Multiple_Linear_Regression.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Multiple Linear Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Contents:\n",
    "* [Load Dataset](#Load-Dataset)\n",
    "* [Part 1: Analytics](#Part-1:-Analytics)\n",
    "    * [Question 1 (a)](#Question-1-(a))\n",
    "    * [Question 1 (b)](#Question-1-(b))\n",
    "    * [Question 1 (c)](#Question-1-(c))\n",
    "    * [Question 1 (d)](#Question-1-(d))\n",
    "* [Part 2: Regression](#Part-2:-Regression)\n",
    "    * [Line Plots (Before Outlier Removal)](#Line-Plots-(Before-Outlier-Removal))\n",
    "    * [Remove Outliers](#Remove-Outliers)\n",
    "    * [Line Plots (After Outlier Removal)](#Line-Plots-(After-Outlier-Removal))\n",
    "    * [Interaction Variable](#Interaction-Variable)\n",
    "    * [Train Test Split](#Train-Test-Split)\n",
    "    * [Regression Model](#Regression-Model)\n",
    "    * [Evaluation Metrics](#Evaluation-Metrics)\n",
    "    * [Plot (True vs Fitted Values)](#Plot-(True-vs-Fitted-Values))    \n",
    "    * [Coefficient Analysis](#Coefficient-Analysis)\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load Dataset "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import os\n",
    "from pathlib import Path\n",
    "path = Path(os.getcwd())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_1 = pd.read_excel('Adops & Data Scientist Sample Data.xlsx', sheet_name = 0)\n",
    "df_2 = pd.read_excel('Adops & Data Scientist Sample Data.xlsx', sheet_name = 1, header = None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Part 1: Analytics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ts</th>\n",
       "      <th>user_id</th>\n",
       "      <th>country_id</th>\n",
       "      <th>site_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-02-01 00:01:24</td>\n",
       "      <td>LC36FC</td>\n",
       "      <td>TL6</td>\n",
       "      <td>N0OTG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-02-01 00:10:19</td>\n",
       "      <td>LC39B6</td>\n",
       "      <td>TL6</td>\n",
       "      <td>N0OTG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-02-01 00:21:50</td>\n",
       "      <td>LC3500</td>\n",
       "      <td>TL6</td>\n",
       "      <td>N0OTG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-02-01 00:22:50</td>\n",
       "      <td>LC374F</td>\n",
       "      <td>TL6</td>\n",
       "      <td>N0OTG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-02-01 00:23:44</td>\n",
       "      <td>LCC1C3</td>\n",
       "      <td>TL6</td>\n",
       "      <td>QGO3G</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    ts user_id country_id site_id\n",
       "0  2019-02-01 00:01:24  LC36FC        TL6   N0OTG\n",
       "1  2019-02-01 00:10:19  LC39B6        TL6   N0OTG\n",
       "2  2019-02-01 00:21:50  LC3500        TL6   N0OTG\n",
       "3  2019-02-01 00:22:50  LC374F        TL6   N0OTG\n",
       "4  2019-02-01 00:23:44  LCC1C3        TL6   QGO3G"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ts            0\n",
       "user_id       0\n",
       "country_id    0\n",
       "site_id       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 1 (a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "844"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1_BDV = df_1[df_1['country_id'] == 'BDV']\n",
    "len(df_1_BDV)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "site_id\n",
       "5NPAU    544\n",
       "N0OTG     90\n",
       "3POLC      2\n",
       "Name: user_id, dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1_BDV.groupby('site_id')['user_id'].nunique().sort_values(ascending = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\"5NPAU\" has the highest number of unique visitors (544)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 1 (b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "object\n",
      "datetime64[ns]\n"
     ]
    }
   ],
   "source": [
    "print(df_1['ts'].dtype)\n",
    "df_1['ts'] = pd.to_datetime(df_1['ts'])\n",
    "print(df_1['ts'].dtype)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ts</th>\n",
       "      <th>user_id</th>\n",
       "      <th>country_id</th>\n",
       "      <th>site_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1049</th>\n",
       "      <td>2019-02-03 00:02:31</td>\n",
       "      <td>LC3C7E</td>\n",
       "      <td>TL6</td>\n",
       "      <td>3POLC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1050</th>\n",
       "      <td>2019-02-03 00:03:09</td>\n",
       "      <td>LC3C7E</td>\n",
       "      <td>TL6</td>\n",
       "      <td>3POLC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1051</th>\n",
       "      <td>2019-02-03 00:03:46</td>\n",
       "      <td>LC3C7E</td>\n",
       "      <td>TL6</td>\n",
       "      <td>3POLC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1052</th>\n",
       "      <td>2019-02-03 00:04:12</td>\n",
       "      <td>LC3C7E</td>\n",
       "      <td>TL6</td>\n",
       "      <td>3POLC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1053</th>\n",
       "      <td>2019-02-03 00:04:25</td>\n",
       "      <td>LC3C7E</td>\n",
       "      <td>TL6</td>\n",
       "      <td>3POLC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2075</th>\n",
       "      <td>2019-02-04 23:54:56</td>\n",
       "      <td>LC34B0</td>\n",
       "      <td>XA7</td>\n",
       "      <td>N0OTG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2076</th>\n",
       "      <td>2019-02-04 23:55:46</td>\n",
       "      <td>LC3DEA</td>\n",
       "      <td>TL6</td>\n",
       "      <td>N0OTG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2077</th>\n",
       "      <td>2019-02-04 23:56:12</td>\n",
       "      <td>LC06C3</td>\n",
       "      <td>TL6</td>\n",
       "      <td>N0OTG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2078</th>\n",
       "      <td>2019-02-04 23:56:54</td>\n",
       "      <td>LC06C3</td>\n",
       "      <td>TL6</td>\n",
       "      <td>N0OTG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2079</th>\n",
       "      <td>2019-02-04 23:57:03</td>\n",
       "      <td>LC06C3</td>\n",
       "      <td>TL6</td>\n",
       "      <td>N0OTG</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1031 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      ts user_id country_id site_id\n",
       "1049 2019-02-03 00:02:31  LC3C7E        TL6   3POLC\n",
       "1050 2019-02-03 00:03:09  LC3C7E        TL6   3POLC\n",
       "1051 2019-02-03 00:03:46  LC3C7E        TL6   3POLC\n",
       "1052 2019-02-03 00:04:12  LC3C7E        TL6   3POLC\n",
       "1053 2019-02-03 00:04:25  LC3C7E        TL6   3POLC\n",
       "...                  ...     ...        ...     ...\n",
       "2075 2019-02-04 23:54:56  LC34B0        XA7   N0OTG\n",
       "2076 2019-02-04 23:55:46  LC3DEA        TL6   N0OTG\n",
       "2077 2019-02-04 23:56:12  LC06C3        TL6   N0OTG\n",
       "2078 2019-02-04 23:56:54  LC06C3        TL6   N0OTG\n",
       "2079 2019-02-04 23:57:03  LC06C3        TL6   N0OTG\n",
       "\n",
       "[1031 rows x 4 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_temp = df_1[(df_1['ts'] >= '2019-02-03 00:00:00') & (df_1['ts'] <= '2019-02-04 23:59:59')]\n",
    "df_temp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>site_id</th>\n",
       "      <th>number of visits</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LC06C3</td>\n",
       "      <td>N0OTG</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>417</th>\n",
       "      <td>LC3A59</td>\n",
       "      <td>N0OTG</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>485</th>\n",
       "      <td>LC3C7E</td>\n",
       "      <td>3POLC</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>493</th>\n",
       "      <td>LC3C9D</td>\n",
       "      <td>N0OTG</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    user_id site_id  number of visits\n",
       "3    LC06C3   N0OTG                25\n",
       "417  LC3A59   N0OTG                26\n",
       "485  LC3C7E   3POLC                15\n",
       "493  LC3C9D   N0OTG                17"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1_Q2 = df_temp.groupby(['user_id', 'site_id'])['ts'].count().reset_index()\n",
    "df_1_Q2.columns = ['user_id', 'site_id', 'number of visits']\n",
    "df_1_Q2[df_1_Q2['number of visits'] > 10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 1 (c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>site_id</th>\n",
       "      <th>num_of_users</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5NPAU</td>\n",
       "      <td>992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>N0OTG</td>\n",
       "      <td>561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>QGO3G</td>\n",
       "      <td>289</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  site_id  num_of_users\n",
       "0   5NPAU           992\n",
       "1   N0OTG           561\n",
       "2   QGO3G           289"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1_Q3 = df_1[['user_id','ts']].groupby('user_id').max()\n",
    "df_1_Q3 = df_1_Q3.merge(df_1, on=['ts','user_id'])\n",
    "df_1_Q3 = df_1_Q3.groupby('site_id').user_id.nunique()\n",
    "df_1_Q3 = df_1_Q3.reset_index().rename(columns={'user_id':'num_of_users'})\n",
    "df_1_Q3.sort_values(by = 'num_of_users', ascending = False).reset_index(drop=True).head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 1 (d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ts_x</th>\n",
       "      <th>country_id_x</th>\n",
       "      <th>site_id</th>\n",
       "      <th>ts_y</th>\n",
       "      <th>country_id_y</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>user_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LC00C3</th>\n",
       "      <td>2019-02-03 18:52:50</td>\n",
       "      <td>QLT</td>\n",
       "      <td>5NPAU</td>\n",
       "      <td>2019-02-03 18:52:50</td>\n",
       "      <td>QLT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LC01C3</th>\n",
       "      <td>2019-02-04 11:35:10</td>\n",
       "      <td>QLT</td>\n",
       "      <td>5NPAU</td>\n",
       "      <td>2019-02-04 11:35:10</td>\n",
       "      <td>QLT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LC05C3</th>\n",
       "      <td>2019-02-02 14:14:44</td>\n",
       "      <td>BDV</td>\n",
       "      <td>5NPAU</td>\n",
       "      <td>2019-02-02 14:14:44</td>\n",
       "      <td>BDV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LC06C3</th>\n",
       "      <td>2019-02-01 22:49:39</td>\n",
       "      <td>TL6</td>\n",
       "      <td>N0OTG</td>\n",
       "      <td>2019-02-07 01:16:12</td>\n",
       "      <td>TL6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LC07C3</th>\n",
       "      <td>2019-02-05 19:06:42</td>\n",
       "      <td>BDV</td>\n",
       "      <td>5NPAU</td>\n",
       "      <td>2019-02-05 19:06:42</td>\n",
       "      <td>BDV</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       ts_x country_id_x site_id                ts_y  \\\n",
       "user_id                                                                \n",
       "LC00C3  2019-02-03 18:52:50          QLT   5NPAU 2019-02-03 18:52:50   \n",
       "LC01C3  2019-02-04 11:35:10          QLT   5NPAU 2019-02-04 11:35:10   \n",
       "LC05C3  2019-02-02 14:14:44          BDV   5NPAU 2019-02-02 14:14:44   \n",
       "LC06C3  2019-02-01 22:49:39          TL6   N0OTG 2019-02-07 01:16:12   \n",
       "LC07C3  2019-02-05 19:06:42          BDV   5NPAU 2019-02-05 19:06:42   \n",
       "\n",
       "        country_id_y  \n",
       "user_id               \n",
       "LC00C3           QLT  \n",
       "LC01C3           QLT  \n",
       "LC05C3           BDV  \n",
       "LC06C3           TL6  \n",
       "LC07C3           BDV  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1_first = df_1.groupby('user_id').first()\n",
    "df_1_last = df_1.groupby('user_id').last()\n",
    "\n",
    "df_result = df_1_first.merge(df_1_last, on=['user_id','site_id'], how='inner')\n",
    "df_result.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1670"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_result)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Part 2: Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>A</th>\n",
       "      <th>B</th>\n",
       "      <th>C</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.490142</td>\n",
       "      <td>-0.179654</td>\n",
       "      <td>11.536508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-1.414793</td>\n",
       "      <td>-1.225605</td>\n",
       "      <td>11.828531</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.943066</td>\n",
       "      <td>4.506148</td>\n",
       "      <td>-3.235349</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3.569090</td>\n",
       "      <td>5.068347</td>\n",
       "      <td>-23.891922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-1.702460</td>\n",
       "      <td>6.905051</td>\n",
       "      <td>-22.125437</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          A         B          C\n",
       "0  0.490142 -0.179654  11.536508\n",
       "1 -1.414793 -1.225605  11.828531\n",
       "2  0.943066  4.506148  -3.235349\n",
       "3  3.569090  5.068347 -23.891922\n",
       "4 -1.702460  6.905051 -22.125437"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_2.columns = ['A', 'B', 'C']\n",
    "df_2.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Line Plots (Before Outlier Removal)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "for i, col in enumerate(df_2.columns):\n",
    "    df_2[col].plot(fig=plt.figure(i))\n",
    "    plt.title(col)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Remove Outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def replace_outlier(val, mean, std):\n",
    "    '''\n",
    "    Replaces extreme values with median.\n",
    "    \n",
    "    PARAMETERS:\n",
    "    val (float): Value.\n",
    "    mean (float): Mean of the column.\n",
    "    std (float): Standard deviation of the column.\n",
    "\n",
    "    RETURNS:\n",
    "    median (float): Median value which is used to replace outliers.\n",
    "    '''\n",
    "    if val > mean + 3*std:\n",
    "        return median \n",
    "    elif val < mean - 3*std:\n",
    "        return median \n",
    "    return val\n",
    "\n",
    "for col in df_2.columns:\n",
    "    median = df_2[col].median()\n",
    "    mean = df_2[col].mean()\n",
    "    std_dev = df_2[col].std(axis=0)\n",
    "    df_2[col] = df_2[col].map(lambda x: replace_outlier(x, mean, std_dev))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Line Plots (After Outlier Removal)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i, col in enumerate(df_2.columns):\n",
    "    df_2[col].plot(fig=plt.figure(i))\n",
    "    plt.title(col)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Interaction Variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x228f7e5ccc0>]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_2['AxB'] = df_2['A'] * df_2['B']\n",
    "plt.plot(df_2['AxB'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train Test Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_test_split(df, ratio):\n",
    "    '''\n",
    "    Split data into train and test data.\n",
    "    \n",
    "    PARAMETERS:\n",
    "    df (pandas.core.frame.DataFrame): Original data to be split.\n",
    "    ratio (float): Split ratio of the data (Between 0.25 to 1; default = 0.75).\n",
    "\n",
    "    RETURNS:\n",
    "    train_set (pandas.core.frame.DataFrame): Train data.\n",
    "    test_set (pandas.core.frame.DataFrame): Test data.\n",
    "    '''\n",
    "    if((ratio > 1) and (ratio < 0.25)):\n",
    "        ratio = 0.75\n",
    "    num_of_rows = int(len(df) * ratio)\n",
    "    df = df.sample(frac = 1)\n",
    "    train_set = df[:num_of_rows]\n",
    "    test_set = df[num_of_rows:]\n",
    "    return train_set, test_set\n",
    "\n",
    "train_set, test_set = train_test_split(df_2, 0.70)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = train_set[['A', 'B', 'AxB']]\n",
    "y_train = train_set[['C']]\n",
    "\n",
    "X_test = test_set[['A', 'B', 'AxB']]\n",
    "y_test = test_set[['C']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-9.53579730e-02,  2.36096366e+00, -2.25136709e-01],\n",
       "       [-7.97415386e-01,  6.05381976e+00, -4.82740902e+00],\n",
       "       [-2.58928061e+00,  2.27701109e+00, -5.89582067e+00],\n",
       "       [ 2.12152570e-01,  3.20132598e+00,  6.79169535e-01],\n",
       "       [ 7.01464061e+00,  3.18254772e+00,  2.23244284e+01],\n",
       "       [-2.39718926e+00,  1.28994027e+00, -3.09223097e+00],\n",
       "       [-2.43752271e+00,  6.86001064e+00, -1.67214317e+01],\n",
       "       [-1.73616435e+00,  5.76653603e+00, -1.00116543e+01],\n",
       "       [-4.45298073e+00,  6.50607147e+00, -2.89714109e+01],\n",
       "       [-3.72816236e+00,  5.22537243e+00, -1.94810368e+01],\n",
       "       [ 4.16185239e+00,  3.90394829e+00,  1.62476565e+01],\n",
       "       [-1.96618455e+00,  5.42118923e+00, -1.06590585e+01],\n",
       "       [-2.68686259e+00,  6.02552907e+00, -1.61897686e+01],\n",
       "       [-2.48521449e-01,  5.49647391e+00, -1.36599166e+00],\n",
       "       [-6.67232231e-01,  6.50061967e+00, -4.33742296e+00],\n",
       "       [-4.17313279e+00,  4.52374529e+00, -1.88781898e+01],\n",
       "       [-6.28912047e+00,  4.00969131e+00, -2.52174317e+01],\n",
       "       [ 1.10387932e+01,  6.20872401e+00,  6.85368206e+01],\n",
       "       [ 1.79384036e+00,  6.86639579e+00,  1.23172179e+01],\n",
       "       [-3.40683181e+00,  5.90263784e+00, -2.01092943e+01],\n",
       "       [ 4.69037895e+00,  2.60937972e+00,  1.22389797e+01],\n",
       "       [-3.03238909e-01, -1.70840585e+00,  5.18055125e-01],\n",
       "       [ 1.00710465e+01,  5.17072071e+00,  5.20745685e+01],\n",
       "       [-1.70241087e+00,  1.12617677e+00, -1.91721557e+00],\n",
       "       [-4.69259295e+00,  1.68693160e+00, -7.91608332e+00],\n",
       "       [ 1.14200148e+00,  4.83653013e+00,  5.52332457e+00],\n",
       "       [ 1.36125381e+00,  3.71573196e+00,  5.05805429e+00],\n",
       "       [-1.87508125e+00,  1.48445522e+00, -2.78347415e+00],\n",
       "       [-3.10615928e+00,  6.69302536e+00, -2.07896028e+01],\n",
       "       [ 8.37563325e+00, -1.87961098e+00, -1.57429322e+01],\n",
       "       [ 6.18652488e+00,  3.40874745e+00,  2.10883009e+01],\n",
       "       [ 4.19712874e-01,  9.65427823e-01,  4.05202486e-01],\n",
       "       [-6.87901037e+00,  1.94420305e+00, -1.33741929e+01],\n",
       "       [ 1.22668518e+01,  4.99410399e+00,  6.12619335e+01],\n",
       "       [-1.79397050e+00,  5.05875588e-01, -9.07525882e-01],\n",
       "       [ 1.05413579e+01,  4.56672833e-01,  4.81395176e+00],\n",
       "       [ 4.20777292e-01, -1.49583109e+00, -6.29411755e-01],\n",
       "       [-1.67732890e+00, -2.89162349e+00,  4.85020364e+00],\n",
       "       [ 1.33238199e+01, -1.62240942e-01, -2.16166909e+00],\n",
       "       [-4.57391049e+00,  3.18128240e+00, -1.45509010e+01],\n",
       "       [ 7.84405921e+00, -1.47140861e+00, -1.15418162e+01],\n",
       "       [-7.23802020e-01,  4.89171239e+00, -3.54063131e+00],\n",
       "       [-1.37466710e-02,  5.91143114e+00, -8.12624991e-02],\n",
       "       [ 1.18620053e+01, -1.07710981e+00, -1.27766823e+01],\n",
       "       [-1.09045980e-01,  6.44765730e+00, -7.03091111e-01],\n",
       "       [ 1.54318323e+01, -1.15478981e+00, -1.78205226e+01],\n",
       "       [-3.26120849e+00,  1.03482866e+00, -3.37479202e+00],\n",
       "       [-2.40842316e+00,  4.08035403e-01, -9.82721913e-01],\n",
       "       [ 9.55936294e+00,  1.84522985e+00,  1.76392219e+01],\n",
       "       [ 1.59439757e+01, -9.36662816e-01, -1.49341292e+01],\n",
       "       [-3.92404501e+00, -1.86442408e+00,  7.31608400e+00],\n",
       "       [ 1.34546862e+00,  6.99717673e+00,  9.41448169e+00],\n",
       "       [ 7.54729778e+00,  2.20791656e-01,  1.66638037e+00],\n",
       "       [-2.55481065e+00,  6.63222539e+00, -1.69440801e+01],\n",
       "       [ 1.10587806e+01,  2.03417271e+00,  2.24954696e+01],\n",
       "       [-4.80457222e-01,  1.59135756e+00, -7.64579234e-01],\n",
       "       [-1.92763713e+00,  7.06421471e-01, -1.36172425e+00],\n",
       "       [ 1.53991001e+01, -1.26626405e+00, -1.94993268e+01],\n",
       "       [-4.98455815e+00, -2.42441240e+00,  1.20846246e+01],\n",
       "       [ 1.20973958e+01,  7.72841631e-01,  9.34937109e+00],\n",
       "       [ 8.35028867e-01, -2.18651219e+00, -1.82580080e+00],\n",
       "       [ 3.61410970e+00,  1.68660642e+00,  6.09558062e+00],\n",
       "       [ 7.67986654e+00,  1.31659546e+00,  1.01112774e+01],\n",
       "       [-6.17475350e+00, -1.96876131e+00,  1.21566158e+01],\n",
       "       [-2.26193597e+00, -1.90368104e-01,  4.30600462e-01],\n",
       "       [ 4.68779852e-01,  1.73961640e+00,  8.15497120e-01],\n",
       "       [-2.39025308e+00,  5.58412752e+00, -1.33474780e+01],\n",
       "       [ 8.49080751e-02, -1.88802518e+00, -1.60308584e-01],\n",
       "       [-9.84659630e-01,  2.84314312e+00, -2.79952825e+00],\n",
       "       [-7.24717670e-01,  6.62172548e+00, -4.79888147e+00],\n",
       "       [ 1.12769077e+01, -2.14096610e-01, -2.41434770e+00],\n",
       "       [ 1.22491537e+01, -1.71954161e+00, -2.10629295e+01],\n",
       "       [ 3.20838293e+00,  6.49799913e+00,  2.08480695e+01],\n",
       "       [ 1.25579910e+00, -1.37383061e+00, -1.72525524e+00],\n",
       "       [ 8.24229555e+00, -2.13079712e+00, -1.75626596e+01],\n",
       "       [-5.54454167e+00,  6.49520624e+00, -3.60129417e+01],\n",
       "       [-2.77480918e-02, -2.27236994e+00,  6.30539296e-02],\n",
       "       [-7.38858795e-01,  4.46044879e+00, -3.29564182e+00],\n",
       "       [ 7.04001230e+00,  2.32113485e+00,  1.63408179e+01],\n",
       "       [ 8.54464471e+00, -2.19766254e+00, -1.87782456e+01],\n",
       "       [-3.14305425e+00,  3.06174634e+00, -9.62323487e+00],\n",
       "       [-2.51042696e+00, -2.73633026e+00,  6.86935725e+00],\n",
       "       [-4.85895156e-01,  5.87704183e+00, -2.85562616e+00],\n",
       "       [ 1.14385864e+01,  6.29375989e+00,  7.19917160e+01],\n",
       "       [ 2.92142826e+00, -2.65929281e-01, -7.76893317e-01],\n",
       "       [-2.80191607e+00,  3.31837212e+00, -9.29780017e+00],\n",
       "       [ 3.08548687e-02, -2.15893194e+00, -6.66135615e-02],\n",
       "       [-1.11639169e-01,  1.54541065e+00, -1.72528360e-01],\n",
       "       [ 6.92127121e+00, -4.20971730e-01, -2.91365951e+00],\n",
       "       [ 3.56908957e+00,  5.06834739e+00,  1.80893858e+01],\n",
       "       [ 6.21920859e+00,  2.93592454e+00,  1.82591271e+01],\n",
       "       [ 1.16987996e+01, -1.29112413e+00, -1.51046024e+01],\n",
       "       [ 1.37309584e+00,  6.10927184e+00,  8.38861576e+00],\n",
       "       [ 4.65855770e+00, -2.26188170e-01, -1.05371064e+00],\n",
       "       [-3.66854329e+00, -1.65984772e+00,  6.08922320e+00],\n",
       "       [-8.25373845e-01,  3.33400854e+00, -2.75180345e+00],\n",
       "       [-1.48385713e+00,  1.55656753e+00, -2.30972383e+00],\n",
       "       [-2.16834183e-01,  3.27558080e+00, -7.10257888e-01],\n",
       "       [-6.73984073e+00,  4.54542874e+00, -3.06354658e+01],\n",
       "       [ 1.90593497e+00,  2.53854284e+00,  4.83829758e+00],\n",
       "       [ 1.06823536e+01, -1.71120278e+00, -1.82796732e+01],\n",
       "       [ 1.05149618e+01, -2.63316797e+00, -2.76876607e+01],\n",
       "       [-2.74113185e-01,  4.50871068e+00, -1.23589705e+00],\n",
       "       [ 4.69241355e+00,  6.30822964e-01,  2.96008222e+00],\n",
       "       [-7.01045905e-01, -2.54781330e+00,  1.78613408e+00],\n",
       "       [-1.90331109e+00, -1.82932984e+00,  3.48178376e+00],\n",
       "       [-2.17632446e+00,  3.29398638e+00, -7.16878313e+00],\n",
       "       [-2.20351617e-01,  5.12901009e+00, -1.13018567e+00],\n",
       "       [ 3.43368213e+00, -7.64041615e-01, -2.62347604e+00],\n",
       "       [ 6.05547405e+00,  5.60967259e-01,  3.39692268e+00],\n",
       "       [ 2.34559244e-02,  3.20309551e+00,  7.51315664e-02],\n",
       "       [-1.23130513e+00,  4.55137256e+00, -5.60412837e+00],\n",
       "       [-5.24611223e+00, -2.54553620e+00,  1.33541686e+01],\n",
       "       [-3.15953263e+00,  4.61510632e+00, -1.45815790e+01],\n",
       "       [ 3.39694631e+00,  8.92016787e-01,  3.03013313e+00],\n",
       "       [-3.75827270e+00,  2.94130715e+00, -1.10542344e+01],\n",
       "       [-6.20970579e-03,  5.12799567e+00, -3.18433444e-02],\n",
       "       [-3.17620196e-01,  1.14819502e+00, -3.64689929e-01],\n",
       "       [-2.33954486e+00, -1.52926519e+00,  3.57778451e+00],\n",
       "       [ 1.28007334e+01, -8.85519930e-01, -1.13353046e+01],\n",
       "       [-2.80511984e+00, -6.78922830e-02,  1.90445990e-01],\n",
       "       [ 6.85408887e+00,  5.73578624e+00,  3.93135886e+01],\n",
       "       [-4.96136984e+00,  2.55431706e+00, -1.26729116e+01],\n",
       "       [-3.46204696e+00,  4.72318392e+00, -1.63518845e+01],\n",
       "       [-2.63314817e+00,  1.93136376e-01, -5.08556695e-01],\n",
       "       [-4.21267749e+00,  6.04158694e+00, -2.54512574e+01],\n",
       "       [-2.27348828e-01,  1.63698405e+00, -3.72166405e-01],\n",
       "       [-3.54038115e+00,  5.62363709e+00, -1.99098188e+01],\n",
       "       [-3.57147267e+00,  1.37474922e+00, -4.90987927e+00],\n",
       "       [ 1.21614062e+01, -1.48097306e+00, -1.80107149e+01],\n",
       "       [ 1.08888865e+01,  2.56762893e+00,  2.79586201e+01],\n",
       "       [ 1.03314583e+01, -1.80237868e+00, -1.86212002e+01],\n",
       "       [ 1.27094055e-01,  2.73437888e+00,  3.47523300e-01],\n",
       "       [-3.03076600e+00,  4.06242227e+00, -1.23122513e+01],\n",
       "       [-4.09416292e-01,  2.49528882e+00, -1.02161190e+00],\n",
       "       [ 5.39802299e-01,  6.92964796e+00,  3.74063990e+00],\n",
       "       [-9.60994324e-01, -1.63628524e+00,  1.57246083e+00],\n",
       "       [-1.21848674e+00,  3.22086700e+00, -3.92458373e+00],\n",
       "       [ 5.76676567e+00, -2.18240968e+00, -1.25854452e+01],\n",
       "       [ 8.65579068e+00,  4.47470938e+00,  3.87321477e+01],\n",
       "       [-4.46098424e-01,  5.64722376e+00, -2.51921762e+00],\n",
       "       [ 8.82142393e+00,  3.79102319e+00,  3.34422227e+01],\n",
       "       [-1.55697693e+00,  4.53378185e+00, -7.05899375e+00],\n",
       "       [ 5.25665045e+00,  2.37106542e+00,  1.24638621e+01],\n",
       "       [-1.07954163e+00, -2.22265363e+00,  2.39944711e+00],\n",
       "       [ 1.09017571e+01, -9.06506767e-01, -9.88251655e+00],\n",
       "       [ 1.55952139e-01,  6.80032575e+00,  1.06052535e+00],\n",
       "       [-4.50603411e+00,  5.13795020e+00, -2.31517789e+01],\n",
       "       [ 4.44189010e+00, -2.79928802e+00, -1.24341298e+01],\n",
       "       [ 3.69393097e+00, -1.81182084e+00, -6.69274110e+00],\n",
       "       [ 4.21498763e-01,  3.71700684e+00,  1.56671379e+00],\n",
       "       [-6.96270674e+00,  7.48705795e-01, -5.21301889e+00],\n",
       "       [-5.27424456e+00, -2.08713323e+00,  1.10080511e+01],\n",
       "       [ 7.16050750e+00,  2.89917687e+00,  2.07595777e+01],\n",
       "       [ 7.57516407e+00,  2.11677442e+00,  1.60349135e+01],\n",
       "       [ 1.17535858e+01, -5.75294964e-01, -6.76177874e+00],\n",
       "       [ 1.46763474e+00,  4.91579044e+00,  7.21458480e+00],\n",
       "       [-7.08767352e-01, -2.26203435e+00,  1.60325610e+00],\n",
       "       [ 1.89012839e+00,  2.20163501e+00,  4.16137283e+00],\n",
       "       [ 1.46618048e+00,  2.51906839e+00,  3.69340890e+00],\n",
       "       [ 1.09541738e+01, -2.59131384e+00, -2.83857020e+01],\n",
       "       [ 6.38972634e+00,  6.86210744e+00,  4.38469887e+01],\n",
       "       [ 3.64980322e+00,  8.08908566e-01,  2.95235709e+00],\n",
       "       [-4.18691114e+00,  6.87276129e+00, -2.87756408e+01],\n",
       "       [-1.22333775e+00,  5.33522280e-01, -6.52677944e-01],\n",
       "       [ 7.60571281e-01,  3.13415196e+00,  2.38374597e+00],\n",
       "       [-1.41479290e+00, -1.22560456e+00,  1.73397664e+00],\n",
       "       [ 1.00733621e+01, -2.24136672e+00, -2.25780985e+01],\n",
       "       [ 1.92663538e+00,  6.47248577e+00,  1.24701201e+01],\n",
       "       [ 8.41868165e-02, -2.88646355e+00, -2.43002177e-01],\n",
       "       [ 7.99649629e+00,  5.95206838e+00,  4.75956927e+01],\n",
       "       [ 1.56919638e+00,  6.26587625e+00,  9.83239035e+00],\n",
       "       [-1.21603036e+00,  1.24222009e+00, -1.51057735e+00],\n",
       "       [ 7.58420440e+00, -2.60687860e+00, -1.97711002e+01],\n",
       "       [ 1.24894913e+01,  3.45917241e+00,  4.32033036e+01],\n",
       "       [ 8.05419227e+00, -1.31064937e+00, -1.05562220e+01],\n",
       "       [ 1.30230419e+00,  4.76412961e+00,  6.20434594e+00],\n",
       "       [-1.89702205e+00,  2.83368765e+00, -5.37556796e+00],\n",
       "       [ 1.24066969e+01, -1.27680129e+00, -1.58408866e+01],\n",
       "       [ 7.13190144e+00,  6.82378617e+00,  4.86665704e+01],\n",
       "       [ 3.73763845e+00,  7.20180858e-01,  2.69177566e+00],\n",
       "       [-3.34975988e+00,  6.69914398e+00, -2.24405237e+01],\n",
       "       [-4.73721634e+00, -4.17556117e-01,  1.97805366e+00],\n",
       "       [-4.76266562e-01, -1.11878840e+00,  5.32841506e-01],\n",
       "       [ 4.70957587e+00,  4.45439474e+00,  2.09783100e+01],\n",
       "       [-6.75631365e+00,  2.83656112e+00, -1.91646966e+01],\n",
       "       [-5.39054484e+00,  3.95748689e+00, -2.13330106e+01],\n",
       "       [ 2.42846844e+00,  3.84731173e+00,  9.34307512e+00],\n",
       "       [ 6.80890010e+00, -2.38922040e+00, -1.62679630e+01],\n",
       "       [ 9.24562242e+00,  1.60778768e+00,  1.48649978e+01],\n",
       "       [-3.72407223e+00,  2.00496010e-01, -7.46661624e-01],\n",
       "       [ 4.86699190e+00, -1.61172827e+00, -7.84426845e+00],\n",
       "       [ 5.57136688e+00,  1.18243036e+00,  6.58775336e+00],\n",
       "       [ 7.54390936e+00, -1.79364129e+00, -1.35310673e+01],\n",
       "       [ 6.87699160e+00,  2.74973522e-01,  1.89099060e+00],\n",
       "       [ 5.65824697e-01,  4.68987415e+00,  2.65364662e+00],\n",
       "       [ 4.90142459e-01, -1.79654274e-01, -8.80561878e-02],\n",
       "       [ 1.15972656e+01, -4.75630557e-01, -5.51601388e+00],\n",
       "       [ 2.47578674e+00,  2.20307701e+00,  5.45434884e+00],\n",
       "       [-3.04007416e+00,  1.50544135e+00, -4.57665336e+00],\n",
       "       [ 3.14373660e+00,  5.96765425e+00,  1.87607331e+01],\n",
       "       [-4.42891089e+00, -5.98543812e-01,  2.65089721e+00],\n",
       "       [ 7.19934937e+00, -6.27732083e-01, -4.51926257e+00],\n",
       "       [-8.85923531e+00, -1.82473753e+00,  1.61657792e+01],\n",
       "       [ 5.28655350e+00, -1.49048313e-01, -7.87951882e-01],\n",
       "       [-4.71085213e+00,  6.96636837e+00, -3.28175313e+01],\n",
       "       [-4.58861987e+00, -2.16499283e+00,  9.93432914e+00],\n",
       "       [ 9.84297560e+00,  1.10396827e+00,  1.08663327e+01],\n",
       "       [-1.10413531e+00,  9.52761629e-02, -1.05197776e-01],\n",
       "       [ 9.69660826e-01,  5.81627180e-01,  5.63981092e-01]])"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train = X_train.values\n",
    "X_test = X_test.values\n",
    "y_train = y_train.values\n",
    "y_test = y_test.values\n",
    "\n",
    "X_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Regression Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "class Multiple_Linear_Regression():   \n",
    "    def __init__ (self):\n",
    "        self.theta=np.zeros(int(np.random.random()),float)[:,np.newaxis]; \n",
    "    \n",
    "    def fit(self, X_train, y_train):\n",
    "        '''\n",
    "        Fit data to the model based on the matrix approach to Linear Regression. Actual magic happens here!  \n",
    "        \n",
    "        PARAMETERS:\n",
    "        X_train (numpy.ndarray): Independent variables train data.\n",
    "        y_train (numpy.ndarray): Dependent variable of train data.\n",
    "\n",
    "        RETURNS:\n",
    "        None\n",
    "        '''\n",
    "        X_b = np.c_[np.ones(len(X_train)), X_train] \n",
    "        theta_bst = np.linalg.pinv(X_b.T.dot(X_b)).dot(X_b.T).dot(y_train)\n",
    "        self.theta = theta_bst\n",
    "    \n",
    "    def predict(self, X_test):\n",
    "        '''\n",
    "        Fucntion to predict on new data.\n",
    "        \n",
    "        PARAMETERS:\n",
    "        X_test (numpy.ndarray): Independent variables data.\n",
    "\n",
    "        RETURNS:\n",
    "        y_predict (numpy.ndarray): Predicted dependent variable.\n",
    "        '''\n",
    "        X_test = np.c_[np.ones((len(X_test), 1)), X_test]\n",
    "        y_predict = np.dot(X_test, self.theta)\n",
    "        \n",
    "        return y_predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Multiple_Linear_Regression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_preds = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluation Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def mse(y_preds, y):\n",
    "    '''\n",
    "    Mean squared error.\n",
    "    \n",
    "    PARAMETERS:\n",
    "    y_preds (numpy.ndarray): Predictions.\n",
    "    y (numpy.ndarray): Original dependent variable values.\n",
    "    \n",
    "    RETURNS:\n",
    "    mse (float): Mean squared error.\n",
    "    '''\n",
    "    mse = ((y - y_preds)**2).mean()\n",
    "    return mse\n",
    "\n",
    "def rmse(y_preds, y):\n",
    "    '''\n",
    "    Root Mean squared error.\n",
    "    \n",
    "    PARAMETERS:\n",
    "    y_preds (numpy.ndarray): Predictions.\n",
    "    y (numpy.ndarray): Original dependent variable values.\n",
    "    \n",
    "    RETURNS:\n",
    "    rmse (float): Root Mean Squared Error.\n",
    "    '''\n",
    "    rmse = (((y - y_preds)**2).mean())**(1/2)\n",
    "    return rmse\n",
    "\n",
    "def r2(y_preds, y):\n",
    "    '''\n",
    "    R square score.\n",
    "    \n",
    "    PARAMETERS:\n",
    "    y_preds (numpy.ndarray): Predictions.\n",
    "    y (numpy.ndarray): Original dependent variable values.\n",
    "    \n",
    "    RETURNS:\n",
    "    r2 (float): R-Squared score.\n",
    "    '''\n",
    "    ssr = sum((y - y_preds) ** 2) \n",
    "    sst = sum((y - y.mean()) ** 2) \n",
    "    return 1 - ssr/sst"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1486.5680564854385, 38.55603787327529, array([0.77058015]))"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mse(y_preds, y_test), rmse(y_preds, y_test), r2(y_preds ,y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here, RMSE and the R2 value are helpful in assesing the model performance. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot (True vs Fitted Values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "ax.scatter(y_test, y_preds)\n",
    "ax.set_xlabel('True Values')\n",
    "ax.set_ylabel('Fitted Values')\n",
    "ax.set_title('True Value vs. Fitted Values');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks pretty good. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Coefficient Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 10.491129  ],\n",
       "       [  2.71755829],\n",
       "       [-10.41377649],\n",
       "       [ -2.19281552]])"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.theta"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Interesting. This means that:\n",
    "* Unit increase in 'A' will lead to an increase in the 'C' and the effect is positive. \n",
    "* Unit increase in 'B' will lead to a decrease in the 'C' and the effect is negative. \n",
    "* Unit increase in 'AxB' will lead to a decrease in the 'C' and the effect is negative. \n",
    "\n",
    "One thing to note is that, using the interaction variable 'AxB' has signifacntly increased the model performance.\n",
    "\n",
    "With more information regarding the actual nature of the data, their meaning and a background we can be able to further explore this data and build a stronger explanatory model."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}