg25ui8DF0TE4hTXYuBbwFm5NpcCZ0iqp9+nmtWUk5yVwbnAZ4CpwLdT2bCK3ZXHAEgaD2wfEb8nu5bXMVWWeRDw04qyG4FjJQ0FJgP35er2AB6oaD8vlbd6BrgXOL4Lr+2d+ddBNnI1s07yN0orvIhYJ+km4NWIeD0Vv5Z2AVY6hj9fqPJGsjPG/Guu/teSRgOvAt+oWM/D6bRMx5GN6rrjW2TnMbyjk+2fyr8OSed0c71mdckjOSuLjenWkeOA6ZKWkB0DmyxpYq7+w2RXWn+IbIRY6TayY283VJQvAPapKNsHeDRfEBFPpmX7Eitm/cBJzuqGpHcBwyNibEQ0RUQT2chqkwkoEdEC/D1wQhrV5V0FnBsRj1SUfwf4ahrpkf5+jU1Hia3+mewYnpn1Me+utLIalo5htZoNvAb8v4p2PwZuAs7LF0bEc5JuIJup+U+58uXAdytXFhEPKbvS8+2SNgPeBL7SekLtiraPSnoQGPDXADQrOp/Wy8zMSsu7K83MrLSc5MzMrLSc5MzMrLSc5MzMrLSc5MzMrLSc5MzMrLSc5MzMrLT+P4SwJiOvu2Y2AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "trial_control_dic = {77:233, 86:155, 88:40}\n",
    "for key, val in trial_control_dic.items():\n",
    "    pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"].isin([key, val])].groupby(\n",
    "        [\"YEARMONTH\", \"STORE_NBR\"]).sum()[\"TOT_SALES\"].unstack().plot.bar()\n",
    "    plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))\n",
    "    plt.title(\"Trial Store \"+str(key)+\" and Control Store \"+str(val)+\" - TOT_SALES\")\n",
    "    plt.show()\n",
    "    pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"].isin([key, val])].groupby(\n",
    "    [\"YEARMONTH\", \"STORE_NBR\"]).sum()[\"nCustomers\"].unstack().plot.bar()\n",
    "    plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))\n",
    "    plt.title(\"Trial Store \"+str(key)+\" and Control Store \"+str(val)+\" - nCustomer\")\n",
    "    plt.show()\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5hTbfRGyEeeT"
   },
   "source": [
    "Next we'll compare the performance of Trial stores to Control stores during the trial period. To ensure their performance is comparable during Trial period, we need to scale (multiply to ratio of trial / control) all of Control stores' performance to Trial store's performance during pre-trial. Starting with TOT_SALES."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "y3GrRtu6Ebkn"
   },
   "outputs": [],
   "source": [
    "#Ratio of Store 77 and its Control store.\n",
    "sales_ratio_77 = pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 77][\"TOT_SALES\"].sum() / pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 233][\"TOT_SALES\"].sum()\n",
    "\n",
    "#Ratio of Store 86 and its Control store.\n",
    "sales_ratio_86 = pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 86][\"TOT_SALES\"].sum() / pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 155][\"TOT_SALES\"].sum()\n",
    "\n",
    "#Ratio of Store 77 and its Control store.\n",
    "sales_ratio_88 = pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 88][\"TOT_SALES\"].sum() / pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 40][\"TOT_SALES\"].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 935
    },
    "id": "ek6lRxkdEbY2",
    "outputId": "21f48f3f-9d47-4c3a-f667-365eb31a2823"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "trial_full_observ = full_observ[(full_observ[\"YEARMONTH\"] >= 201902) & (full_observ[\"YEARMONTH\"] <= 201904)]\n",
    "scaled_sales_control_stores = full_observ[full_observ[\"STORE_NBR\"].isin([233, 155, 40])][[\"STORE_NBR\", \"YEARMONTH\", \"TOT_SALES\"]]\n",
    "\n",
    "def scaler(row):\n",
    "    if row[\"STORE_NBR\"] == 233:\n",
    "        return row[\"TOT_SALES\"] * sales_ratio_77\n",
    "    elif row[\"STORE_NBR\"] == 155:\n",
    "        return row[\"TOT_SALES\"] * sales_ratio_86\n",
    "    elif row[\"STORE_NBR\"] == 40:\n",
    "        return row[\"TOT_SALES\"] * sales_ratio_88\n",
    "\n",
    "scaled_sales_control_stores[\"ScaledSales\"] = scaled_sales_control_stores.apply(lambda row: scaler(row), axis=1)\n",
    "\n",
    "trial_scaled_sales_control_stores = scaled_sales_control_stores[(scaled_sales_control_stores[\"YEARMONTH\"] >= 201902) & (scaled_sales_control_stores[\"YEARMONTH\"] <= 201904)]\n",
    "pretrial_scaled_sales_control_stores = scaled_sales_control_stores[scaled_sales_control_stores[\"YEARMONTH\"] < 201902]\n",
    "\n",
    "percentage_diff = {}\n",
    "\n",
    "for trial, control in trial_control_dic.items():\n",
    "    a = trial_scaled_sales_control_stores[trial_scaled_sales_control_stores[\"STORE_NBR\"] == control]\n",
    "    b = trial_full_observ[trial_full_observ[\"STORE_NBR\"] == trial][[\"STORE_NBR\", \"YEARMONTH\", \"TOT_SALES\"]]\n",
    "    percentage_diff[trial] = b[\"TOT_SALES\"].sum() / a[\"ScaledSales\"].sum()\n",
    "    b[[\"YEARMONTH\", \"TOT_SALES\"]].merge(a[[\"YEARMONTH\", \"ScaledSales\"]],on=\"YEARMONTH\").set_index(\"YEARMONTH\").rename(columns={\"ScaledSales\":\"Scaled_Control_Sales\", \"TOT_SALES\":\"Trial_Sales\"}).plot.bar()\n",
    "    plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))\n",
    "    plt.title(\"Trial Store \"+str(trial)+\" and Control Store \"+str(control))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6Iimmo7qEbVi",
    "outputId": "56defdaa-eb61-4331-f628-37dfd8b6b064"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{77: 1.2615468650086274, 86: 1.13150143573637, 88: 1.0434583458542188}"
      ]
     },
     "execution_count": 26,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "percentage_diff"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 316
    },
    "id": "VWWbBZ3QEbSP",
    "outputId": "698e0ba6-7ae5-4b02-cfa0-2953a46ee940"
   },
   "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>c_STORE_NBR</th>\n",
       "      <th>YEARMONTH</th>\n",
       "      <th>c_ScaledSales</th>\n",
       "      <th>t_STORE_NBR</th>\n",
       "      <th>t_TOT_SALES</th>\n",
       "      <th>Sales_Percentage_Diff</th>\n",
       "      <th>trial_period</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>233</td>\n",
       "      <td>201902</td>\n",
       "      <td>249.762622</td>\n",
       "      <td>77</td>\n",
       "      <td>235.0</td>\n",
       "      <td>-0.060907</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>233</td>\n",
       "      <td>201903</td>\n",
       "      <td>203.802205</td>\n",
       "      <td>77</td>\n",
       "      <td>278.5</td>\n",
       "      <td>0.309755</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>233</td>\n",
       "      <td>201904</td>\n",
       "      <td>162.345704</td>\n",
       "      <td>77</td>\n",
       "      <td>263.5</td>\n",
       "      <td>0.475075</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>155</td>\n",
       "      <td>201902</td>\n",
       "      <td>864.522060</td>\n",
       "      <td>86</td>\n",
       "      <td>913.2</td>\n",
       "      <td>0.054764</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>155</td>\n",
       "      <td>201903</td>\n",
       "      <td>780.320405</td>\n",
       "      <td>86</td>\n",
       "      <td>1026.8</td>\n",
       "      <td>0.272787</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>155</td>\n",
       "      <td>201904</td>\n",
       "      <td>819.317024</td>\n",
       "      <td>86</td>\n",
       "      <td>848.2</td>\n",
       "      <td>0.034642</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>40</td>\n",
       "      <td>201902</td>\n",
       "      <td>1434.399269</td>\n",
       "      <td>88</td>\n",
       "      <td>1370.2</td>\n",
       "      <td>-0.045781</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>40</td>\n",
       "      <td>201903</td>\n",
       "      <td>1352.064709</td>\n",
       "      <td>88</td>\n",
       "      <td>1477.2</td>\n",
       "      <td>0.088458</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>40</td>\n",
       "      <td>201904</td>\n",
       "      <td>1321.797762</td>\n",
       "      <td>88</td>\n",
       "      <td>1439.4</td>\n",
       "      <td>0.085182</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    c_STORE_NBR  YEARMONTH  ...  Sales_Percentage_Diff  trial_period\n",
       "7           233     201902  ...              -0.060907         trial\n",
       "8           233     201903  ...               0.309755         trial\n",
       "9           233     201904  ...               0.475075         trial\n",
       "19          155     201902  ...               0.054764         trial\n",
       "20          155     201903  ...               0.272787         trial\n",
       "21          155     201904  ...               0.034642         trial\n",
       "31           40     201902  ...              -0.045781         trial\n",
       "32           40     201903  ...               0.088458         trial\n",
       "33           40     201904  ...               0.085182         trial\n",
       "\n",
       "[9 rows x 7 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp1 = scaled_sales_control_stores.sort_values(by=[\"STORE_NBR\", \"YEARMONTH\"], ascending=[False, True]).reset_index().drop([\"TOT_SALES\", \"index\"], axis=1)\n",
    "temp2 = full_observ[full_observ[\"STORE_NBR\"].isin([77,86,88])][[\"STORE_NBR\", \"YEARMONTH\", \"TOT_SALES\"]].reset_index().drop([\"index\", \"YEARMONTH\"], axis=1)\n",
    "scaledsales_vs_trial = pd.concat([temp1, temp2], axis=1)\n",
    "scaledsales_vs_trial.columns = [\"c_STORE_NBR\", \"YEARMONTH\", \"c_ScaledSales\", \"t_STORE_NBR\", \"t_TOT_SALES\"]\n",
    "scaledsales_vs_trial[\"Sales_Percentage_Diff\"] = (scaledsales_vs_trial[\"t_TOT_SALES\"] - scaledsales_vs_trial[\"c_ScaledSales\"]) / (((scaledsales_vs_trial[\"t_TOT_SALES\"] + scaledsales_vs_trial[\"c_ScaledSales\"])/2))\n",
    "def label_period(cell):\n",
    "    if cell < 201902:\n",
    "        return \"pre\"\n",
    "    elif cell > 201904:\n",
    "        return \"post\"\n",
    "    else:\n",
    "        return \"trial\"\n",
    "scaledsales_vs_trial[\"trial_period\"] = scaledsales_vs_trial[\"YEARMONTH\"].apply(lambda cell: label_period(cell))\n",
    "scaledsales_vs_trial[scaledsales_vs_trial[\"trial_period\"] == \"trial\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LhuyCIMiFZCD"
   },
   "source": [
    "Check significance of Trial minus Control stores TOT_SALES Percentage Difference Pre-Trial vs Trial.\n",
    "\n",
    "Step 1: Check null hypothesis of 0 difference between control store's Pre-Trial and Trial period performance.\n",
    "\n",
    "Step 2: Proof control and trial stores are similar statistically\n",
    "\n",
    "Check p-value of control store's Pre-Trial vs Trial store's Pre-Trial.\n",
    "If <5%, it is significantly different. If >5%, it is not significantly different (similar).\n",
    "\n",
    "Step 3: After checking Null Hypothesis of first 2 step to be true, we can check Null Hypothesis of Percentage Difference between Trial and Control stores during pre-trial is the same as during trial.\n",
    "\n",
    "Check T-Value of Percentage Difference of each Trial month (Feb, March, April 2019).\n",
    "Mean is mean of Percentage Difference during pre-trial.\n",
    "Standard deviation is stdev of Percentage Difference during pre-trial.\n",
    "Formula is Trial month's Percentage Difference minus Mean, divided by Standard deviation.\n",
    "Compare each T-Value with 95% percentage significance critical t-value of 6 degrees of freedom (7 months of sample - 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "-C4BolQEEbPA",
    "outputId": "2080adaf-b19e-4d08-c6c5-b9e2bd22c1e9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Store 40\n",
      "Ttest_indResult(statistic=-0.5958372343168585, pvalue=0.5722861621434009) \n",
      "\n",
      "Store 155\n",
      "Ttest_indResult(statistic=1.429195687929098, pvalue=0.19727058651603258) \n",
      "\n",
      "Store 233\n",
      "Ttest_indResult(statistic=1.1911026010974504, pvalue=0.29445006064862156) \n",
      "\n",
      "Critical t-value for 95% confidence interval:\n",
      "[-4.30265273  4.30265273]\n"
     ]
    }
   ],
   "source": [
    "from scipy.stats import ttest_ind, t\n",
    "\n",
    "# Step 1\n",
    "for num in [40, 155, 233]:\n",
    "    print(\"Store\", num)\n",
    "    print(ttest_ind(pretrial_scaled_sales_control_stores[pretrial_scaled_sales_control_stores[\"STORE_NBR\"] == num][\"ScaledSales\"],\n",
    "                   trial_scaled_sales_control_stores[trial_scaled_sales_control_stores[\"STORE_NBR\"] == num][\"ScaledSales\"],\n",
    "                   equal_var=False), '\\n')\n",
    "    #print(len(pretrial_scaled_sales_control_stores[pretrial_scaled_sales_control_stores[\"STORE_NBR\"] == num][\"ScaledSales\"]), len(trial_scaled_sales_control_stores[trial_scaled_sales_control_stores[\"STORE_NBR\"] == num][\"ScaledSales\"]))\n",
    "    \n",
    "alpha = 0.05\n",
    "print(\"Critical t-value for 95% confidence interval:\")\n",
    "print(t.ppf((alpha/2, 1-alpha/2), df=min([len(pretrial_scaled_sales_control_stores[pretrial_scaled_sales_control_stores[\"STORE_NBR\"] == num]),\n",
    "                       len(trial_scaled_sales_control_stores[trial_scaled_sales_control_stores[\"STORE_NBR\"] == num])])-1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "id": "EF9azpK6EbLm"
   },
   "outputs": [],
   "source": [
    "\n",
    "a = pretrial_scaled_sales_control_stores[pretrial_scaled_sales_control_stores[\"STORE_NBR\"] == 40][\"ScaledSales\"]\n",
    "b = trial_scaled_sales_control_stores[trial_scaled_sales_control_stores[\"STORE_NBR\"] == 40][\"ScaledSales\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "p9r3iolZFlzG"
   },
   "source": [
    "\n",
    "Null hypothesis is true. There isn't any statistically significant difference between control store's scaled Pre-Trial and Trial period sales."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "xfJsTQ1bEa_2",
    "outputId": "cd8f69ad-9540-49dc-b69e-c38498f47fa9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial store: 77 , Control store: 233\n",
      "Ttest_indResult(statistic=-1.2533353315065926e-15, pvalue=0.999999999999999) \n",
      "\n",
      "Trial store: 86 , Control store: 155\n",
      "Ttest_indResult(statistic=0.0, pvalue=1.0) \n",
      "\n",
      "Trial store: 88 , Control store: 40\n",
      "Ttest_indResult(statistic=0.0, pvalue=1.0) \n",
      "\n",
      "Critical t-value for 95% confidence interval:\n",
      "[-2.44691185  2.44691185]\n"
     ]
    }
   ],
   "source": [
    "# Step 2\n",
    "for trial, cont in trial_control_dic.items():\n",
    "    print(\"Trial store:\", trial, \", Control store:\", cont)\n",
    "    print(ttest_ind(pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == trial][\"TOT_SALES\"],\n",
    "                   pretrial_scaled_sales_control_stores[pretrial_scaled_sales_control_stores[\"STORE_NBR\"] == cont][\"ScaledSales\"],\n",
    "                   equal_var=True), '\\n')\n",
    "    #print(len(pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == trial][\"TOT_SALES\"]),len(pretrial_scaled_sales_control_stores[pretrial_scaled_sales_control_stores[\"STORE_NBR\"] == cont][\"ScaledSales\"]))\n",
    "\n",
    "alpha = 0.05\n",
    "print(\"Critical t-value for 95% confidence interval:\")\n",
    "print(t.ppf((alpha/2, 1-alpha/2), df=len(pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == trial])-1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2Zu137QPFsbC"
   },
   "source": [
    "Null hypothesis is true. There isn't any statistically significant difference between Trial store's sales and Control store's scaled-sales performance during pre-trial."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "kigmTotcFnX1",
    "outputId": "b7edbd6e-71eb-437e-ebc7-ee39dfd2e3b8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial store: 77 , Control store: 233\n",
      "201902 : -0.7171038288055888\n",
      "201903 : 3.035317928855662\n",
      "201904 : 4.708944418758203\n",
      "\n",
      "\n",
      "Trial store: 86 , Control store: 155\n",
      "201902 : 1.4133618775921797\n",
      "201903 : 7.123063846042149\n",
      "201904 : 0.8863824572944162\n",
      "\n",
      "\n",
      "Trial store: 88 , Control store: 40\n",
      "201902 : -0.5481633746817604\n",
      "201903 : 1.0089992743637755\n",
      "201904 : 0.9710006270463645\n",
      "\n",
      "\n",
      "Critical t-value for 95% confidence interval:\n",
      "1.9431802803927816\n"
     ]
    }
   ],
   "source": [
    "# Step 3\n",
    "for trial, cont in trial_control_dic.items():\n",
    "    print(\"Trial store:\", trial, \", Control store:\", cont)\n",
    "    temp_pre = scaledsales_vs_trial[(scaledsales_vs_trial[\"c_STORE_NBR\"] == cont) & (scaledsales_vs_trial[\"trial_period\"]==\"pre\")]\n",
    "    std = temp_pre[\"Sales_Percentage_Diff\"].std()\n",
    "    mean = temp_pre[\"Sales_Percentage_Diff\"].mean()\n",
    "    #print(std, mean)\n",
    "    for t_month in scaledsales_vs_trial[scaledsales_vs_trial[\"trial_period\"] == \"trial\"][\"YEARMONTH\"].unique():\n",
    "        pdif = scaledsales_vs_trial[(scaledsales_vs_trial[\"YEARMONTH\"] == t_month) & (scaledsales_vs_trial[\"t_STORE_NBR\"] == trial)][\"Sales_Percentage_Diff\"]\n",
    "        print(t_month,\":\",(float(pdif)-mean)/std)\n",
    "    print('\\n')\n",
    "    \n",
    "print(\"Critical t-value for 95% confidence interval:\")\n",
    "conf_intv_95 = t.ppf(0.95, df=len(temp_pre)-1)\n",
    "print(conf_intv_95)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ipBQBBA7F1bM"
   },
   "source": [
    "\n",
    "There are 3 months' increase in performance that are statistically significant (Above the 95% confidence interval t-score):\n",
    "\n",
    "March and April trial months for trial store 77\n",
    "\n",
    "March trial months for trial store 86"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 935
    },
    "id": "w7CuojO8FnVU",
    "outputId": "93da568e-e31b-43e7-a9ad-a25922ea128a"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for trial, control in trial_control_dic.items():\n",
    "    a = trial_scaled_sales_control_stores[trial_scaled_sales_control_stores[\"STORE_NBR\"] == control].rename(columns={\"TOT_SALES\": \"control_TOT_SALES\"})\n",
    "    b = trial_full_observ[trial_full_observ[\"STORE_NBR\"] == trial][[\"STORE_NBR\", \"YEARMONTH\", \"TOT_SALES\"]].rename(columns={\"TOT_SALES\": \"trial_TOT_SALES\"})\n",
    "    comb = b[[\"YEARMONTH\", \"trial_TOT_SALES\"]].merge(a[[\"YEARMONTH\", \"control_TOT_SALES\"]],on=\"YEARMONTH\").set_index(\"YEARMONTH\")\n",
    "    comb.plot.bar()\n",
    "    cont_sc_sales = trial_scaled_sales_control_stores[trial_scaled_sales_control_stores[\"STORE_NBR\"] == control][\"TOT_SALES\"]\n",
    "    std = scaledsales_vs_trial[(scaledsales_vs_trial[\"c_STORE_NBR\"] == control) & (scaledsales_vs_trial[\"trial_period\"]==\"pre\")][\"Sales_Percentage_Diff\"].std()\n",
    "    thresh95 = cont_sc_sales.mean() + (cont_sc_sales.mean() * std * 2)\n",
    "    thresh5 = cont_sc_sales.mean() - (cont_sc_sales.mean() * std * 2)\n",
    "    plt.axhline(y=thresh95,linewidth=1, color='b', label=\"95% threshold\")\n",
    "    plt.axhline(y=thresh5,linewidth=1, color='r', label=\"5% threshold\")\n",
    "    plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))\n",
    "    plt.title(\"Trial Store \"+str(trial)+\" and Control Store \"+str(control)+\" - Total Sales\")\n",
    "    plt.savefig(\"TS {} and CS {} - TOT_SALES.png\".format(trial,control), bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "id": "236RaKOSFnSz"
   },
   "outputs": [],
   "source": [
    "#Ratio of Store 77 and its Control store.\n",
    "ncust_ratio_77 = pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 77][\"nCustomers\"].sum() / pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 233][\"nCustomers\"].sum()\n",
    "\n",
    "#Ratio of Store 86 and its Control store.\n",
    "ncust_ratio_86 = pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 86][\"nCustomers\"].sum() / pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 155][\"nCustomers\"].sum()\n",
    "\n",
    "#Ratio of Store 77 and its Control store.\n",
    "ncust_ratio_88 = pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 88][\"nCustomers\"].sum() / pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == 40][\"nCustomers\"].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 935
    },
    "id": "wutX9m6SFnPW",
    "outputId": "febf692b-10f7-4c4f-ba8d-a4d2a0e68d50"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAd0AAAEyCAYAAAC/Lwo5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de5xVdb3/8fcbhqsYOjgSgjqKMMMAGoIcw0zK7JSZ+vOCosdLaWa/tKwMzY63CtKMLPVkktfSQ5J1Uo9dfhpe0tQCiTsoXlCQyyAIoigM8/n9sdbodpxhhpk9a88eXs/HYx6z9/quvdZnr++eee/v2muv5YgQAABoe50KXQAAADsKQhcAgIwQugAAZITQBQAgI4QuAAAZIXQBAMgIodvO2T7U9uJmznum7cfbuqaOznbY3q/QdbSE7Udsn13oOgA0jNAtANsbc35qbW/KuX9q7rwR8beIqMjTes+yvcj2G7ZX2f6j7Z3Ttttt/yAf62lmLd1s/yKtY63t+233rzfPybYX2n7T9vO2D82qvm2x/e+2H0u3Y7XtR20fnYflXmH7znzU2Mjyd7F9q+2Vae3P2r44pz3TNxu2v297ru0a21fUaxub/m3k/q2ckdP+iO23c9qa9cYUKDRCtwAiolfdj6SXJX0+Z9pddfPZLsnXOm0fJmmSpPERsbOkIZLuzuPyO2/nQ74u6aOS9pe0h6R1kq7PWd4Rkq6W9AVJO0v6uKQX8lJsK9g+QdJvJf1K0gBJfSVdJunzGazbtlvzN3utpF5K+r63pKMlLclTbS15rS6RNEHSA420v5r7txIRd9RrPy+nLS9vTIG2Rui2I+m7+2W2L7K9UtJtddNy5rk4HfW9YXuB7f/TzMUfJOnJiJglSRGxNiLuiIg3bJ8j6VRJE9JRw/3puoakI4rXbc/PHc2lI+Mb09Hym5I+YXsP279LR38v2v7aNurZR9JfImJVRLyt5A3A0Jz2KyV9LyKeiojaiFgeEcsb2W4DbU+3/ZrtNbbvsr1LTvtLti+0Pcf2ett32+6e0/5t2ytsv2r7i40VbNuSfiLp+xFxc0SsT2t7NCK+lM7TyfZ/2l5qe7XtX9nunbaVp6PJM2y/nNb63bTtM5IukXRS2gez0+mP2J5o+wlJb0na1/YY2/9Mn8s/bY/ZxnbOdZCk/46IdWndiyLinnQ9j6XzzE7Xf1I6/Uu2l6R7I+6zvUfO9gjbX7X9nKTn0mlH2f5X+pr5u+39Gysmff39SdIbzawfKH4RwU8BfyS9JOlT6e2xkmqUjPC6SeqRTluWM/+JSkaGnSSdJOlNSf3StjMlPd7Ieg6VtElJmB0iqVu99tsl/SDnfhclI5FLJHWV9Ekl/xwrcuZfny6rk6SekmYqGfV1lbSvkpHpvzdSzyhJT6TPpaek/5b007Sts6TNki5Oa1gm6QZJPRpZ1n6Sjki3WZmkx+qWlbON/5Guq1TSQknnpm2fkbRK0jBJO6V1hKT9GlhPZdq2zzb684tpzfsqGVX+XtKv07by9PG/TPv2AEnvSBqStl8h6c56y3tEyd6QoZJKlIys10k6Lb0/Pr3fJ2f+sxup7WZJ85XsPRjUQPv7nnfa52skHZhu2+slPVZv/gfTbdpD0ghJqyX9W9qHZ6Tbvltj2ytdzp2Srqg3bWz6Glgl6UUlo/Sd6m2X6rS+JySNLfTfMj/8NOeHkW77Uyvp8oh4JyI21W+MiN9GxKuRjFTuVjLCGN3UQiPib5KOU/IP9AFJr9n+yTZ2Cx+sJDSuiojNETFd0v8q+Sdf596IeCIiaiUNl1QWEd9L539BSbic3Mjyn5P0iqTlkjYo2eX5vbStr5LQP0HJm4WPKPmH/p+NPLclEfFgus2qlYxGD6s323Xpdlsr6f50mZI0TtJtETEvIt5UEnyN6ZP+XrGNeU6V9JOIeCEiNkr6jqST6+1+vTIiNkXEbEmzlYTvttweEfMjokbSpyU9FxG/joiaiJgqaZGat3v7fEl3STpP0oJ0BPvZJp7LrRHxTES8kz6Xj9ouz5nnh5HsNdkk6RxJN0XE0xGxNZLdwe8oeS1tr0VK+qifkvAfqaRf61yk5I1Nf0lTJN1ve2AL1gNkitBtf6oj2d3aINun5+y+e13JCG235iw4Iv4UEZ9XMjI5RsnIuLEjXfeQ9EoaqHWWKvknV+eVnNt7S9qjrq60tkuUBGhD/kvJ6KmPkhHm7yX9KW2re7NxfUSsiIg1Sv7hHtnQgmz3tf0b28ttb1Aycqq/TVbm3H5LyRuKd59nvefYmNfS3/22Mc8e9ZaxVO+NUJuqpTG59dVfft06+qsJadBPioiRSrb7NEm/tV3ayEPet670TcRr2vZr4Fv1XgN7psvZLhGxMiIWpG8uX1Ty2e/xOe1PR8Qb6RutO5SMdht8fQDtCaHb/jR62SfbeysZPZ6nZHfiLpLmSfJ2rSD5R/ZXSdOVhHZD631V0p71DtzZS8nItKFaX5H0YkTskvOzc0Q09o/wI0pGcGvTUdT1kkbb3i0i1inZpZy7/G1dDmtS2j48Ij4k6T/U/G2yQkkw1NlrG/MuVvI8j9/GPK8qCZ/c5dUo2U3alMaeY+70+suvW0eDn3c3uqKIDUq2205KPl9vyPvWZXsnJWG9rdfAxHqvgZ7paLy1Qtv+fxXazr8DoBAI3eKyk5J/LtWSZPsLei80t8n2MU6+grNrehTsaCW7YJ9KZ1mlZHddnaeVjMIm2O5ie6ySXZi/aWQV/5D0RnoQWA/bnW0Ps31QI/P/U9Lptnvb7iLp/yo5WnVN2n6bpPNt7257V0nfULJ7uyE7S9ooab2Trx19e9tb432mSTrTdpXtnpIub2zGiAhJ35R0qe0v2P5QeuDUx2xPSWebKukbtvex3UtJsN2d7hpuyipJ5d72Ecp/lDTY9im2S9IDnqrU+LZ5l+1LbR9ku2t6INnXJb2u5M1E3fpzXwNTJX3B9kdsd0ufy9MR8VIjq/ilpHNt/1v6GtvJ9uecfi2tgXq6pHV0klRiu3vdxx22P2F773Q5e0q6StK9adsuTr621T3dBqcqObr9z01tA6DQCN0iEhELJE2W9KSSf5DDlexWa451kr6k5LPUul2w18R7X1G6RVJVulvwDxGxWUnIflbJwSo/l3R6RCxqpLatko5SMoJ9MX3MzUq+mtKQCyW9ndZTrWTXYO6R2N9XEszPKjnwaZakiY0s60oln1WvV/J59e8b3QofrPtPkn6qZNS/JP29rfnvUXIA2xeVjARXSfqB0kCQdKukXys5mOvF9Dme38xyfpv+fs32M42s/zUl2/lbSnb1TpB0VM6blW2Wr+TNzJq09iMkfS7dbSwln2ffkb4GxkXEQ5IulfQ7JXsEBqrxz+gVETOUvMZuUPJ6W6LkI4zG/FLJRwnjJX03vX1a2jZC0t+VHCj4d0lzJdUdDd9FyTavO5DqfEnHRsSzzdgGQEE5efMOAADaGiNdAAAyQugCAJARQhcAgIw0GbpOTpC+2va8nGnXODlx/hzb/+P3n3LvO+mX7hfb/ve2KhwAgGLT5IFUtj+u5OsYv4qIYem0T0uaHhE1tq+WpIi4yHaVkq8ZjFbyhfiHJA1Oj2xt1G677Rbl5eWtfS4AsEOZOXPmmogoK3QdaL4mrwwSEY/VO+2bIuL/5dx9Ssnp+qTkLEe/SU928KLtJUoC+MltraO8vFwzZszYjrIBALa3dQY1tEP5+Ez3i3rv9H399f7Twi1TI6ens32O7Rm2Z1RXV+ehDAAA2rdWha6Ty5LVKDmJ+naJiCkRMSoiRpWVsXcEANDxtfgi6bbPVHJmnMPjvQ+Gl+v957EdoO08JywAAB1Vi0a6Ti64PUHS0RHxVk7TfUouY9bN9j6SBik5Jy8AADu8Jke6tqcquaD0braXKTkh/HeUXJbtQduS9FREnBsR821Pk7RAyW7nrzZ15DIAADuKdnHu5VGjRgVHLwPA9rE9MyJGFboONB9npAIAICOELgAAGSF0AQDISIu/MgQUSvnFD2S6vpeu+lym6wPQcTHSBQAgI4QuAAAZIXQBAMgIoQsAQEYIXQAAMkLoAgCQEUIXAICMELoAAGSE0AUAICOELgAAGSF0AQDICKELAEBGCF0AADKyQ15liKvUAAAKgZEuAAAZIXQBAMgIoQsAQEZ2yM90ARQGx1NgR8dIFwCAjBC6AABkhNAFACAjhC4AABkhdAEAyEiToWv7Vturbc/LmVZq+0Hbz6W/d02n2/Z1tpfYnmP7wLYsHgCAYtKcke7tkj5Tb9rFkv4aEYMk/TW9L0mflTQo/TlH0o35KRMAgOLXZOhGxGOS1tabfIykO9Lbd0g6Nmf6ryLxlKRdbPfLV7EAABSzln6m2zciVqS3V0rqm97uL+mVnPmWpdM+wPY5tmfYnlFdXd3CMgAAKB6tPpAqIkJStOBxUyJiVESMKisra20ZAAC0ey09DeQq2/0iYkW6+3h1On25pD1z5huQTgOA7F3RO+P1rc92fSg6LR3p3ifpjPT2GZLuzZl+enoU88GS1ufshgYAYIfW5EjX9lRJYyXtZnuZpMslXSVpmu2zJC2VNC6d/Y+SjpS0RNJbkr7QBjUDAFCUmgzdiBjfSNPhDcwbkr7a2qIAAOiIOCMVAAAZIXQBAMgIoQsAQEZa+pUhbA++tgAAECNdAAAyw0gXaAp7KgDkCSNdAAAyQugCAJARQhcAgIwQugAAZITQBQAgI4QuAAAZIXQBAMgIoQsAQEYIXQAAMkLoAgCQEUIXAICMELoAAGSE0AUAICOELgAAGSF0AQDICKELAEBGCF0AADJC6AIAkBFCFwCAjBC6AABkhNAFACAjrQpd29+wPd/2PNtTbXe3vY/tp20vsX237a75KhYAgGLW4tC13V/S1ySNiohhkjpLOlnS1ZKujYj9JK2TdFY+CgUAoNi1dvdyiaQetksk9ZS0QtInJd2Ttt8h6dhWrgMAgA6hxaEbEcsl/VjSy0rCdr2kmZJej4iadLZlkvo39Hjb59ieYXtGdXV1S8sAAKBotGb38q6SjpG0j6Q9JO0k6TPNfXxETImIURExqqysrKVlAABQNFqze/lTkl6MiOqI2CLp95IOkbRLurtZkgZIWt7KGgEA6BBaE7ovSzrYdk/blnS4pAWSHpZ0QjrPGZLubV2JAAB0DK35TPdpJQdMPSNpbrqsKZIukvRN20sk9ZF0Sx7qBACg6JU0PUvjIuJySZfXm/yCpNGtWS4AAB0RZ6QCACAjhC4AABkhdAEAyAihCwBARghdAAAyQugCAJARQhcAgIwQugAAZITQBQAgI4QuAAAZIXQBAMgIoQsAQEYIXQAAMtKqqwwBANqXmTNn7l5SUnKzpGFiYFUItZLm1dTUnD1y5MjV9RsJXQDoQEpKSm7+8Ic/PKSsrGxdp06dotD17Ghqa2tdXV1dtXLlypslHV2/nXdBANCxDCsrK9tA4BZGp06doqysbL2SPQ0fbM+4HgBA2+pE4BZWuv0bzFdCFwCQmcsuu6zvPvvsM7SysrJq2LBhQ2644YY+27uMxYsXd/3FL35R2hb1tTU+0wWADqz84gdG5nN5L131uZktfeyPfvSjsunTp39o5syZC0tLS2vXrl3b6a677tp1e5fz3HPPdbv77rtLzz333LUtrWV7bdmyRV26dGn1chjpAgDyavHixV333XffoSeffPLe++2339BDDjlk0MaNG33ttdd+eMqUKUtLS0trJam0tLT2/PPPf02S+vfvP3zFihUlkvTYY4/1HD16dIUkPfDAA70qKyurKisrq4YMGVK1bt26Tt/97nf7z5gxo1dlZWXVlVdeuftbb73lE044oXzw4MFVQ4YMqbr//vt3lqTrrruuz6c+9amBY8aMGdS/f//hkyZNKrviiiv6DhkypOqAAw6oXLVqVWdJmj9/frdDDz100NChQ4eMHDmyYtasWd0l6fjjjy8/5ZRT9tp///0rv/KVrwxoqJbt3TaMdAEAeffyyy93v/POO18YM2bM0iOPPHLf22+/vfTNN9/sXFVVtXl7ljN58uQPX3fddUs//elPv7l+/fpOPXv2rJ04ceLyyZMn93344YeXSNLll1/e17aeffbZBbNmzep+5JFHDnr++efnSdKzzz7bY/bs2Qs2bdrUqaKiYtill166fOHChQvOOuusPW+66aY+l1122eqzzz577ylTpiwdPnz4O9OnT9/pK1/5yl5PPfXUs5K0YsWKrs8888yikpISffKTn9yvfi3bu10IXQBA3vXv3/+dMWPGbJKkESNGvPXSSy91bclyDj744I0XXnjhnuPGjVs7fvz4dQMHDvxA0P3973/vdf75569O1/X2HnvssXnu3LndJWnMmDFv7LrrrrW77rprba9evbaeeOKJr0vS8OHD35ozZ07P9evXd5o1a1avE088cWDd8jZv3uy628cdd9y6kpKSZtfSFHYvAwDyrmvXru8eQd25c+eora11z549axcsWNBg+KbzSJI2bdr0bjZNmjRp5c0337x006ZNnQ499NDKul2/LamjU6dO6t69e9Tdrqmp8datW7XzzjvXLFq0aEHdzwsvvDC/7jG9evV6N1hbW4tE6AIAMnLBBResOPfcc/deu3ZtJ0lav359p7qjlwcMGLD5iSee6ClJ06ZNe/fgqvnz53cbPXr0pokTJ67cf//935w3b1733r17b924cWPnunkOOeSQjXfeeWepJM2ZM6fbihUruu6///5vN6em0tLS2gEDBmy+9dZbd5Wk2tpaPfnkkz0amrehWrZ3GxC6AIBMTJgwofrjH//4hgMPPLBq0KBBQw8++ODKuu8UX3bZZa9OmDBhr2HDhg3p3Lnzu6PTH/3oR7sPGjRo6ODBg6u6dOkSJ5xwwvrRo0dv6ty5c1RUVFRdeeWVu0+YMGF1bW2tBw8eXHXSSScNvOmmm17q0aNHs7+rPHXq1Bduu+223SoqKqoGDRo09He/+90uDc3XUC3buw0cUfjvUI8aNSpmzJiR2frKL34gs3VJ0kvdT8l0fbpiu18HRYX+K170XX7ZnhkRo3KnzZ49+6UDDjhgTaaF4ANmz5692wEHHFBefzojXQAAMtKq0LW9i+17bC+yvdD2R22X2n7Q9nPp7+3+4jMAAB1Ra0e6P5P054iolHSApIWSLpb014gYJOmv6X0AAHZ4LQ5d270lfVzSLZIUEZsj4nVJx0i6I53tDknHtrZIAAA6gtaMdPeRVC3pNtuzbN9seydJfSNiRTrPSkl9G3qw7XNsz7A9o7q6uhVlAABQHFoTuiWSDpR0Y0SMkPSm6u1KjuTQ6AYPj46IKRExKiJGlZWVtaIMAACKQ2tCd5mkZRHxdHr/HiUhvMp2P0lKf69uXYkAgGJz0UUXfXi//fYbOnjw4KrKysqq6dOn77Q9j1+8eHHXQYMGDd2exxx//PHlt912266SNHr06Iphw4YNqWvLvYhCPrT08oItPvdyRKy0/YrtiohYLOlwSQvSnzMkXZX+vrel6wAAtNIVvfN6aT9dsb7JS/s99NBDO/3lL3/ZZe7cuQt69OgRK1asKHnnnXfc1OPy7bXXXiuZNm3ah8aNG7ch38tu6eUFW3v08vmS7rI9R9JHJE1SErZH2H5O0qfS+wCAHcTy5cu7lJaW1tSdFapfv3415eXlWx599NGeI0aMqKyoqKgaPnz4kHXr1nVavHhx15EjR1ZUVVUNqaqqGvLggw9+YERcU1OjL3/5ywOGDRs2ZPDgwVXXXHPNblJyysbTTz99r/Ly8mFjxowZvGbNmvcNJM8777xVV199db+GlnfOOecMqDu71MSJE3eXWnd5weZum1ZdZSgi/iVpVANNh7dmuQCA4nXsscdu+OEPf7hHeXn5sI997GMbxo8fv/bwww9/89RTTx141113PX/YYYe9tXbt2k69evWqLSkpqfnb3/72bM+ePWPu3Lndxo8fv++8efMW5i7vpz/96W69e/feOm/evIWbNm3yQQcdVPn5z39+w9NPP91zyZIl3ZYsWTJv2bJlXYYPHz70zDPPfK3ucYceeujG+++/f5f7779/5969e2+tmz558uSyl19+ueuCBQvmd+nSRXXX1W1Mcy4v2FyckQoAkFe9e/eunTdv3oIbbrhhaVlZWc0ZZ5wxcPLkyWW77777lsMOO+wtKbnQQJcuXbR582afcsop5YMHD6468cQTBz7//PMfuIjAQw899KFp06b1qaysrBoxYsSQdevWlSxYsKD7o48+uvO4cePWlpSUqLy8fMtHP/rRN+o/9pJLLlkxadKk9412p0+f/qEvf/nLa7p06SJJ6tu379b6j8tVd0m/H/zgB7uvWbOmc93jWoLQBQDkXUlJiY466qg3rr322levueaal++7774GLyIwceLEvrvvvvuWhQsXLpg7d+6CLVu2fCCXIsKTJ09+ue7Se8uXL5973HHHNetz2qOPPvqNt99+u9Pjjz/e5IFcbXV5wVyELgAgr2bPnt1t7ty53eruz5o1q8egQYPeXr16dZdHH320pyStW7eu05YtW7R+/frO/fr129K5c2f9/Oc/77N16wcHnUccccT6G2+8sazuYKw5c+Z027BhQ6fDDjvsjXvuuae0pqZGS5cu7fLUU0/t3FA9F1988Yrrr7/+w3X3Dz/88A033XTTblu2bJGkd3cvt+bygs1F6AIA8mrDhg2dTz/99H0GDhw4dPDgwVWLFi3qcc0117x61113Pf+1r31tr4qKiqqxY8cOfuuttzpdcMEFq6dOndqnoqKiatGiRd179OhRW3953/jGN9ZUVla+PXz48CGDBg0a+qUvfWnvLVu2+LTTTnt93333fWe//fYbNn78+PIRI0ZsbKiek046aX1paWlNzvKqBwwYsLmysnJoRUVF1S233FIqte7ygs3dNlzaLwMd/fJiWaP/ihd9l19c2q/94tJ+AAAUGKELAEBGCF0AADJC6AJAx1JbW1ub+SkX8Z50+3/ggDCJ0AWAjmZedXV1b4K3MGpra11dXd1b0ryG2lt1GkgAQPtSU1Nz9sqVK29euXLlMDGwKoRaSfNqamrObqiR0AWADmTkyJGrJR1d6DrQMN4FAQCQEUIXAICMELoAAGSE0AUAICOELgAAGSF0AQDICKELAEBGCF0AADJC6AIAkBFCFwCAjBC6AABkhNAFACAjhC4AABkhdAEAyAihCwBARlodurY7255l+3/T+/vYftr2Ett32+7a+jIBACh++Rjpfl3Swpz7V0u6NiL2k7RO0ll5WAcAAEWvVaFre4Ckz0m6Ob1vSZ+UdE86yx2Sjm3NOgAA6ChaO9L9qaQJkmrT+30kvR4RNen9ZZL6N/RA2+fYnmF7RnV1dSvLAACg/Wtx6No+StLqiJjZksdHxJSIGBURo8rKylpaBgAARaOkFY89RNLRto+U1F3ShyT9TNIutkvS0e4ASctbXyYAAMWvxSPdiPhORAyIiHJJJ0uaHhGnSnpY0gnpbGdIurfVVQIA0AG0xfd0L5L0TdtLlHzGe0sbrAMAgKLTmt3L74qIRyQ9kt5+QdLofCwXAICOhDNSAQCQEUIXAICMELoAAGSE0AUAICOELgAAGSF0AQDICKELAEBGCF0AADJC6AIAkBFCFwCAjBC6AABkhNAFACAjhC4AABkhdAEAyAihCwBARghdAAAyQugCAJARQhcAgIwQugAAZITQBQAgI4QuAAAZIXQBAMgIoQsAQEYIXQAAMkLoAgCQEUIXAICMELoAAGSkxaFre0/bD9teYHu+7a+n00ttP2j7ufT3rvkrFwCA4tWakW6NpG9FRJWkgyV91XaVpIsl/TUiBkn6a3ofAIAdXotDNyJWRMQz6e03JC2U1F/SMZLuSGe7Q9KxrS0SAICOIC+f6doulzRC0tOS+kbEirRppaS+jTzmHNszbM+orq7ORxkAALRrrQ5d270k/U7SBRGxIbctIkJSNPS4iJgSEaMiYlRZWVlrywAAoN1rVeja7qIkcO+KiN+nk1fZ7pe295O0unUlAgDQMbTm6GVLukXSwoj4SU7TfZLOSG+fIenelpcHAEDHUdKKxx4i6TRJc23/K512iaSrJE2zfZakpZLGta5EAAA6hhaHbkQ8LsmNNB/e0uUCANBRcUYqAAAyQugCAJARQhcAgIwQugAAZITQBQAgI4QuAAAZIXQBAMgIoQsAQEYIXQAAMkLoAgCQEUIXAICMELoAAGSE0AUAICOELgAAGSF0AQDICKELAEBGCF0AADJC6AIAkBFCFwCAjBC6AABkhNAFACAjhC4AABkhdAEAyAihCwBARghdAAAyQugCAJARQhcAgIy0Weja/oztxbaX2L64rdYDAECxaJPQtd1Z0n9J+qykKknjbVe1xboAACgWbTXSHS1pSUS8EBGbJf1G0jFttC4AAIpCSRstt7+kV3LuL5P0b7kz2D5H0jnp3Y22F7dRLQVnaTdJazJb4ZXObFU7AvqveO0Afbd31itE67RV6DYpIqZImlKo9WfJ9oyIGFXoOtAy9F/xou/Q3rTV7uXlkvbMuT8gnQYAwA6rrUL3n5IG2d7HdldJJ0u6r43WBQBAUWiT3csRUWP7PEl/kdRZ0q0RMb8t1lUkdojd6B0Y/Ve86Du0K46IQtcAAMAOgTNSAQCQEUIXAICMELoAAGSE0AUAICOELrANtksLXQNazvbRha4ByFWwM1J1VLaHS/qlklNh/knSRRGxLm37R0SMLmR9aJztQyTdLKlW0hcl/UDSvul3zcdFxJOFrA/bZvu4+pMk/ZftEkmKiN9nXxXwfoRu/t0o6QpJT0k6W9Ljto+OiOcldSlkYWjStZLGSeol6QFJx0bE47YPlHS9pEMKWRyadLeScwOsVhK4krSTpM9LCkmELgqO0M2/nSPiz+ntH9ueKenPtk9T8oeP9qtLRMyVJNvVEfG4JEXEM7Z7FLY0NMMYSVdJ+mdE3ChJtsdGxBcKWxbwHj7TbQO2e9fdjoiHJR0v6dfiiiDtXe7fw3fqtXXNshBsv4j4p6QjJHW1/bDt0eKNLtoZQjf/rpY0JHdCRMyRdLjYvdXeXWq7pyRFxB/qJtoeKOlXBasKzRYRtRHxM0n/If9DI88AAAYBSURBVOnCQtcD1MdpIAEAyAgj3Tyz3dv2VbYX2V5r+zXbC9NpuxS6PjSOvitu9B+KAaGbf9MkrZM0NiJKI6KPpE+k06YVtDI0hb4rbvQf2j12L+eZ7cURUbG9bSg8+q640X8oBox082+p7Qm2+9ZNsN3X9kWSXilgXWgafVfc6D+0e4Ru/p0kqY+kR22vs71O0iOSSpWceAHtF31X3Og/tHvsXgYAICOckaoN2N5X0nGS9pS0VdKzkv47IjYUtDA0ib4rbvQf2jt2L+eZ7a9JuklSd0kHSeqm5B/AU7bHFrA0NIG+K270H4oBu5fzzPZcSR+JiK3p2Y3+GBFjbe8l6d6IGFHgEtEI+q640X8oBox020bdbvtuSq5Yo4h4WVxlqBjQd8WN/kO7xme6+XezpH/aflrSoUrOxSzbZZLWFrIwNIm+K270H9o9di+3AdtDlVz0YF5ELCp0PWg++q640X9o7wjdDNnuFREbC10Hth99V9zoP7QXfKabrQWFLgAtRt8VN/oP7QKf6eaZ7W821qT0wA60T/RdcaP/UAwY6ebfJEm7Stq53k8vsb3bO/quuNF/aPcY6ebfM5L+EBEz6zfYPrsA9aD56LviRv+h3eNAqjyzXSFpbURUN9DWNyJWFaAsNAN9V9zoPxQDQhcAgIzwOUee2e5t+yrbi2yvtf2a7YXptF0KXR8aR98VN/oPxYDQzb9pktZJGhsRpRHRR9In0mnTCloZmkLfFTf6D+0eu5fzzPbiiKjY3jYUHn1X3Og/FANGuvm31PYE233rJtjua/siSa8UsC40jb4rbvQf2j1CN/9OktRH0qO219leK+kRSaWSxhWyMDSJvitu9B/aPXYvtwHblZIGSHoq93yvtj8TEX8uXGVoCn1X3Og/tHeMdPPM9tck3SvpPEnzbB+T0zypMFWhOei74kb/oRhwRqr8+5KkkRGx0Xa5pHtsl0fEz5ScAxbtF31X3Og/tHuEbv51qtutFREv2R6r5I9/b/GH397Rd8WN/kO7x+7l/Ftl+yN1d9J/AkdJ2k3S8IJVheag74ob/Yd2jwOp8sz2AEk1EbGygbZDIuKJApSFZqDvihv9h2JA6AIAkBF2LwMAkBFCFwCAjBC6aPeceNz2Z3OmnWj7z7a32v5Xzs/FOfPsZnuL7XPrLe8l23Ntz7H9aHp0a11b2L4z536J7Wrb/5sz7dj0sQvT5Ryb03a77eW2u+XU8JLt4Tk1rrX9Ynr7IdvltufVq/EK2xfmaxsCaB8IXbR7kRx4cK6kn9jubruXkpMdfFXSpoj4SM7PVTkPPVHSU5LGN7DYT0TE/kpOE/ifOdPflDTMdo/0/hGSltc12j5A0o8lHRMRQyQdLenHtvfPWcZWSV+s9xzm1tUo6T5J307vf2r7tgaAYkbooihExDxJ90u6SNJlkn4VEc838bDxkr4lqX96ZGtDnpTUv960P0r6XM4ypua0XShpUkS8mNb1oqQfSvp2zjw/lfQN23wPHsD7ELooJldKOkXSZyX9KJ3Wo97u5ZMkyfaekvpFxD+UXEv1pEaW+RlJf6g37TeSTrbdXdL+kp7OaRsqaWa9+Wek0+u8LOlxSadtx3MbmPs8lIzsAXQwvBNH0YiIN23fLWljRLyTTt6U7rKt7yS9d+Hy30i6VdLknPaHbZdK2ijp0nrrmZOeRnC8klFvS/xQyXmAH2jm/M/nPg/bV7RwvQDaMUa6KDa16U9Txks60/ZLSj5D3d/2oJz2T0jaW9K/lIyg67tPyWe3U+tNXyBpZL1pIyXNz50QEc+ly+aScgDeReiiw7E9WFKviOgfEeURUa5k5Pm+A6oiokbSBZJOT0e9uW6VdGVEzK03/ceSvpOOhJX+vkTvH0XXmajkM2AAkMTuZRS/HulnoHX+LGmTpP+pN9/vJN0t6Xu5EyNihe2pSo6E/n7O9GWSrqu/soj4l+2LJN1vu4ukLZImRMS/Gph3vu1nJB3YomcGoMPhNJAAAGSE3csAAGSE0AUAICOELgAAGSF0AQDICKELAEBGCF0AADJC6AIAkJH/D9FJNjbNN0oaAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#trial_full_observ = full_observ[(full_observ[\"YEARMONTH\"] >= 201902) & (full_observ[\"YEARMONTH\"] <= 201904)]\n",
    "scaled_ncust_control_stores = full_observ[full_observ[\"STORE_NBR\"].isin([233, 155, 40])][[\"STORE_NBR\", \"YEARMONTH\", \"nCustomers\"]]\n",
    "\n",
    "def scaler_c(row):\n",
    "    if row[\"STORE_NBR\"] == 233:\n",
    "        return row[\"nCustomers\"] * ncust_ratio_77\n",
    "    elif row[\"STORE_NBR\"] == 155:\n",
    "        return row[\"nCustomers\"] * ncust_ratio_86\n",
    "    elif row[\"STORE_NBR\"] == 40:\n",
    "        return row[\"nCustomers\"] * ncust_ratio_88\n",
    "\n",
    "scaled_ncust_control_stores[\"ScaledNcust\"] = scaled_ncust_control_stores.apply(lambda row: scaler_c(row), axis=1)\n",
    "\n",
    "trial_scaled_ncust_control_stores = scaled_ncust_control_stores[(scaled_ncust_control_stores[\"YEARMONTH\"] >= 201902) & (scaled_ncust_control_stores[\"YEARMONTH\"] <= 201904)]\n",
    "pretrial_scaled_ncust_control_stores = scaled_ncust_control_stores[scaled_ncust_control_stores[\"YEARMONTH\"] < 201902]\n",
    "\n",
    "ncust_percentage_diff = {}\n",
    "\n",
    "for trial, control in trial_control_dic.items():\n",
    "    a = trial_scaled_ncust_control_stores[trial_scaled_ncust_control_stores[\"STORE_NBR\"] == control]\n",
    "    b = trial_full_observ[trial_full_observ[\"STORE_NBR\"] == trial][[\"STORE_NBR\", \"YEARMONTH\", \"nCustomers\"]]\n",
    "    ncust_percentage_diff[trial] = b[\"nCustomers\"].sum() / a[\"ScaledNcust\"].sum()\n",
    "    b[[\"YEARMONTH\", \"nCustomers\"]].merge(a[[\"YEARMONTH\", \"ScaledNcust\"]],on=\"YEARMONTH\").set_index(\"YEARMONTH\").rename(columns={\"ScaledSales\":\"Scaled_Control_nCust\", \"TOT_SALES\":\"Trial_nCust\"}).plot.bar()\n",
    "    plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))\n",
    "    plt.title(\"Trial Store \"+str(trial)+\" and Control Store \"+str(control))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "lPJZq6xhFnNY",
    "outputId": "67ff149d-028e-4496-fad4-d474b97f64b9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{77: 1.2306529009742622, 86: 1.1354166666666667, 88: 1.0444876946258161}"
      ]
     },
     "execution_count": 35,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ncust_percentage_diff"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 316
    },
    "id": "6MewaZpdFnJo",
    "outputId": "882dd611-4aa6-4a73-ee8b-cc2cb0c3a661"
   },
   "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>c_STORE_NBR</th>\n",
       "      <th>YEARMONTH</th>\n",
       "      <th>c_ScaledNcust</th>\n",
       "      <th>t_STORE_NBR</th>\n",
       "      <th>t_nCustomers</th>\n",
       "      <th>nCust_Percentage_Diff</th>\n",
       "      <th>trial_period</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>233</td>\n",
       "      <td>201902</td>\n",
       "      <td>45.151007</td>\n",
       "      <td>77</td>\n",
       "      <td>45</td>\n",
       "      <td>-0.003350</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>233</td>\n",
       "      <td>201903</td>\n",
       "      <td>40.134228</td>\n",
       "      <td>77</td>\n",
       "      <td>50</td>\n",
       "      <td>0.218913</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>233</td>\n",
       "      <td>201904</td>\n",
       "      <td>30.100671</td>\n",
       "      <td>77</td>\n",
       "      <td>47</td>\n",
       "      <td>0.438370</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>155</td>\n",
       "      <td>201902</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>86</td>\n",
       "      <td>107</td>\n",
       "      <td>0.118812</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>155</td>\n",
       "      <td>201903</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>86</td>\n",
       "      <td>115</td>\n",
       "      <td>0.200957</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>155</td>\n",
       "      <td>201904</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>86</td>\n",
       "      <td>105</td>\n",
       "      <td>0.058824</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>40</td>\n",
       "      <td>201902</td>\n",
       "      <td>127.610209</td>\n",
       "      <td>88</td>\n",
       "      <td>124</td>\n",
       "      <td>-0.028697</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>40</td>\n",
       "      <td>201903</td>\n",
       "      <td>120.464037</td>\n",
       "      <td>88</td>\n",
       "      <td>134</td>\n",
       "      <td>0.106388</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>40</td>\n",
       "      <td>201904</td>\n",
       "      <td>121.484919</td>\n",
       "      <td>88</td>\n",
       "      <td>128</td>\n",
       "      <td>0.052228</td>\n",
       "      <td>trial</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    c_STORE_NBR  YEARMONTH  ...  nCust_Percentage_Diff  trial_period\n",
       "7           233     201902  ...              -0.003350         trial\n",
       "8           233     201903  ...               0.218913         trial\n",
       "9           233     201904  ...               0.438370         trial\n",
       "19          155     201902  ...               0.118812         trial\n",
       "20          155     201903  ...               0.200957         trial\n",
       "21          155     201904  ...               0.058824         trial\n",
       "31           40     201902  ...              -0.028697         trial\n",
       "32           40     201903  ...               0.106388         trial\n",
       "33           40     201904  ...               0.052228         trial\n",
       "\n",
       "[9 rows x 7 columns]"
      ]
     },
     "execution_count": 36,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp1 = scaled_ncust_control_stores.sort_values(by=[\"STORE_NBR\", \"YEARMONTH\"], ascending=[False, True]).reset_index().drop([\"nCustomers\", \"index\"], axis=1)\n",
    "temp2 = full_observ[full_observ[\"STORE_NBR\"].isin([77,86,88])][[\"STORE_NBR\", \"YEARMONTH\", \"nCustomers\"]].reset_index().drop([\"index\", \"YEARMONTH\"], axis=1)\n",
    "scaledncust_vs_trial = pd.concat([temp1, temp2], axis=1)\n",
    "scaledncust_vs_trial.columns = [\"c_STORE_NBR\", \"YEARMONTH\", \"c_ScaledNcust\", \"t_STORE_NBR\", \"t_nCustomers\"]\n",
    "scaledncust_vs_trial[\"nCust_Percentage_Diff\"] = (scaledncust_vs_trial[\"t_nCustomers\"] - scaledncust_vs_trial[\"c_ScaledNcust\"]) / (((scaledncust_vs_trial[\"t_nCustomers\"] + scaledncust_vs_trial[\"c_ScaledNcust\"])/2))\n",
    "\n",
    "scaledncust_vs_trial[\"trial_period\"] = scaledncust_vs_trial[\"YEARMONTH\"].apply(lambda cell: label_period(cell))\n",
    "scaledncust_vs_trial[scaledncust_vs_trial[\"trial_period\"] == \"trial\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vtcHZRLFGQSk"
   },
   "source": [
    "Check significance of Trial minus Control stores nCustomers Percentage Difference Pre-Trial vs Trial.\n",
    "\n",
    "Step 1: Check null hypothesis of 0 difference between control store's Pre-Trial and Trial period performance.\n",
    "\n",
    "Step 2: Proof control and trial stores are similar statistically\n",
    "\n",
    "Step 3: After checking Null Hypothesis of first 2 step to be true, we can check Null Hypothesis of Percentage Difference between Trial and Control stores during pre-trial is the same as during trial."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "3HNHGLHwFnHA",
    "outputId": "18977f49-7d15-40a1-d55f-b3fedc5ba55c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Store 40\n",
      "Ttest_indResult(statistic=0.644732693420032, pvalue=0.5376573016017127) \n",
      "\n",
      "Store 155\n",
      "Ttest_indResult(statistic=1.3888888888888882, pvalue=0.204345986327886) \n",
      "\n",
      "Store 233\n",
      "Ttest_indResult(statistic=0.8442563765225701, pvalue=0.4559280037660254) \n",
      "\n",
      "Critical t-value for 95% confidence interval:\n",
      "[-4.30265273  4.30265273]\n"
     ]
    }
   ],
   "source": [
    "# Step 1\n",
    "for num in [40, 155, 233]:\n",
    "    print(\"Store\", num)\n",
    "    print(ttest_ind(pretrial_scaled_ncust_control_stores[pretrial_scaled_ncust_control_stores[\"STORE_NBR\"] == num][\"ScaledNcust\"],\n",
    "                   trial_scaled_ncust_control_stores[trial_scaled_ncust_control_stores[\"STORE_NBR\"] == num][\"ScaledNcust\"],\n",
    "                   equal_var=False), '\\n')\n",
    "    \n",
    "alpha = 0.05\n",
    "print(\"Critical t-value for 95% confidence interval:\")\n",
    "print(t.ppf((alpha/2, 1-alpha/2), df=min([len(pretrial_scaled_ncust_control_stores[pretrial_scaled_ncust_control_stores[\"STORE_NBR\"] == num]),\n",
    "                       len(trial_scaled_ncust_control_stores[trial_scaled_ncust_control_stores[\"STORE_NBR\"] == num])])-1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "l4ZRTLLtGVQu",
    "outputId": "04edecb1-2f89-4465-8c90-0731a4d70f2c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial store: 77 , Control store: 233\n",
      "Ttest_indResult(statistic=0.0, pvalue=1.0) \n",
      "\n",
      "Trial store: 86 , Control store: 155\n",
      "Ttest_indResult(statistic=0.0, pvalue=1.0) \n",
      "\n",
      "Trial store: 88 , Control store: 40\n",
      "Ttest_indResult(statistic=-7.648483953264653e-15, pvalue=0.999999999999994) \n",
      "\n",
      "Critical t-value for 95% confidence interval:\n",
      "[-2.44691185  2.44691185]\n"
     ]
    }
   ],
   "source": [
    "# Step 2\n",
    "for trial, cont in trial_control_dic.items():\n",
    "    print(\"Trial store:\", trial, \", Control store:\", cont)\n",
    "    print(ttest_ind(pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == trial][\"nCustomers\"],\n",
    "                   pretrial_scaled_ncust_control_stores[pretrial_scaled_ncust_control_stores[\"STORE_NBR\"] == cont][\"ScaledNcust\"],\n",
    "                   equal_var=True), '\\n')\n",
    "\n",
    "alpha = 0.05\n",
    "print(\"Critical t-value for 95% confidence interval:\")\n",
    "print(t.ppf((alpha/2, 1-alpha/2), df=len(pretrial_full_observ[pretrial_full_observ[\"STORE_NBR\"] == trial])-1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "o0VdPXArGVtm",
    "outputId": "78e60ef9-2aa2-40cd-99c6-b5231b3a5bb2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial store: 77 , Control store: 233\n",
      "201902 : -0.19886295797440687\n",
      "201903 : 8.009609025380932\n",
      "201904 : 16.114474772873923\n",
      "\n",
      "\n",
      "Trial store: 86 , Control store: 155\n",
      "201902 : 6.220524882227514\n",
      "201903 : 10.52599074274189\n",
      "201904 : 3.0763575852842706\n",
      "\n",
      "\n",
      "Trial store: 88 , Control store: 40\n",
      "201902 : -0.3592881735131531\n",
      "201903 : 1.2575196020616801\n",
      "201904 : 0.6092905590514273\n",
      "\n",
      "\n",
      "Critical t-value for 95% confidence interval:\n",
      "1.9431802803927816\n"
     ]
    }
   ],
   "source": [
    "# Step 3\n",
    "for trial, cont in trial_control_dic.items():\n",
    "    print(\"Trial store:\", trial, \", Control store:\", cont)\n",
    "    temp_pre = scaledncust_vs_trial[(scaledncust_vs_trial[\"c_STORE_NBR\"] == cont) & (scaledncust_vs_trial[\"trial_period\"]==\"pre\")]\n",
    "    std = temp_pre[\"nCust_Percentage_Diff\"].std()\n",
    "    mean = temp_pre[\"nCust_Percentage_Diff\"].mean()\n",
    "    #print(std, mean)\n",
    "    for t_month in scaledncust_vs_trial[scaledncust_vs_trial[\"trial_period\"] == \"trial\"][\"YEARMONTH\"].unique():\n",
    "        pdif = scaledncust_vs_trial[(scaledncust_vs_trial[\"YEARMONTH\"] == t_month) & (scaledncust_vs_trial[\"t_STORE_NBR\"] == trial)][\"nCust_Percentage_Diff\"]\n",
    "        print(t_month,\":\",(float(pdif)-mean)/std)\n",
    "    print('\\n')\n",
    "    \n",
    "print(\"Critical t-value for 95% confidence interval:\")\n",
    "conf_intv_95 = t.ppf(0.95, df=len(temp_pre)-1)\n",
    "print(conf_intv_95)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "PizmI6PzGfmt"
   },
   "source": [
    "\n",
    "There are 5 months' increase in performance that are statistically significant (Above the 95% confidence interval t-score):\n",
    "\n",
    "March and April trial months for trial store 77\n",
    "\n",
    "Feb, March and April trial months for trial store 86"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 935
    },
    "id": "pQwuBwljGhb8",
    "outputId": "c639f95a-fb54-4e58-a223-ccaa1b88d9d8"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for trial, control in trial_control_dic.items():\n",
    "    a = trial_scaled_ncust_control_stores[trial_scaled_ncust_control_stores[\"STORE_NBR\"] == control].rename(columns={\"nCustomers\": \"control_nCustomers\"})\n",
    "    b = trial_full_observ[trial_full_observ[\"STORE_NBR\"] == trial][[\"STORE_NBR\", \"YEARMONTH\", \"nCustomers\"]].rename(columns={\"nCustomers\": \"trial_nCustomers\"})\n",
    "    comb = b[[\"YEARMONTH\", \"trial_nCustomers\"]].merge(a[[\"YEARMONTH\", \"control_nCustomers\"]],on=\"YEARMONTH\").set_index(\"YEARMONTH\")\n",
    "    comb.plot.bar()\n",
    "    cont_sc_ncust = trial_scaled_ncust_control_stores[trial_scaled_ncust_control_stores[\"STORE_NBR\"] == control][\"nCustomers\"]\n",
    "    std = scaledncust_vs_trial[(scaledncust_vs_trial[\"c_STORE_NBR\"] == control) & (scaledncust_vs_trial[\"trial_period\"]==\"pre\")][\"nCust_Percentage_Diff\"].std()\n",
    "    thresh95 = cont_sc_ncust.mean() + (cont_sc_ncust.mean() * std * 2)\n",
    "    thresh5 = cont_sc_ncust.mean() - (cont_sc_ncust.mean() * std * 2)\n",
    "    plt.axhline(y=thresh95,linewidth=1, color='b', label=\"95% threshold\")\n",
    "    plt.axhline(y=thresh5,linewidth=1, color='r', label=\"5% threshold\")\n",
    "    plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))\n",
    "    plt.title(\"Trial Store \"+str(trial)+\" and Control Store \"+str(control)+\" - Number of Customers\")\n",
    "    plt.savefig(\"TS {} and CS {} - nCustomers.png\".format(trial,control), bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "QFKJDMtsGmUa"
   },
   "source": [
    "We can see that Trial store 77 sales for Feb, March, and April exceeds 95% threshold of control store. Same goes to store 86 sales for all 3 trial months.\n",
    "\n",
    "1. Trial store 77: Control store 233\n",
    "2. Trial store 86: Control store 155\n",
    "3. Trial store 88: Control store 40\n",
    "4. Both trial store 77 and 86 showed significant increase in Total Sales and Number of Customers during trial period. But not for trial store 88. Perhaps the client knows if there's anything about trial 88 that differs it from the other two trial.\n",
    "5. Overall the trial showed positive significant result."
   ]
  }
 ],
 "metadata": {
  "colab": {
   "name": "Task2.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}