{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Chapter 3 - Handling Missing Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.3. Handling Missing Numerical Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.3.1. Mean and Median Imputation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>survived</th>\n",
       "      <th>pclass</th>\n",
       "      <th>sex</th>\n",
       "      <th>age</th>\n",
       "      <th>sibsp</th>\n",
       "      <th>parch</th>\n",
       "      <th>fare</th>\n",
       "      <th>embarked</th>\n",
       "      <th>class</th>\n",
       "      <th>who</th>\n",
       "      <th>adult_male</th>\n",
       "      <th>deck</th>\n",
       "      <th>embark_town</th>\n",
       "      <th>alive</th>\n",
       "      <th>alone</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>S</td>\n",
       "      <td>Third</td>\n",
       "      <td>man</td>\n",
       "      <td>True</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>no</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C</td>\n",
       "      <td>First</td>\n",
       "      <td>woman</td>\n",
       "      <td>False</td>\n",
       "      <td>C</td>\n",
       "      <td>Cherbourg</td>\n",
       "      <td>yes</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>S</td>\n",
       "      <td>Third</td>\n",
       "      <td>woman</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>yes</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>S</td>\n",
       "      <td>First</td>\n",
       "      <td>woman</td>\n",
       "      <td>False</td>\n",
       "      <td>C</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>yes</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>S</td>\n",
       "      <td>Third</td>\n",
       "      <td>man</td>\n",
       "      <td>True</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>no</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   survived  pclass     sex   age  sibsp  parch     fare embarked  class  \\\n",
       "0         0       3    male  22.0      1      0   7.2500        S  Third   \n",
       "1         1       1  female  38.0      1      0  71.2833        C  First   \n",
       "2         1       3  female  26.0      0      0   7.9250        S  Third   \n",
       "3         1       1  female  35.0      1      0  53.1000        S  First   \n",
       "4         0       3    male  35.0      0      0   8.0500        S  Third   \n",
       "\n",
       "     who  adult_male deck  embark_town alive  alone  \n",
       "0    man        True  NaN  Southampton    no  False  \n",
       "1  woman       False    C    Cherbourg   yes  False  \n",
       "2  woman       False  NaN  Southampton   yes   True  \n",
       "3  woman       False    C  Southampton   yes  False  \n",
       "4    man        True  NaN  Southampton    no   True  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = [8,6]\n",
    "sns.set_style(\"darkgrid\")\n",
    "\n",
    "titanic_data = sns.load_dataset('titanic')\n",
    "\n",
    "titanic_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>survived</th>\n",
       "      <th>pclass</th>\n",
       "      <th>age</th>\n",
       "      <th>fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>22.0</td>\n",
       "      <td>7.2500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>71.2833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.9250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>53.1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>35.0</td>\n",
       "      <td>8.0500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   survived  pclass   age     fare\n",
       "0         0       3  22.0   7.2500\n",
       "1         1       1  38.0  71.2833\n",
       "2         1       3  26.0   7.9250\n",
       "3         1       1  35.0  53.1000\n",
       "4         0       3  35.0   8.0500"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "titanic_data  = titanic_data[[\"survived\", \"pclass\", \"age\", \"fare\"]]\n",
    "titanic_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "survived    0.000000\n",
       "pclass      0.000000\n",
       "age         0.198653\n",
       "fare        0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "titanic_data.isnull().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "28.0\n",
      "29.69911764705882\n"
     ]
    }
   ],
   "source": [
    "median = titanic_data.age.median()\n",
    "print(median)\n",
    "\n",
    "mean = titanic_data.age.mean()\n",
    "print(mean)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\usman\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:3: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n",
      "C:\\Users\\usman\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:5: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"\n",
      "C:\\Users\\usman\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:7: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  import sys\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>survived</th>\n",
       "      <th>pclass</th>\n",
       "      <th>age</th>\n",
       "      <th>fare</th>\n",
       "      <th>Median_Age</th>\n",
       "      <th>Mean_Age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>22.0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>22.0</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>38.0</td>\n",
       "      <td>38.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>26.0</td>\n",
       "      <td>26.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>35.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>35.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>35.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.4583</td>\n",
       "      <td>28.0</td>\n",
       "      <td>29.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>54.0</td>\n",
       "      <td>51.8625</td>\n",
       "      <td>54.0</td>\n",
       "      <td>54.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>21.0750</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>27.0</td>\n",
       "      <td>11.1333</td>\n",
       "      <td>27.0</td>\n",
       "      <td>27.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>14.0</td>\n",
       "      <td>30.0708</td>\n",
       "      <td>14.0</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4.0</td>\n",
       "      <td>16.7000</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>58.0</td>\n",
       "      <td>26.5500</td>\n",
       "      <td>58.0</td>\n",
       "      <td>58.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>20.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>20.0</td>\n",
       "      <td>20.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>39.0</td>\n",
       "      <td>31.2750</td>\n",
       "      <td>39.0</td>\n",
       "      <td>39.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>14.0</td>\n",
       "      <td>7.8542</td>\n",
       "      <td>14.0</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>16.0000</td>\n",
       "      <td>55.0</td>\n",
       "      <td>55.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>29.1250</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>28.0</td>\n",
       "      <td>29.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>31.0</td>\n",
       "      <td>18.0000</td>\n",
       "      <td>31.0</td>\n",
       "      <td>31.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.2250</td>\n",
       "      <td>28.0</td>\n",
       "      <td>29.7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    survived  pclass   age     fare  Median_Age  Mean_Age\n",
       "0          0       3  22.0   7.2500        22.0      22.0\n",
       "1          1       1  38.0  71.2833        38.0      38.0\n",
       "2          1       3  26.0   7.9250        26.0      26.0\n",
       "3          1       1  35.0  53.1000        35.0      35.0\n",
       "4          0       3  35.0   8.0500        35.0      35.0\n",
       "5          0       3   NaN   8.4583        28.0      29.7\n",
       "6          0       1  54.0  51.8625        54.0      54.0\n",
       "7          0       3   2.0  21.0750         2.0       2.0\n",
       "8          1       3  27.0  11.1333        27.0      27.0\n",
       "9          1       2  14.0  30.0708        14.0      14.0\n",
       "10         1       3   4.0  16.7000         4.0       4.0\n",
       "11         1       1  58.0  26.5500        58.0      58.0\n",
       "12         0       3  20.0   8.0500        20.0      20.0\n",
       "13         0       3  39.0  31.2750        39.0      39.0\n",
       "14         0       3  14.0   7.8542        14.0      14.0\n",
       "15         1       2  55.0  16.0000        55.0      55.0\n",
       "16         0       3   2.0  29.1250         2.0       2.0\n",
       "17         1       2   NaN  13.0000        28.0      29.7\n",
       "18         0       3  31.0  18.0000        31.0      31.0\n",
       "19         1       3   NaN   7.2250        28.0      29.7"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "titanic_data['Median_Age'] = titanic_data.age.fillna(median)\n",
    "\n",
    "titanic_data['Mean_Age'] = titanic_data.age.fillna(mean)\n",
    "\n",
    "titanic_data['Mean_Age']  = np.round(titanic_data['Mean_Age'], 1)\n",
    "\n",
    "titanic_data.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2181c88f948>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = [8,6]\n",
    "\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "titanic_data['age'] .plot(kind='kde', ax=ax)\n",
    "\n",
    "titanic_data['Median_Age'] .plot(kind='kde', ax=ax, color='red')\n",
    "\n",
    "titanic_data['Mean_Age'] .plot(kind='kde', ax=ax, color='green')\n",
    "\n",
    "lines, labels = ax.get_legend_handles_labels()\n",
    "ax.legend(lines, labels, loc='best')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.3.2. End of Distribution Imputation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "survived    0.000000\n",
       "pclass      0.000000\n",
       "age         0.198653\n",
       "fare        0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = [8,6]\n",
    "sns.set_style(\"darkgrid\")\n",
    "\n",
    "titanic_data = sns.load_dataset('titanic')\n",
    "\n",
    "\n",
    "\n",
    "titanic_data  = titanic_data[[\"survived\", \"pclass\", \"age\", \"fare\"]]\n",
    "\n",
    "\n",
    "titanic_data.isnull().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2181c9c17c8>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "titanic_data.age.hist(bins=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "73.27860964406095\n"
     ]
    }
   ],
   "source": [
    "eod_value = titanic_data.age.mean() + 3 * titanic_data.age.std()\n",
    "print(eod_value)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>survived</th>\n",
       "      <th>pclass</th>\n",
       "      <th>age</th>\n",
       "      <th>fare</th>\n",
       "      <th>age_eod</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>22.0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>22.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>38.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>26.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>35.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>35.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>35.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.4583</td>\n",
       "      <td>73.27861</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>54.0</td>\n",
       "      <td>51.8625</td>\n",
       "      <td>54.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>21.0750</td>\n",
       "      <td>2.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>27.0</td>\n",
       "      <td>11.1333</td>\n",
       "      <td>27.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>14.0</td>\n",
       "      <td>30.0708</td>\n",
       "      <td>14.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4.0</td>\n",
       "      <td>16.7000</td>\n",
       "      <td>4.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>58.0</td>\n",
       "      <td>26.5500</td>\n",
       "      <td>58.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>20.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>20.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>39.0</td>\n",
       "      <td>31.2750</td>\n",
       "      <td>39.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>14.0</td>\n",
       "      <td>7.8542</td>\n",
       "      <td>14.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>16.0000</td>\n",
       "      <td>55.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>29.1250</td>\n",
       "      <td>2.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>73.27861</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>31.0</td>\n",
       "      <td>18.0000</td>\n",
       "      <td>31.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.2250</td>\n",
       "      <td>73.27861</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    survived  pclass   age     fare   age_eod\n",
       "0          0       3  22.0   7.2500  22.00000\n",
       "1          1       1  38.0  71.2833  38.00000\n",
       "2          1       3  26.0   7.9250  26.00000\n",
       "3          1       1  35.0  53.1000  35.00000\n",
       "4          0       3  35.0   8.0500  35.00000\n",
       "5          0       3   NaN   8.4583  73.27861\n",
       "6          0       1  54.0  51.8625  54.00000\n",
       "7          0       3   2.0  21.0750   2.00000\n",
       "8          1       3  27.0  11.1333  27.00000\n",
       "9          1       2  14.0  30.0708  14.00000\n",
       "10         1       3   4.0  16.7000   4.00000\n",
       "11         1       1  58.0  26.5500  58.00000\n",
       "12         0       3  20.0   8.0500  20.00000\n",
       "13         0       3  39.0  31.2750  39.00000\n",
       "14         0       3  14.0   7.8542  14.00000\n",
       "15         1       2  55.0  16.0000  55.00000\n",
       "16         0       3   2.0  29.1250   2.00000\n",
       "17         1       2   NaN  13.0000  73.27861\n",
       "18         0       3  31.0  18.0000  31.00000\n",
       "19         1       3   NaN   7.2250  73.27861"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "titanic_data['age_eod'] = titanic_data.age.fillna(eod_value)\n",
    "titanic_data.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2181cb0be88>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = [8,6]\n",
    "\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "titanic_data['age'] .plot(kind='kde', ax=ax)\n",
    "titanic_data['age_eod'] .plot(kind='kde', ax=ax)\n",
    "lines, labels = ax.get_legend_handles_labels()\n",
    "ax.legend(lines, labels, loc='best')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.3.3. Aribitrary Value  Imputation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "survived    0.000000\n",
       "pclass      0.000000\n",
       "age         0.198653\n",
       "fare        0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = [8,6]\n",
    "sns.set_style(\"darkgrid\")\n",
    "\n",
    "titanic_data = sns.load_dataset('titanic')\n",
    "\n",
    "\n",
    "\n",
    "titanic_data  = titanic_data[[\"survived\", \"pclass\", \"age\", \"fare\"]]\n",
    "\n",
    "\n",
    "titanic_data.isnull().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2181c883e88>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "titanic_data.age.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>survived</th>\n",
       "      <th>pclass</th>\n",
       "      <th>age</th>\n",
       "      <th>fare</th>\n",
       "      <th>age_99</th>\n",
       "      <th>age_minus1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>22.0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>22.0</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>38.0</td>\n",
       "      <td>38.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>26.0</td>\n",
       "      <td>26.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>35.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>35.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>35.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.4583</td>\n",
       "      <td>99.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>54.0</td>\n",
       "      <td>51.8625</td>\n",
       "      <td>54.0</td>\n",
       "      <td>54.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>21.0750</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>27.0</td>\n",
       "      <td>11.1333</td>\n",
       "      <td>27.0</td>\n",
       "      <td>27.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>14.0</td>\n",
       "      <td>30.0708</td>\n",
       "      <td>14.0</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4.0</td>\n",
       "      <td>16.7000</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>58.0</td>\n",
       "      <td>26.5500</td>\n",
       "      <td>58.0</td>\n",
       "      <td>58.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>20.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>20.0</td>\n",
       "      <td>20.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>39.0</td>\n",
       "      <td>31.2750</td>\n",
       "      <td>39.0</td>\n",
       "      <td>39.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>14.0</td>\n",
       "      <td>7.8542</td>\n",
       "      <td>14.0</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>16.0000</td>\n",
       "      <td>55.0</td>\n",
       "      <td>55.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>29.1250</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>99.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>31.0</td>\n",
       "      <td>18.0000</td>\n",
       "      <td>31.0</td>\n",
       "      <td>31.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.2250</td>\n",
       "      <td>99.0</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    survived  pclass   age     fare  age_99  age_minus1\n",
       "0          0       3  22.0   7.2500    22.0        22.0\n",
       "1          1       1  38.0  71.2833    38.0        38.0\n",
       "2          1       3  26.0   7.9250    26.0        26.0\n",
       "3          1       1  35.0  53.1000    35.0        35.0\n",
       "4          0       3  35.0   8.0500    35.0        35.0\n",
       "5          0       3   NaN   8.4583    99.0        -1.0\n",
       "6          0       1  54.0  51.8625    54.0        54.0\n",
       "7          0       3   2.0  21.0750     2.0         2.0\n",
       "8          1       3  27.0  11.1333    27.0        27.0\n",
       "9          1       2  14.0  30.0708    14.0        14.0\n",
       "10         1       3   4.0  16.7000     4.0         4.0\n",
       "11         1       1  58.0  26.5500    58.0        58.0\n",
       "12         0       3  20.0   8.0500    20.0        20.0\n",
       "13         0       3  39.0  31.2750    39.0        39.0\n",
       "14         0       3  14.0   7.8542    14.0        14.0\n",
       "15         1       2  55.0  16.0000    55.0        55.0\n",
       "16         0       3   2.0  29.1250     2.0         2.0\n",
       "17         1       2   NaN  13.0000    99.0        -1.0\n",
       "18         0       3  31.0  18.0000    31.0        31.0\n",
       "19         1       3   NaN   7.2250    99.0        -1.0"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "titanic_data['age_99'] = titanic_data.age.fillna(99)\n",
    "\n",
    "titanic_data['age_minus1'] = titanic_data.age.fillna(-1)\n",
    "\n",
    "\n",
    "titanic_data.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2181c9dfb08>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = [8,6]\n",
    "\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "titanic_data['age'] .plot(kind='kde', ax=ax)\n",
    "\n",
    "titanic_data['age_99'] .plot(kind='kde', ax=ax, color='red')\n",
    "\n",
    "titanic_data['age_minus1'] .plot(kind='kde', ax=ax, color='green')\n",
    "\n",
    "lines, labels = ax.get_legend_handles_labels()\n",
    "ax.legend(lines, labels, loc='best')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.4. Handling Categorical Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.4.1. Frequent Category Imputation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "embark_town    0.002245\n",
       "age            0.198653\n",
       "fare           0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = [8,6]\n",
    "sns.set_style(\"darkgrid\")\n",
    "\n",
    "titanic_data = sns.load_dataset('titanic')\n",
    "\n",
    "titanic_data  = titanic_data[[\"embark_town\", \"age\", \"fare\"]]\n",
    "titanic_data.head()\n",
    "titanic_data.isnull().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Number of Passengers')"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "titanic_data.embark_town.value_counts().sort_values(ascending=False).plot.bar()\n",
    "plt.xlabel('Embark Town')\n",
    "plt.ylabel('Number of Passengers')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    Southampton\n",
       "dtype: object"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "titanic_data.embark_town.mode()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "titanic_data.embark_town.fillna('Southampton', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    24.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "titanic_data.age.mode()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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>embark_town</th>\n",
       "      <th>age</th>\n",
       "      <th>fare</th>\n",
       "      <th>age_mode</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>22.0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Cherbourg</td>\n",
       "      <td>38.0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>38.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>26.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>35.0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>35.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>Queenstown</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.4583</td>\n",
       "      <td>24.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>54.0</td>\n",
       "      <td>51.8625</td>\n",
       "      <td>54.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>2.0</td>\n",
       "      <td>21.0750</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>27.0</td>\n",
       "      <td>11.1333</td>\n",
       "      <td>27.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>Cherbourg</td>\n",
       "      <td>14.0</td>\n",
       "      <td>30.0708</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>4.0</td>\n",
       "      <td>16.7000</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>58.0</td>\n",
       "      <td>26.5500</td>\n",
       "      <td>58.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>20.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>20.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>39.0</td>\n",
       "      <td>31.2750</td>\n",
       "      <td>39.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>14.0</td>\n",
       "      <td>7.8542</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>55.0</td>\n",
       "      <td>16.0000</td>\n",
       "      <td>55.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>Queenstown</td>\n",
       "      <td>2.0</td>\n",
       "      <td>29.1250</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>24.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>31.0</td>\n",
       "      <td>18.0000</td>\n",
       "      <td>31.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>Cherbourg</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.2250</td>\n",
       "      <td>24.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    embark_town   age     fare  age_mode\n",
       "0   Southampton  22.0   7.2500      22.0\n",
       "1     Cherbourg  38.0  71.2833      38.0\n",
       "2   Southampton  26.0   7.9250      26.0\n",
       "3   Southampton  35.0  53.1000      35.0\n",
       "4   Southampton  35.0   8.0500      35.0\n",
       "5    Queenstown   NaN   8.4583      24.0\n",
       "6   Southampton  54.0  51.8625      54.0\n",
       "7   Southampton   2.0  21.0750       2.0\n",
       "8   Southampton  27.0  11.1333      27.0\n",
       "9     Cherbourg  14.0  30.0708      14.0\n",
       "10  Southampton   4.0  16.7000       4.0\n",
       "11  Southampton  58.0  26.5500      58.0\n",
       "12  Southampton  20.0   8.0500      20.0\n",
       "13  Southampton  39.0  31.2750      39.0\n",
       "14  Southampton  14.0   7.8542      14.0\n",
       "15  Southampton  55.0  16.0000      55.0\n",
       "16   Queenstown   2.0  29.1250       2.0\n",
       "17  Southampton   NaN  13.0000      24.0\n",
       "18  Southampton  31.0  18.0000      31.0\n",
       "19    Cherbourg   NaN   7.2250      24.0"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "\n",
    "titanic_data['age_mode'] = titanic_data.age.fillna(24)\n",
    "\n",
    "\n",
    "titanic_data.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2181c7df848>"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = [8,6]\n",
    "\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "titanic_data['age'] .plot(kind='kde', ax=ax)\n",
    "\n",
    "titanic_data['age_mode'] .plot(kind='kde', ax=ax, color='red')\n",
    "\n",
    "\n",
    "lines, labels = ax.get_legend_handles_labels()\n",
    "ax.legend(lines, labels, loc='best')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.4.2. Missing Category Imputation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "embark_town    0.002245\n",
       "age            0.198653\n",
       "fare           0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = [8,6]\n",
    "sns.set_style(\"darkgrid\")\n",
    "\n",
    "titanic_data = sns.load_dataset('titanic')\n",
    "\n",
    "titanic_data  = titanic_data[[\"embark_town\", \"age\", \"fare\"]]\n",
    "titanic_data.head()\n",
    "titanic_data.isnull().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "titanic_data.embark_town.fillna('Missing', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Number of Passengers')"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "titanic_data.embark_town.value_counts().sort_values(ascending=False).plot.bar()\n",
    "plt.xlabel('Embark Town')\n",
    "plt.ylabel('Number of Passengers')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exercise 3.1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Question 1\n",
    "\n",
    "What is the major disadvantage of mean and median imputation?\n",
    "\n",
    "A. Distorts the data distribution \\\n",
    "B. Distorts the data variance \\\n",
    "C. Distorts the data covariance \\\n",
    "D. All of the Above\n",
    "\n",
    "Answer: D"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Question 1\n",
    "\n",
    "Which imputation should be used when the data is not missing at random.\n",
    "\n",
    "A. Mean and Median Imputation \\\n",
    "B. Aribitrary Value Imputation \\\n",
    "C. End of Distribution Imputation \\\n",
    "D. Missing Value Imputation\n",
    "\n",
    "Answer: A"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Question 3\n",
    "\n",
    "How should the end of tail distribution should be calculated for normal distribution?\n",
    "\n",
    "A. IQR Rule \\\n",
    "B. Mean x 3 Standard deviations\\\n",
    "C. Mean \\\n",
    "D. Median\n",
    "\n",
    "Answer: B"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exercise 3.2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Replace the missing values in the \"deck\" column of the Titanic dataset by the most \n",
    "frequently occuring categories in that column. Plot a bar plot for the updated \"deck\" column. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = [8,6]\n",
    "sns.set_style(\"darkgrid\")\n",
    "\n",
    "titanic_data = sns.load_dataset('titanic')\n",
    "\n",
    "titanic_data  = titanic_data[[\"deck\"]]\n",
    "titanic_data.head()\n",
    "titanic_data.isnull().mean()\n",
    "\n",
    "titanic_data.deck.value_counts().sort_values(ascending=False).plot.bar()\n",
    "plt.xlabel('deck')\n",
    "plt.ylabel('Number of Passengers')\n",
    "\n",
    "titanic_data.deck.mode()\n",
    "\n",
    "titanic_data.deck.fillna('C', inplace=True)\n",
    "\n",
    "titanic_data.deck.value_counts().sort_values(ascending=False).plot.bar()\n",
    "plt.xlabel('deck')\n",
    "plt.ylabel('Number of Passengers')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}