Group Project - 1_Regression (Group G)v2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#     Machine Learning in Finance\n",
    "##     NOVA IMS\n",
    "####     Group G: Carlos Cardoso - 20211220 | Carlota Reis - 20211208 | Guilherme Miranda - 20210420 | Mariana Garcia - 20210838 "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Project 1 - Regression\n",
    "## Forecasting the number of motor insurance claims\n",
    "### This notebook uses the dateset *freMTPL2freq.csv*\n",
    "\n",
    "(c) Nuno António 2022 - Rev. 1.0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dataset description\n",
    "\n",
    "- **IDpol**: The policy ID (used to link with the claims dataset).\n",
    "- **ClaimNb**: Number of claims during the exposure period.\n",
    "- **Exposure**: The exposure period.\n",
    "- **Area**: The area code.\n",
    "- **VehPower**: The power of the car (ordered categorical).\n",
    "- **VehAge**: The vehicle age, in years.\n",
    "- **DrivAge**: The driver age, in years (in France, people can drive a car at 18).\n",
    "- **BonusMalus**: Bonus/malus, between 50 and 350: <100 means bonus, >100 means malus in France.\n",
    "- **VehBrand**: The car brand (unknown categories).\n",
    "- **VehGas**: The car gas, Diesel or regular.\n",
    "- **Density**: The density of inhabitants (number of inhabitants per km2) in the city the driver of the car lives in.\n",
    "- **Region**: The policy regions in France (based on a standard French classification)\n",
    "\n",
    "For additional information on the dataset check https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3164764"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Work description\n",
    "\n",
    "### Overview\n",
    "<p>You should organize into groups of 3 to 5 students, where you will assume the role of a consultant. You are asked to develop a model to forecast how many claims will each policy holder from a car insurer in France have in the following year. The insurance company wants to use this model to improve the policies' premiums (pricing).</p>\n",
    "<p>Employing the CRISP-DM process model, you are expected to define, describe and explain the model built. Simultaneous, you should explain how your model can help the insurance company reaching its objectives.</p>\n",
    "\n",
    "### Questions or additional informations\n",
    "For any additional questions, don't hesitate to get in touch with the instructor. The instructor will also act as the insurance company/project stakeholder."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ----------------"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Project 1 objective consists in developing a predictive data analytics solution for a French insurance company, following a CRISP-DM methodology.\n",
    "The developed model seeks to forecast the number of claims each policyholder will have in the following year. By having this information, the insurance company could adjust its pricing model for the next year’s premiums according to the predicted number of claims.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Business Understading"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "This phase of the CRIPS-DM framework is somehow compromised since we can’t contact the insurance French company and interact with its associates and workers. It would be interesting to contact a sales team member and observe their daily activity. Besides the top management, it would be also fundamental to contact the IT department to assess the analytics and the data availability. One could also already get some key insights of the company’s data.\n",
    "Given the business objective, the target variable defined in this problem is the number of claims.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Initializations and data loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Loading packages\n",
    "import os\n",
    "import csv\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import category_encoders as ce\n",
    "from sklearn import preprocessing\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import datasets, linear_model\n",
    "from sklearn import metrics\n",
    "from yellowbrick.model_selection import RFECV\n",
    "from yellowbrick.model_selection import LearningCurve\n",
    "from yellowbrick.regressor import ResidualsPlot\n",
    "from yellowbrick.regressor import PredictionError\n",
    "from sklearn.metrics import r2_score, mean_absolute_error\n",
    "from sklearn import tree\n",
    "from sklearn.tree import export_graphviz\n",
    "import graphviz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "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>count</th>\n",
       "      <th>unique</th>\n",
       "      <th>top</th>\n",
       "      <th>freq</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>IDpol</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2621856.921071</td>\n",
       "      <td>1641782.752655</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1157951.0</td>\n",
       "      <td>2272152.0</td>\n",
       "      <td>4046274.0</td>\n",
       "      <td>6114330.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ClaimNb</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.053247</td>\n",
       "      <td>0.240117</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>16.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Exposure</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.52875</td>\n",
       "      <td>0.364442</td>\n",
       "      <td>0.002732</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.99</td>\n",
       "      <td>2.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Area</th>\n",
       "      <td>678013</td>\n",
       "      <td>6</td>\n",
       "      <td>C</td>\n",
       "      <td>191880</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehPower</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.454631</td>\n",
       "      <td>2.050906</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>15.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehAge</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.044265</td>\n",
       "      <td>5.666232</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DrivAge</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>45.499122</td>\n",
       "      <td>14.137444</td>\n",
       "      <td>18.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BonusMalus</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>59.761502</td>\n",
       "      <td>15.636658</td>\n",
       "      <td>50.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>230.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehBrand</th>\n",
       "      <td>678013</td>\n",
       "      <td>11</td>\n",
       "      <td>B12</td>\n",
       "      <td>166024</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehGas</th>\n",
       "      <td>678013</td>\n",
       "      <td>2</td>\n",
       "      <td>Regular</td>\n",
       "      <td>345877</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Density</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1792.422405</td>\n",
       "      <td>3958.646564</td>\n",
       "      <td>1.0</td>\n",
       "      <td>92.0</td>\n",
       "      <td>393.0</td>\n",
       "      <td>1658.0</td>\n",
       "      <td>27000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Region</th>\n",
       "      <td>678013</td>\n",
       "      <td>22</td>\n",
       "      <td>R24</td>\n",
       "      <td>160601</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               count unique      top    freq            mean             std  \\\n",
       "IDpol       678013.0    NaN      NaN     NaN  2621856.921071  1641782.752655   \n",
       "ClaimNb     678013.0    NaN      NaN     NaN        0.053247        0.240117   \n",
       "Exposure    678013.0    NaN      NaN     NaN         0.52875        0.364442   \n",
       "Area          678013      6        C  191880             NaN             NaN   \n",
       "VehPower    678013.0    NaN      NaN     NaN        6.454631        2.050906   \n",
       "VehAge      678013.0    NaN      NaN     NaN        7.044265        5.666232   \n",
       "DrivAge     678013.0    NaN      NaN     NaN       45.499122       14.137444   \n",
       "BonusMalus  678013.0    NaN      NaN     NaN       59.761502       15.636658   \n",
       "VehBrand      678013     11      B12  166024             NaN             NaN   \n",
       "VehGas        678013      2  Regular  345877             NaN             NaN   \n",
       "Density     678013.0    NaN      NaN     NaN     1792.422405     3958.646564   \n",
       "Region        678013     22      R24  160601             NaN             NaN   \n",
       "\n",
       "                 min        25%        50%        75%        max  \n",
       "IDpol            1.0  1157951.0  2272152.0  4046274.0  6114330.0  \n",
       "ClaimNb          0.0        0.0        0.0        0.0       16.0  \n",
       "Exposure    0.002732       0.18       0.49       0.99       2.01  \n",
       "Area             NaN        NaN        NaN        NaN        NaN  \n",
       "VehPower         4.0        5.0        6.0        7.0       15.0  \n",
       "VehAge           0.0        2.0        6.0       11.0      100.0  \n",
       "DrivAge         18.0       34.0       44.0       55.0      100.0  \n",
       "BonusMalus      50.0       50.0       50.0       64.0      230.0  \n",
       "VehBrand         NaN        NaN        NaN        NaN        NaN  \n",
       "VehGas           NaN        NaN        NaN        NaN        NaN  \n",
       "Density          1.0       92.0      393.0     1658.0    27000.0  \n",
       "Region           NaN        NaN        NaN        NaN        NaN  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Loading the dataset and visualizing summary statistics\n",
    "ds = pd.read_csv('freMTPL2freq.csv')\n",
    "ds.describe(include='all').T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>IDpol</th>\n",
       "      <th>ClaimNb</th>\n",
       "      <th>Exposure</th>\n",
       "      <th>Area</th>\n",
       "      <th>VehPower</th>\n",
       "      <th>VehAge</th>\n",
       "      <th>DrivAge</th>\n",
       "      <th>BonusMalus</th>\n",
       "      <th>VehBrand</th>\n",
       "      <th>VehGas</th>\n",
       "      <th>Density</th>\n",
       "      <th>Region</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.10</td>\n",
       "      <td>D</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>55</td>\n",
       "      <td>50</td>\n",
       "      <td>B12</td>\n",
       "      <td>Regular</td>\n",
       "      <td>1217</td>\n",
       "      <td>R82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.77</td>\n",
       "      <td>D</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>55</td>\n",
       "      <td>50</td>\n",
       "      <td>B12</td>\n",
       "      <td>Regular</td>\n",
       "      <td>1217</td>\n",
       "      <td>R82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.75</td>\n",
       "      <td>B</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>52</td>\n",
       "      <td>50</td>\n",
       "      <td>B12</td>\n",
       "      <td>Diesel</td>\n",
       "      <td>54</td>\n",
       "      <td>R22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.09</td>\n",
       "      <td>B</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>46</td>\n",
       "      <td>50</td>\n",
       "      <td>B12</td>\n",
       "      <td>Diesel</td>\n",
       "      <td>76</td>\n",
       "      <td>R72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.84</td>\n",
       "      <td>B</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>46</td>\n",
       "      <td>50</td>\n",
       "      <td>B12</td>\n",
       "      <td>Diesel</td>\n",
       "      <td>76</td>\n",
       "      <td>R72</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   IDpol  ClaimNb  Exposure Area  VehPower  VehAge  DrivAge  BonusMalus  \\\n",
       "0    1.0        1      0.10    D         5       0       55          50   \n",
       "1    3.0        1      0.77    D         5       0       55          50   \n",
       "2    5.0        1      0.75    B         6       2       52          50   \n",
       "3   10.0        1      0.09    B         7       0       46          50   \n",
       "4   11.0        1      0.84    B         7       0       46          50   \n",
       "\n",
       "  VehBrand   VehGas  Density Region  \n",
       "0      B12  Regular     1217    R82  \n",
       "1      B12  Regular     1217    R82  \n",
       "2      B12   Diesel       54    R22  \n",
       "3      B12   Diesel       76    R72  \n",
       "4      B12   Diesel       76    R72  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show top rows\n",
    "ds.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Summary Statistics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The data presented in freMTPL2freq has a total of 678 013 observations, described along 12 different variables/attributes/descriptors/features.\n",
    "The minimum and maximum values of all the numeric features seem to be in line with the variables’ units. Although it is possible to start questioning the maximum values of the features: Claim Numbers, Exposure and Vehicle Age.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Understanding"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The previous summary statistics and data overview also concern the Data Understanding phase.\n",
    "It is crucial to deepen the data understanding and examine the dataset and hence it is further verified the presence/absence of missing values, followed by a graphical visualization.\n",
    "It is confirmed that the data doesn’t present missing values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "IDpol         0\n",
      "ClaimNb       0\n",
      "Exposure      0\n",
      "Area          0\n",
      "VehPower      0\n",
      "VehAge        0\n",
      "DrivAge       0\n",
      "BonusMalus    0\n",
      "VehBrand      0\n",
      "VehGas        0\n",
      "Density       0\n",
      "Region        0\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Confirm that there are not any missing values\n",
    "print(ds.isnull().sum())\n",
    "\n",
    "# No missing values on the dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Regarding the attributes’ type, there is 4 categorical variables: Area, Vehicle Brand, Vehicle Gas and Region, with the remaining variables being numeric."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 678013 entries, 0 to 678012\n",
      "Data columns (total 12 columns):\n",
      " #   Column      Non-Null Count   Dtype  \n",
      "---  ------      --------------   -----  \n",
      " 0   IDpol       678013 non-null  float64\n",
      " 1   ClaimNb     678013 non-null  int64  \n",
      " 2   Exposure    678013 non-null  float64\n",
      " 3   Area        678013 non-null  object \n",
      " 4   VehPower    678013 non-null  int64  \n",
      " 5   VehAge      678013 non-null  int64  \n",
      " 6   DrivAge     678013 non-null  int64  \n",
      " 7   BonusMalus  678013 non-null  int64  \n",
      " 8   VehBrand    678013 non-null  object \n",
      " 9   VehGas      678013 non-null  object \n",
      " 10  Density     678013 non-null  int64  \n",
      " 11  Region      678013 non-null  object \n",
      "dtypes: float64(2), int64(6), object(4)\n",
      "memory usage: 62.1+ MB\n"
     ]
    }
   ],
   "source": [
    "ds.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Categorical Variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 678013 entries, 0 to 678012\n",
      "Data columns (total 12 columns):\n",
      " #   Column      Non-Null Count   Dtype   \n",
      "---  ------      --------------   -----   \n",
      " 0   IDpol       678013 non-null  float64 \n",
      " 1   ClaimNb     678013 non-null  int64   \n",
      " 2   Exposure    678013 non-null  float64 \n",
      " 3   Area        678013 non-null  category\n",
      " 4   VehPower    678013 non-null  int64   \n",
      " 5   VehAge      678013 non-null  int64   \n",
      " 6   DrivAge     678013 non-null  int64   \n",
      " 7   BonusMalus  678013 non-null  int64   \n",
      " 8   VehBrand    678013 non-null  category\n",
      " 9   VehGas      678013 non-null  category\n",
      " 10  Density     678013 non-null  int64   \n",
      " 11  Region      678013 non-null  category\n",
      "dtypes: category(4), float64(2), int64(6)\n",
      "memory usage: 44.0 MB\n"
     ]
    }
   ],
   "source": [
    "# Transforming those variables into categorical\n",
    "cols = ['Area', 'VehBrand', 'VehGas','Region']\n",
    "ds[cols] = ds[cols].apply(lambda x:x.astype('category'))\n",
    "ds.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The number of levels from the Area, Vehicle Brand, Vehicle Gas and Region are 6, 11, 2 and 22, respectively.\n",
    "The Area’s levels are attributed alphabetically from A to F, with the areas C, D, E and A being the ones with the most observations. Concerning the Vehicle Brands, there are unknown and classified with the letter “B” followed by a number. The most observed Vehicle Brands are B12, B1 and B2. The Vehicle Gas classification is straightforward being determined as Regular or Diesel and is almost equally distributed with 48.99% being Diesel and the remaining as Regular. The Region is the feature with the highest levels (22). These are named with an “R” followed by a number, and according to the description, it follows the standard French Classification. The most observed regions among the policyholders are R24, R82, R93 and R11.\n",
    "It was further computed a graphical visualization of some relationships between variables, to gain further insights. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Area:\n",
      "C    191880\n",
      "D    151596\n",
      "E    137167\n",
      "A    103957\n",
      "B     75459\n",
      "F     17954\n",
      "Name: Area, dtype: int64\n",
      "VehBrand:\n",
      "B12    166024\n",
      "B1     162736\n",
      "B2     159861\n",
      "B3      53395\n",
      "B5      34753\n",
      "B6      28548\n",
      "B4      25179\n",
      "B10     17707\n",
      "B11     13585\n",
      "B13     12178\n",
      "B14      4047\n",
      "Name: VehBrand, dtype: int64\n",
      "VehGas:\n",
      "Regular    345877\n",
      "Diesel     332136\n",
      "Name: VehGas, dtype: int64\n",
      "Region:\n",
      "R24    160601\n",
      "R82     84752\n",
      "R93     79315\n",
      "R11     69791\n",
      "R53     42122\n",
      "R52     38751\n",
      "R91     35805\n",
      "R72     31329\n",
      "R31     27285\n",
      "R54     19046\n",
      "R73     17141\n",
      "R41     12990\n",
      "R25     10893\n",
      "R26     10492\n",
      "R23      8784\n",
      "R22      7994\n",
      "R83      5287\n",
      "R74      4567\n",
      "R94      4516\n",
      "R21      3026\n",
      "R42      2200\n",
      "R43      1326\n",
      "Name: Region, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Counts per category level\n",
    "ds['Area'].value_counts()\n",
    "ds['VehBrand'].value_counts()\n",
    "ds['VehGas'].value_counts()\n",
    "ds['Region'].value_counts()\n",
    "\n",
    "# Top Counts \n",
    "categorical = ['Area', 'VehBrand', 'VehGas','Region']\n",
    "categorical=ds.select_dtypes(exclude=[\"number\",\"bool_\",\"object_\"]).columns.tolist()  #select all variables excep number, bool and obj\n",
    "categorical\n",
    "for var in ds[categorical]:\n",
    "    print(var,\":\\n\",ds[var].value_counts(), sep=\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "48.98667134700957"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Percentage of Diesel presence in the Dataframe\n",
    "ds['VehGas'].value_counts().Diesel/678013*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Exploration of categorical variables\n",
    "\n",
    "#Region and Vehicle Brand\n",
    "tb1 = pd.crosstab(index=ds['Region'], columns=ds['VehBrand'])\n",
    "\n",
    "#Area and Vehicle Brand\n",
    "tb2 = pd.crosstab(index=ds['Area'], columns=ds['VehBrand'])\n",
    "\n",
    "#Region and ClaimNb\n",
    "tb3 = pd.crosstab(index=ds['Region'], columns=ds['ClaimNb'])\n",
    "\n",
    "#VehBrand and ClaimNb\n",
    "tb5 = pd.crosstab(index=ds['VehBrand'], columns=ds['ClaimNb'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='VehBrand'>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualization\n",
    "\n",
    "#Region and Vehicle Brand\n",
    "tb1.plot(kind=\"bar\", figsize=(8,8), stacked=True)\n",
    "\n",
    "#Area and Vehicle Brand\n",
    "tb2.plot(kind=\"bar\", figsize=(8,8), stacked=True)\n",
    "\n",
    "#Region and ClaimNb\n",
    "tb3.plot(kind=\"bar\", figsize=(8,8), stacked=True)\n",
    "\n",
    "#VehBrand and ClaimNb\n",
    "tb5.plot(kind=\"bar\", figsize=(8,8), stacked=True)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1st graphic: As one can observe, regions R24, R82, R94 and R11 are the areas with the most observations. Brands B12, B1 and B2 are the ones with more emphasis on the 1st graphic. There are many regions with lower observations, for instance below 20,000 which are possible candidates to group later."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2nd graphic: When comparing the areas with the vehicle brands, it is possible to infer that areas C, D and E are the most populated among the policyholders with vehicles of brands B1, B2 and B12. Area F is the one with the least observations which later could be also grouped with another region, possibly region B."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "3rd graphic: It is compared the number of claims with the policyholders’ regions. Region R24 is the region with the highest observations. There are also many regions with few observations. So, it is important to consolidate this graphic with the percentage of claims of each region. The region with the highest percentage of zero claims is R23 (96.7668%) but it has few observations and we can see on the graphic. Region R53 is the one with the lowest zero % claims (93.8868%) and it has a significant number of observations. Additionally, the region with the highest number of claims of an individual policyholder (16) is region R91 and with 11 individuals’ claims is again region R91 and also region R24."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "4th graphic: From this graphic brands B1, B12 and B2 are the ones with the most observations and probably with the highest presence of different claim numbers. By consolidating with the cross table, we can see that brand B1 is the brand with the largest number of claims per policyholder, followed by brand B2.  There’s also some brand with few observations that might be candidates for category grouping (B14, B13, B11, B10)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Numerical Variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 16 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "numerical=ds.select_dtypes(include=[np.number]).columns.tolist()\n",
    "ds[numerical].hist(bins=20, figsize=(15, 15), layout=(4, 4), xlabelsize=8, ylabelsize=8);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the histograms it is possible to get a glance of the numeric variables’ distributions.\n",
    "Clearly the predominant number of claims is zero, with few observations having one or more claims. This is a sign that the data is not balanced but also it makes sense given the context that this is an insurance company and the number of claims is usually way less than the number of policies. \n",
    "The vehicle power demonstrates a positive skew, as well as vehicle age, driver’s age, bonus malus and density. This is also in line with the median and mean values.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x504 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Do boxplots on all numeric variables\n",
    "fig, ax = plt.subplots(2, 4, figsize=(20, 7))\n",
    "for var, subplot in zip(ds[numerical], ax.flatten()):\n",
    "    sns.boxplot(x=ds[var], ax=subplot)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Boxplots provide insights regarding the possible existance of outliers for each variables. Using the ones plotted above (for the numerical variables) some of them clearly present outliers: ClaimNb, Exposure, VehPower, VehAge, DrivAge and BonusMalus. As for the density variable, higher density does not necessarily represent the existance of outliers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\lotar\\AppData\\Local\\Temp\\ipykernel_20548\\22763989.py:3: DeprecationWarning: `np.bool` is a deprecated alias for the builtin `bool`. To silence this warning, use `bool` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.bool_` here.\n",
      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
      "  mask = np.zeros_like(corr, dtype=np.bool)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Pearson Correlation only for numerical\n",
    "corr = ds[numerical].corr()\n",
    "mask = np.zeros_like(corr, dtype=np.bool)\n",
    "mask[np.triu_indices_from(mask)]= True\n",
    "\n",
    "f, ax = plt.subplots(figsize=(12, 12))\n",
    "heatmap = sns.heatmap(corr,\n",
    "                      mask = mask,\n",
    "                      square = True,\n",
    "                      linewidths = .5,\n",
    "                      cmap = 'coolwarm',\n",
    "                      vmin = -1,\n",
    "                      vmax = 1,\n",
    "                      annot = True,\n",
    "                      annot_kws = {'size': 10})\n",
    "\n",
    "#add the column names as labels\n",
    "ax.set_yticklabels(corr.columns, rotation = 0)\n",
    "ax.set_xticklabels(corr.columns, rotation = 0)\n",
    "sns.set_style({'xtick.bottom': True}, {'ytick.left': True})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using the figure above, where it is represented the pearson correlation between the numerical features, some pairs present relatively higher correlation between them when compared to others. For instance, BonusMalus and DrivAge (0,48 in absolut terms)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pair plot some of the numerical variables with Vehicle Brand\n",
    "\n",
    "#cols=['ClaimNb','Exposure','VehPower','VehAge','DrivAge','BonusMalus','Density','VehBrand']\n",
    "#g = sns.pairplot(ds[cols], hue=\"VehBrand\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Above follows the code to perform paiplots between all numerical variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='VehAge'>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1224x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1224x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Visualization\n",
    "\n",
    "#Driver's Age and ClaimNb\n",
    "tb4 = pd.crosstab(index=ds['DrivAge'], columns=ds['ClaimNb'])\n",
    "tb4.plot(kind=\"bar\", figsize=(17,7), stacked=True)\n",
    "\n",
    "#VehPower and ClaimNb\n",
    "\n",
    "tb6 = pd.crosstab(index=ds['VehPower'], columns=ds['ClaimNb'])\n",
    "tb6.plot(kind=\"bar\", figsize=(10,7), stacked=True)\n",
    "\n",
    "#VehAge and ClaimNb\n",
    "\n",
    "tb7 = pd.crosstab(index=ds['VehAge'], columns=ds['ClaimNb'])\n",
    "tb7.plot(kind=\"bar\", figsize=(17,7), stacked=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Preparation LR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a modeling dataset from the original dataset\n",
    "X = ds.copy(deep=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>unique</th>\n",
       "      <th>top</th>\n",
       "      <th>freq</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>IDpol</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2621856.921071</td>\n",
       "      <td>1641782.752655</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1157951.0</td>\n",
       "      <td>2272152.0</td>\n",
       "      <td>4046274.0</td>\n",
       "      <td>6114330.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ClaimNb</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.053247</td>\n",
       "      <td>0.240117</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>16.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Exposure</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.52875</td>\n",
       "      <td>0.364442</td>\n",
       "      <td>0.002732</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.99</td>\n",
       "      <td>2.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Area</th>\n",
       "      <td>678013</td>\n",
       "      <td>6</td>\n",
       "      <td>C</td>\n",
       "      <td>191880</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehPower</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.454631</td>\n",
       "      <td>2.050906</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>15.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehAge</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.044265</td>\n",
       "      <td>5.666232</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DrivAge</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>45.499122</td>\n",
       "      <td>14.137444</td>\n",
       "      <td>18.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BonusMalus</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>59.761502</td>\n",
       "      <td>15.636658</td>\n",
       "      <td>50.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>230.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehBrand</th>\n",
       "      <td>678013</td>\n",
       "      <td>11</td>\n",
       "      <td>B12</td>\n",
       "      <td>166024</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehGas</th>\n",
       "      <td>678013</td>\n",
       "      <td>2</td>\n",
       "      <td>Regular</td>\n",
       "      <td>345877</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Density</th>\n",
       "      <td>678013.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1792.422405</td>\n",
       "      <td>3958.646564</td>\n",
       "      <td>1.0</td>\n",
       "      <td>92.0</td>\n",
       "      <td>393.0</td>\n",
       "      <td>1658.0</td>\n",
       "      <td>27000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Region</th>\n",
       "      <td>678013</td>\n",
       "      <td>22</td>\n",
       "      <td>R24</td>\n",
       "      <td>160601</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               count unique      top    freq            mean             std  \\\n",
       "IDpol       678013.0    NaN      NaN     NaN  2621856.921071  1641782.752655   \n",
       "ClaimNb     678013.0    NaN      NaN     NaN        0.053247        0.240117   \n",
       "Exposure    678013.0    NaN      NaN     NaN         0.52875        0.364442   \n",
       "Area          678013      6        C  191880             NaN             NaN   \n",
       "VehPower    678013.0    NaN      NaN     NaN        6.454631        2.050906   \n",
       "VehAge      678013.0    NaN      NaN     NaN        7.044265        5.666232   \n",
       "DrivAge     678013.0    NaN      NaN     NaN       45.499122       14.137444   \n",
       "BonusMalus  678013.0    NaN      NaN     NaN       59.761502       15.636658   \n",
       "VehBrand      678013     11      B12  166024             NaN             NaN   \n",
       "VehGas        678013      2  Regular  345877             NaN             NaN   \n",
       "Density     678013.0    NaN      NaN     NaN     1792.422405     3958.646564   \n",
       "Region        678013     22      R24  160601             NaN             NaN   \n",
       "\n",
       "                 min        25%        50%        75%        max  \n",
       "IDpol            1.0  1157951.0  2272152.0  4046274.0  6114330.0  \n",
       "ClaimNb          0.0        0.0        0.0        0.0       16.0  \n",
       "Exposure    0.002732       0.18       0.49       0.99       2.01  \n",
       "Area             NaN        NaN        NaN        NaN        NaN  \n",
       "VehPower         4.0        5.0        6.0        7.0       15.0  \n",
       "VehAge           0.0        2.0        6.0       11.0      100.0  \n",
       "DrivAge         18.0       34.0       44.0       55.0      100.0  \n",
       "BonusMalus      50.0       50.0       50.0       64.0      230.0  \n",
       "VehBrand         NaN        NaN        NaN        NaN        NaN  \n",
       "VehGas           NaN        NaN        NaN        NaN        NaN  \n",
       "Density          1.0       92.0      393.0     1658.0    27000.0  \n",
       "Region           NaN        NaN        NaN        NaN        NaN  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.describe(include='all').T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      VehPower\n",
      "0.05       4.0\n",
      "0.95      11.0\n",
      "      DrivAge\n",
      "0.00     18.0\n",
      "0.99     80.0\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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>IDpol</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>2.608857e+06</td>\n",
       "      <td>1.639082e+06</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1154843.00</td>\n",
       "      <td>2269955.00</td>\n",
       "      <td>4041436.00</td>\n",
       "      <td>6114330.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ClaimNb</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>5.318536e-02</td>\n",
       "      <td>2.378317e-01</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>6.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Exposure</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>5.279595e-01</td>\n",
       "      <td>3.642680e-01</td>\n",
       "      <td>0.002732</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.99</td>\n",
       "      <td>2.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehPower</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>6.294019e+00</td>\n",
       "      <td>1.780453e+00</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>5.00</td>\n",
       "      <td>6.00</td>\n",
       "      <td>7.00</td>\n",
       "      <td>11.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehAge</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>7.027170e+00</td>\n",
       "      <td>5.510512e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.00</td>\n",
       "      <td>6.00</td>\n",
       "      <td>11.00</td>\n",
       "      <td>35.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DrivAge</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>4.510123e+01</td>\n",
       "      <td>1.372731e+01</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>34.00</td>\n",
       "      <td>44.00</td>\n",
       "      <td>54.00</td>\n",
       "      <td>80.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BonusMalus</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>5.989299e+01</td>\n",
       "      <td>1.568537e+01</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>50.00</td>\n",
       "      <td>50.00</td>\n",
       "      <td>66.00</td>\n",
       "      <td>163.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Density</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>1.771515e+03</td>\n",
       "      <td>3.899493e+03</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>92.00</td>\n",
       "      <td>392.00</td>\n",
       "      <td>1629.00</td>\n",
       "      <td>27000.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               count          mean           std        min         25%  \\\n",
       "IDpol       654729.0  2.608857e+06  1.639082e+06   1.000000  1154843.00   \n",
       "ClaimNb     654729.0  5.318536e-02  2.378317e-01   0.000000        0.00   \n",
       "Exposure    654729.0  5.279595e-01  3.642680e-01   0.002732        0.17   \n",
       "VehPower    654729.0  6.294019e+00  1.780453e+00   4.000000        5.00   \n",
       "VehAge      654729.0  7.027170e+00  5.510512e+00   0.000000        2.00   \n",
       "DrivAge     654729.0  4.510123e+01  1.372731e+01  18.000000       34.00   \n",
       "BonusMalus  654729.0  5.989299e+01  1.568537e+01  50.000000       50.00   \n",
       "Density     654729.0  1.771515e+03  3.899493e+03   1.000000       92.00   \n",
       "\n",
       "                   50%         75%         max  \n",
       "IDpol       2269955.00  4041436.00  6114330.00  \n",
       "ClaimNb           0.00        0.00        6.00  \n",
       "Exposure          0.49        0.99        2.01  \n",
       "VehPower          6.00        7.00       11.00  \n",
       "VehAge            6.00       11.00       35.00  \n",
       "DrivAge          44.00       54.00       80.00  \n",
       "BonusMalus       50.00       66.00      163.00  \n",
       "Density         392.00     1629.00    27000.00  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Find and remove outliers\n",
    "\n",
    "# Function to find outliers on columns based on percentile\n",
    "def removeOutliers(df, colList, lowPercentile=0.05, highPercentile=0.95, verbose=False):\n",
    "    quant_df = df[colList].quantile([lowPercentile, highPercentile])\n",
    "    if verbose:\n",
    "        print(quant_df)\n",
    "    for name in list(df[colList].columns):\n",
    "        df = df[(df[name] >= quant_df.loc[lowPercentile, name]) & (df[name] <= quant_df.loc[highPercentile, name])]\n",
    "    return df\n",
    "\n",
    "# Claim Number\n",
    "X.drop(X[X.ClaimNb > 7].index, inplace=True)\n",
    "\n",
    "# Exposure\n",
    "X.loc[(X['Exposure'] > 1)] == 1\n",
    "\n",
    "# Vehicle Power\n",
    "cols = ['VehPower']\n",
    "X = removeOutliers(X, colList=cols , verbose=True)\n",
    "\n",
    "# Vehicle Age\n",
    "X.drop(X[X.VehAge > 35].index, inplace=True)\n",
    "\n",
    "\n",
    "# Driver's Age\n",
    "cols = ['DrivAge']\n",
    "X = removeOutliers(X, colList=cols, lowPercentile=0.00, highPercentile=0.99, verbose=True)\n",
    "\n",
    "# Bonus Malus\n",
    "X.drop(X[X.BonusMalus > ((max(X.BonusMalus)-100)/2)+100].index, inplace=True)\n",
    "\n",
    "X.describe().T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "DriveAge: As the goal is to infer the number of claims per year, elderly people have higher claim frequency and thus the 1% and 99% percentiles were considered as outliers.\n",
    "\n",
    "Area: As for this variable, higher values do not necessarily represent the existance of outliers since it is correlated with the number of inhabitants.\n",
    "\n",
    "Exposure: The default exposure is one and thus values higher than this were considered as errors and assumed to be 1.\n",
    "\n",
    "ClaimNb: Using the boxplots, number of claims per year higher than 7 were considered as outliers as the number os instances with these values was low.\n",
    "\n",
    "VehAge: The older vehicles in this data set were considered outliers, since they can represent an old car collection or even a museum's collection since those vehicles still need insurance, but for the scope of our project these vehicles are not relevant to number of claims. Based on the graph, the outliers were considered for vehicles with age above 35 years. \n",
    "\n",
    "Driver’s Age: The top and bottom 1% were considered as outliers.\n",
    "\n",
    "Bonus Malus: Taking into consideration that values above 100 are considered \"Malus\" and the maximum value in our sample is 230, observations higher than the intermediate point between those (163) were considered outliers.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.0343415244250479"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Number of removed observations, after outliers\n",
    "diff = X.shape[0]-ds.shape[0]\n",
    "diff/ds.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='VehPower'>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Group Region\n",
    "\n",
    "X['Region'] = X['Region'].replace(['R22','R23','R25','R26','R41','R42','R43','R74','R83','R94','R54','R73'],'R21')\n",
    "\n",
    "#Plot\n",
    "tb3_1 = pd.crosstab(index=X['Region'], columns=X['ClaimNb'])\n",
    "tb3_1.plot(kind=\"bar\", figsize=(8,8), stacked=True)\n",
    "\n",
    "# Bin the Driver's age\n",
    "binsInterval = [17, 30, 52, 75, 81] # Create bins interval based on distribution/quartiles\n",
    "intervalLabels = ['(18, 30]', '(31, 52]',  '(53, 75]', '(76, 80]']\n",
    "X['DrivAge_binned'] = pd.cut(X['DrivAge'], bins = binsInterval, labels=intervalLabels)\n",
    "\n",
    "tb4_1 = pd.crosstab(index=X['DrivAge_binned'], columns=X['ClaimNb'])\n",
    "tb4_1.plot(kind=\"bar\", figsize=(8,8), stacked=True)\n",
    "\n",
    "#Group Vehicle Brand\n",
    "\n",
    "X['VehBrand'] = X['VehBrand'].replace(['B10','B11','B13'],'B14')\n",
    "X['VehBrand'] = X['VehBrand'].replace(['B4','B5'],'B6')\n",
    "\n",
    "\n",
    "#Plot\n",
    "tb5_1 = pd.crosstab(index=X['VehBrand'], columns=X['ClaimNb'])\n",
    "tb5_1.plot(kind=\"bar\", figsize=(8,8), stacked=True)\n",
    "\n",
    "#Group Vehicle Power\n",
    "\n",
    "X['VehPower'] = X['VehPower'].replace(['11','12','13','14'],'15')\n",
    "\n",
    "#Plot\n",
    "tb6_1 = pd.crosstab(index=X['VehPower'], columns=X['ClaimNb'])\n",
    "tb6_1.plot(kind=\"bar\", figsize=(8,8), stacked=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Region:\n",
    "The regions with less claim observations were aggregated, this includes zones with less than 20 000 observations, maintaining the regions with bigger claim quantities. \n",
    "\n",
    "Driver’s Age:\n",
    "Based on the plot, we can identify four stages: growth, stable, descending and residual. We grouped Driver’s Age from age 18 to 30 (young adults), 31 to 52 (adults), 53 to 75 (mature) and above 75 (elderly). \n",
    "\n",
    "Vehicle brand:\n",
    "The vehicle brands were grouped according to the claims observations which brand had, meaning brands with less than 20 000 claim observations and the brands with less than 40 000 were grouped together. \n",
    "\n",
    "Vehicle Power:\n",
    "The vehicle power categories with less than 20 000 claim observations were grouped together.\n",
    "\n",
    "The IDpol variable was eliminated because it does not explain the number of claims.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DrivAge_binned\n",
       "(18, 30]    102894\n",
       "(31, 52]    359645\n",
       "(53, 75]    179684\n",
       "(76, 80]     12506\n",
       "Name: ClaimNb, dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.groupby('DrivAge_binned')['ClaimNb'].agg('count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Grouping Variables according p-values below\n",
    "\n",
    "# Group Region\n",
    "\n",
    "X['Region'] = X['Region'].replace(['R53'],'R24')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After computing the regression itself, using the values obtained for the p-value of each variable, we reached to the conclusion that Region 53 and Region 24 should be group to better explain the target variable (claim frequency)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>IDpol</th>\n",
       "      <th>ClaimNb</th>\n",
       "      <th>Exposure</th>\n",
       "      <th>Area_D</th>\n",
       "      <th>Area_B</th>\n",
       "      <th>Area_E</th>\n",
       "      <th>Area_C</th>\n",
       "      <th>Area_F</th>\n",
       "      <th>Area_A</th>\n",
       "      <th>VehPower</th>\n",
       "      <th>...</th>\n",
       "      <th>Region_R31</th>\n",
       "      <th>Region_R91</th>\n",
       "      <th>Region_R52</th>\n",
       "      <th>Region_R93</th>\n",
       "      <th>Region_R11</th>\n",
       "      <th>Region_R24</th>\n",
       "      <th>DrivAge_binned_(18, 30]</th>\n",
       "      <th>DrivAge_binned_(31, 52]</th>\n",
       "      <th>DrivAge_binned_(53, 75]</th>\n",
       "      <th>DrivAge_binned_(76, 80]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.10000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.77000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.75000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.09000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.84000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678008</th>\n",
       "      <td>6114326.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00274</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678009</th>\n",
       "      <td>6114327.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00274</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678010</th>\n",
       "      <td>6114328.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00274</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678011</th>\n",
       "      <td>6114329.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00274</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678012</th>\n",
       "      <td>6114330.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00274</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>654729 rows × 35 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            IDpol  ClaimNb  Exposure  Area_D  Area_B  Area_E  Area_C  Area_F  \\\n",
       "0             1.0        1   0.10000       1       0       0       0       0   \n",
       "1             3.0        1   0.77000       1       0       0       0       0   \n",
       "2             5.0        1   0.75000       0       1       0       0       0   \n",
       "3            10.0        1   0.09000       0       1       0       0       0   \n",
       "4            11.0        1   0.84000       0       1       0       0       0   \n",
       "...           ...      ...       ...     ...     ...     ...     ...     ...   \n",
       "678008  6114326.0        0   0.00274       0       0       1       0       0   \n",
       "678009  6114327.0        0   0.00274       0       0       1       0       0   \n",
       "678010  6114328.0        0   0.00274       1       0       0       0       0   \n",
       "678011  6114329.0        0   0.00274       0       1       0       0       0   \n",
       "678012  6114330.0        0   0.00274       0       1       0       0       0   \n",
       "\n",
       "        Area_A  VehPower  ...  Region_R31  Region_R91  Region_R52  Region_R93  \\\n",
       "0            0         5  ...           0           0           0           0   \n",
       "1            0         5  ...           0           0           0           0   \n",
       "2            0         6  ...           0           0           0           0   \n",
       "3            0         7  ...           0           0           0           0   \n",
       "4            0         7  ...           0           0           0           0   \n",
       "...        ...       ...  ...         ...         ...         ...         ...   \n",
       "678008       0         4  ...           0           0           0           1   \n",
       "678009       0         4  ...           0           0           0           0   \n",
       "678010       0         6  ...           0           0           0           0   \n",
       "678011       0         4  ...           0           0           0           0   \n",
       "678012       0         7  ...           0           0           0           0   \n",
       "\n",
       "        Region_R11  Region_R24  DrivAge_binned_(18, 30]  \\\n",
       "0                0           0                        0   \n",
       "1                0           0                        0   \n",
       "2                0           0                        0   \n",
       "3                0           0                        0   \n",
       "4                0           0                        0   \n",
       "...            ...         ...                      ...   \n",
       "678008           0           0                        0   \n",
       "678009           1           0                        0   \n",
       "678010           0           0                        0   \n",
       "678011           0           0                        0   \n",
       "678012           0           0                        1   \n",
       "\n",
       "        DrivAge_binned_(31, 52]  DrivAge_binned_(53, 75]  \\\n",
       "0                             0                        1   \n",
       "1                             0                        1   \n",
       "2                             1                        0   \n",
       "3                             1                        0   \n",
       "4                             1                        0   \n",
       "...                         ...                      ...   \n",
       "678008                        0                        1   \n",
       "678009                        1                        0   \n",
       "678010                        1                        0   \n",
       "678011                        0                        1   \n",
       "678012                        0                        0   \n",
       "\n",
       "        DrivAge_binned_(76, 80]  \n",
       "0                             0  \n",
       "1                             0  \n",
       "2                             0  \n",
       "3                             0  \n",
       "4                             0  \n",
       "...                         ...  \n",
       "678008                        0  \n",
       "678009                        0  \n",
       "678010                        0  \n",
       "678011                        0  \n",
       "678012                        0  \n",
       "\n",
       "[654729 rows x 35 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Encode categorical values into dummy variables\n",
    "cols = ['Area', 'VehBrand', 'VehGas', 'Region','DrivAge_binned']\n",
    "ce_one_hot = ce.OneHotEncoder(cols = cols, use_cat_names=True)\n",
    "X = ce_one_hot.fit_transform(X)\n",
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here, the categorical variables were divided into dummy variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['DrivAge_binned_(18, 30]', 'DrivAge_binned_(31, 52]',\n",
       "       'DrivAge_binned_(53, 75]', 'DrivAge_binned_(76, 80]', 'VehPower',\n",
       "       'VehAge', 'DrivAge', 'BonusMalus', 'Density'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Create the target (Claim number per year)\n",
    "y = X['ClaimNb']/X['Exposure']\n",
    "\n",
    "# Drop not used columns\n",
    "cols = ['IDpol', 'ClaimNb', 'Exposure','VehPower','VehAge','DrivAge','BonusMalus','Density']\n",
    "X1 = X.drop(columns=cols)\n",
    "\n",
    "# Normalize all columns\n",
    "# MinMax scaler returns an array, so the dataframe must be recreated\n",
    "X1 = pd.DataFrame(preprocessing.MinMaxScaler().fit_transform(X1.values), columns=X1.columns,index=X1.index)\n",
    "\n",
    "for cols in cols:\n",
    "    X1[cols]=X[cols]\n",
    "\n",
    "\n",
    "\n",
    "X=X1\n",
    "\n",
    "X.drop(\"ClaimNb\", axis=1, inplace=True)  #Remove Claim Nb which is the target variable\n",
    "X.drop(\"IDpol\", axis=1, inplace=True)    #Remove IDpolicy since logicaly doesn't explain the number of claims\n",
    "X.drop(\"Exposure\", axis=1, inplace=True)  #Remove Exposure which is in the target variable\n",
    "\n",
    "\n",
    "#P-Value - removing variables\n",
    "\n",
    "   \n",
    "X.drop(\"Region_R82\", axis=1, inplace=True)   \n",
    "X.drop(\"Region_R21\", axis=1, inplace=True)\n",
    "X.drop(\"Region_R72\", axis=1, inplace=True)\n",
    "X.drop(\"Region_R31\", axis=1, inplace=True)\n",
    "X.drop(\"Region_R91\", axis=1, inplace=True)\n",
    "X.drop(\"Region_R52\", axis=1, inplace=True)\n",
    "X.drop(\"Region_R93\", axis=1, inplace=True)\n",
    "X.drop(\"Region_R11\", axis=1, inplace=True)\n",
    "X.drop(\"Region_R24\", axis=1, inplace=True)\n",
    "\n",
    "X.drop(\"Area_D\", axis=1, inplace=True)    \n",
    "X.drop(\"Area_B\", axis=1, inplace=True) \n",
    "X.drop(\"Area_E\", axis=1, inplace=True) \n",
    "X.drop(\"Area_C\", axis=1, inplace=True) \n",
    "X.drop(\"Area_F\", axis=1, inplace=True) \n",
    "X.drop(\"Area_A\", axis=1, inplace=True) \n",
    "\n",
    "X.drop(\"VehBrand_B12\", axis=1, inplace=True) \n",
    "X.drop(\"VehBrand_B6\", axis=1, inplace=True) \n",
    "X.drop(\"VehBrand_B3\", axis=1, inplace=True) \n",
    "X.drop(\"VehBrand_B2\", axis=1, inplace=True) \n",
    "X.drop(\"VehBrand_B14\", axis=1, inplace=True) \n",
    "X.drop(\"VehBrand_B1\", axis=1, inplace=True) \n",
    "\n",
    "X.drop(\"VehGas_Regular\", axis=1, inplace=True) \n",
    "X.drop(\"VehGas_Diesel\", axis=1, inplace=True) \n",
    "\n",
    "\n",
    "X.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After running the model for the first time, we decided to evaluate the explanatory power of each variable using the obtained p-value for each. We reached to the conclusion that some of these variables, according to this measure, were not statistically significant to explain the target variable. Thus, we proceeded to iteratively eliminate those, analysing the new p-values each time.\n",
    "By the end of all iterations, we decided to drop variables Region, Area, VehBrand and VehGas."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modeling LR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split the dataset intro train and test\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, train_size = 0.75, test_size = 0.25, random_state = 52)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\lotar\\anaconda3\\lib\\site-packages\\statsmodels\\tsa\\base\\tsa_model.py:7: FutureWarning: pandas.Int64Index is deprecated and will be removed from pandas in a future version. Use pandas.Index with the appropriate dtype instead.\n",
      "  from pandas import (to_datetime, Int64Index, DatetimeIndex, Period,\n",
      "C:\\Users\\lotar\\anaconda3\\lib\\site-packages\\statsmodels\\tsa\\base\\tsa_model.py:7: FutureWarning: pandas.Float64Index is deprecated and will be removed from pandas in a future version. Use pandas.Index with the appropriate dtype instead.\n",
      "  from pandas import (to_datetime, Int64Index, DatetimeIndex, Period,\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>            <td>y</td>        <th>  R-squared:         </th>  <td>   0.001</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th>  <td>   0.001</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th>  <td>   59.66</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Wed, 25 May 2022</td> <th>  Prob (F-statistic):</th>  <td>5.82e-98</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>21:50:25</td>     <th>  Log-Likelihood:    </th> <td>-1.4191e+06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>491046</td>      <th>  AIC:               </th>  <td>2.838e+06</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>491037</td>      <th>  BIC:               </th>  <td>2.838e+06</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     8</td>      <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>      <td> </td>     \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "             <td></td>                <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>DrivAge_binned_(18, 30]</th> <td>   -0.0258</td> <td>    0.060</td> <td>   -0.430</td> <td> 0.667</td> <td>   -0.144</td> <td>    0.092</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>DrivAge_binned_(31, 52]</th> <td>   -0.0196</td> <td>    0.064</td> <td>   -0.307</td> <td> 0.759</td> <td>   -0.145</td> <td>    0.105</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>DrivAge_binned_(53, 75]</th> <td>   -0.0459</td> <td>    0.081</td> <td>   -0.566</td> <td> 0.571</td> <td>   -0.205</td> <td>    0.113</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>DrivAge_binned_(76, 80]</th> <td>   -0.1068</td> <td>    0.106</td> <td>   -1.005</td> <td> 0.315</td> <td>   -0.315</td> <td>    0.102</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>VehPower</th>                <td>   -0.0002</td> <td>    0.004</td> <td>   -0.062</td> <td> 0.950</td> <td>   -0.007</td> <td>    0.007</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>VehAge</th>                  <td>   -0.0209</td> <td>    0.001</td> <td>  -18.324</td> <td> 0.000</td> <td>   -0.023</td> <td>   -0.019</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>DrivAge</th>                 <td>    0.0022</td> <td>    0.001</td> <td>    2.029</td> <td> 0.042</td> <td> 7.53e-05</td> <td>    0.004</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>BonusMalus</th>              <td>    0.0057</td> <td>    0.000</td> <td>   11.533</td> <td> 0.000</td> <td>    0.005</td> <td>    0.007</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Density</th>                 <td> 1.891e-06</td> <td> 1.62e-06</td> <td>    1.171</td> <td> 0.242</td> <td>-1.28e-06</td> <td> 5.06e-06</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>1578763.430</td> <th>  Durbin-Watson:     </th>     <td>   2.001</td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th>   <td> 0.000</td>    <th>  Jarque-Bera (JB):  </th> <td>291973469177.910</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>            <td>53.390</td>    <th>  Prob(JB):          </th>     <td>    0.00</td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>       <td>3779.093</td>   <th>  Cond. No.          </th>     <td>1.05e+05</td>    \n",
       "</tr>\n",
       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 1.05e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                      y   R-squared:                       0.001\n",
       "Model:                            OLS   Adj. R-squared:                  0.001\n",
       "Method:                 Least Squares   F-statistic:                     59.66\n",
       "Date:                Wed, 25 May 2022   Prob (F-statistic):           5.82e-98\n",
       "Time:                        21:50:25   Log-Likelihood:            -1.4191e+06\n",
       "No. Observations:              491046   AIC:                         2.838e+06\n",
       "Df Residuals:                  491037   BIC:                         2.838e+06\n",
       "Df Model:                           8                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===========================================================================================\n",
       "                              coef    std err          t      P>|t|      [0.025      0.975]\n",
       "-------------------------------------------------------------------------------------------\n",
       "DrivAge_binned_(18, 30]    -0.0258      0.060     -0.430      0.667      -0.144       0.092\n",
       "DrivAge_binned_(31, 52]    -0.0196      0.064     -0.307      0.759      -0.145       0.105\n",
       "DrivAge_binned_(53, 75]    -0.0459      0.081     -0.566      0.571      -0.205       0.113\n",
       "DrivAge_binned_(76, 80]    -0.1068      0.106     -1.005      0.315      -0.315       0.102\n",
       "VehPower                   -0.0002      0.004     -0.062      0.950      -0.007       0.007\n",
       "VehAge                     -0.0209      0.001    -18.324      0.000      -0.023      -0.019\n",
       "DrivAge                     0.0022      0.001      2.029      0.042    7.53e-05       0.004\n",
       "BonusMalus                  0.0057      0.000     11.533      0.000       0.005       0.007\n",
       "Density                  1.891e-06   1.62e-06      1.171      0.242   -1.28e-06    5.06e-06\n",
       "==============================================================================\n",
       "Omnibus:                  1578763.430   Durbin-Watson:                   2.001\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):     291973469177.910\n",
       "Skew:                          53.390   Prob(JB):                         0.00\n",
       "Kurtosis:                    3779.093   Cond. No.                     1.05e+05\n",
       "==============================================================================\n",
       "\n",
       "Notes:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 1.05e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#P-Value\n",
    "\n",
    "from statsmodels.api import OLS\n",
    "\n",
    "lr_regr = linear_model.LinearRegression()\n",
    "\n",
    "OLS(y_train,X_train).fit().summary()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This summary statistics was relevant to determine which variables to drop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'RFECV for LinearRegression'}, xlabel='Number of Features Selected', ylabel='Score'>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Initialize model\n",
    "lr_regr = linear_model.LinearRegression()\n",
    "\n",
    "# Apply RFECV\n",
    "vis = RFECV(lr_regr, cv=5, scoring='r2')\n",
    "vis.fit(X_train, y_train)\n",
    "vis.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The RFECV graph above allows us to conclude that:\n",
    "    - After 4 features, the score increases almost exponetially (corresponds to the VehAge variable). After 6 features the evolution of the score stabilizes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['DrivAge_binned_(18, 30]', 'DrivAge_binned_(31, 52]',\n",
       "       'DrivAge_binned_(53, 75]', 'DrivAge_binned_(76, 80]', 'VehAge',\n",
       "       'DrivAge', 'BonusMalus'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Features to include (all in this case)\n",
    "X.columns[vis.support_]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Learning Curve for LinearRegression'}, xlabel='Training Instances', ylabel='Score'>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Plot the learning curve\n",
    "cv = 10  #cross validation\n",
    "sizes = np.linspace(0.1, 1.0, 5)\n",
    "visualizer = LearningCurve(estimator=lr_regr, cv=cv, scoring='r2', train_sizes=sizes, n_jobs=4, \n",
    "                           random_state=123)\n",
    "visualizer.fit(X, y)\n",
    "visualizer.show()     "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using the Learning Curve graph above, one can observe that, as the training instances increases, the gap between the training score and the cross validation score decreases. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression()"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create and train the model\n",
    "lr_regr.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Intercept: -0.04953304852394241 \n",
      "\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>Feature</th>\n",
       "      <th>Coefficient</th>\n",
       "      <th>AbsCoefficient</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DrivAge_binned_(76, 80]</td>\n",
       "      <td>-0.057233</td>\n",
       "      <td>0.057233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DrivAge_binned_(31, 52]</td>\n",
       "      <td>0.029915</td>\n",
       "      <td>0.029915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DrivAge_binned_(18, 30]</td>\n",
       "      <td>0.023694</td>\n",
       "      <td>0.023694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>VehAge</td>\n",
       "      <td>-0.020872</td>\n",
       "      <td>0.020872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>BonusMalus</td>\n",
       "      <td>0.005684</td>\n",
       "      <td>0.005684</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DrivAge_binned_(53, 75]</td>\n",
       "      <td>0.003624</td>\n",
       "      <td>0.003624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>DrivAge</td>\n",
       "      <td>0.002199</td>\n",
       "      <td>0.002199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>VehPower</td>\n",
       "      <td>-0.000219</td>\n",
       "      <td>0.000219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Density</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000002</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Feature Coefficient  AbsCoefficient\n",
       "3  DrivAge_binned_(76, 80]   -0.057233        0.057233\n",
       "1  DrivAge_binned_(31, 52]    0.029915        0.029915\n",
       "0  DrivAge_binned_(18, 30]    0.023694        0.023694\n",
       "5                   VehAge   -0.020872        0.020872\n",
       "7               BonusMalus    0.005684        0.005684\n",
       "2  DrivAge_binned_(53, 75]    0.003624        0.003624\n",
       "6                  DrivAge    0.002199        0.002199\n",
       "4                 VehPower   -0.000219        0.000219\n",
       "8                  Density    0.000002        0.000002"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Print the Intercept\n",
    "print('Intercept:',lr_regr.intercept_, '\\n')\n",
    "\n",
    "# Print Coefficients by descending \"importance\"\n",
    "formattedList = [format(float(i),'.6f') for i in lr_regr.coef_]\n",
    "formattedList2 = [format(abs(float(i)),'.6f') for i in lr_regr.coef_]\n",
    "data_tuples = list(zip(X.columns,formattedList,formattedList2))\n",
    "coeff_df = pd.DataFrame(data=data_tuples, columns=['Feature','Coefficient','AbsCoefficient'])\n",
    "coeff_df['AbsCoefficient']=coeff_df['AbsCoefficient'].astype(float)\n",
    "coeff_df.sort_values('AbsCoefficient', axis=0, inplace=True, ascending=False)\n",
    "coeff_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluation LR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# predict y for X_train and X_test\n",
    "y_pred_train = lr_regr.predict(X_train) \n",
    "y_pred_test = lr_regr.predict(X_test) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to create dataframe with metrics\n",
    "def performanceMetricsDF(metricsObj, yTrain, yPredTrain, yTest, yPredTest):\n",
    "  measures_list = ['MAE','RMSE', 'R^2','MAPE (%)','MAX Error']\n",
    "  train_results = [metricsObj.mean_absolute_error(yTrain, yPredTrain),\n",
    "                np.sqrt(metricsObj.mean_squared_error(yTrain, yPredTrain)),\n",
    "                metricsObj.r2_score(yTrain, yPredTrain),\n",
    "                metricsObj.mean_absolute_percentage_error(yTrain, yPredTrain),\n",
    "                metricsObj.max_error(yTrain, yPredTrain)]\n",
    "  test_results = [metricsObj.mean_absolute_error(yTest, yPredTest),\n",
    "                np.sqrt(metricsObj.mean_squared_error(yTest, yPredTest)),\n",
    "                metricsObj.r2_score(yTest, yPredTest),\n",
    "                  metricsObj.mean_absolute_percentage_error(yTest, yPredTest),\n",
    "                metricsObj.max_error(yTest, yPredTest)]\n",
    "  resultsDF = pd.DataFrame({'Measure': measures_list, 'Train': train_results, 'Test':test_results})\n",
    "  return(resultsDF)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>Measure</th>\n",
       "      <th>Train</th>\n",
       "      <th>Test</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MAE</td>\n",
       "      <td>5.038642e-01</td>\n",
       "      <td>5.056426e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>RMSE</td>\n",
       "      <td>4.353633e+00</td>\n",
       "      <td>5.383849e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>R^2</td>\n",
       "      <td>9.710834e-04</td>\n",
       "      <td>6.752516e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>MAPE (%)</td>\n",
       "      <td>1.141845e+15</td>\n",
       "      <td>1.144795e+15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>MAX Error</td>\n",
       "      <td>3.657472e+02</td>\n",
       "      <td>7.316586e+02</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Measure         Train          Test\n",
       "0        MAE  5.038642e-01  5.056426e-01\n",
       "1       RMSE  4.353633e+00  5.383849e+00\n",
       "2        R^2  9.710834e-04  6.752516e-04\n",
       "3   MAPE (%)  1.141845e+15  1.144795e+15\n",
       "4  MAX Error  3.657472e+02  7.316586e+02"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show performance results\n",
    "resultsDF = performanceMetricsDF(metrics, y_train, y_pred_train, y_test, y_pred_test)\n",
    "resultsDF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\lotar\\anaconda3\\lib\\site-packages\\sklearn\\base.py:450: UserWarning: X does not have valid feature names, but LinearRegression was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Residuals for LinearRegression Model'}, xlabel='Predicted Value', ylabel='Residuals'>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Visualize the residuals\n",
    "residuals_vis = ResidualsPlot(lr_regr)\n",
    "residuals_vis.fit(X_train, y_train)  # Fit the training data to the visualizer\n",
    "residuals_vis.score(X_test, y_test)  # Evaluate the model on the test data\n",
    "residuals_vis.show() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\lotar\\anaconda3\\lib\\site-packages\\sklearn\\base.py:450: UserWarning: X does not have valid feature names, but LinearRegression was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Prediction Error for LinearRegression'}, xlabel='$y$', ylabel='$\\\\hat{y}$'>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Visualize the prediction error\n",
    "predError_vis = PredictionError(lr_regr)\n",
    "predError_vis.fit(X_train, y_train)  # Fit the training data to the visualizer\n",
    "predError_vis.score(X_test, y_test)  # Evaluate the model on the test data\n",
    "predError_vis.show() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Despite the fact that the results obtained by the model in the training set are similar to the ones obtained using the test set, all obtained measures show us that the model predictive power can be considered weak. One possible reason for such results is the fact that the initial dataset is extremely inbalanced (many more observations reporting zero claims than higher than zero)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Preparation DT"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Data Preparation process for the Decision tree followed the same rationale used before, for the linear regression model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a modeling dataset from the original dataset\n",
    "X = ds.copy(deep=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      VehPower\n",
      "0.05       4.0\n",
      "0.95      11.0\n",
      "      DrivAge\n",
      "0.00     18.0\n",
      "0.99     80.0\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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>IDpol</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>2.608857e+06</td>\n",
       "      <td>1.639082e+06</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1154843.00</td>\n",
       "      <td>2269955.00</td>\n",
       "      <td>4041436.00</td>\n",
       "      <td>6114330.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ClaimNb</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>5.318536e-02</td>\n",
       "      <td>2.378317e-01</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>6.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Exposure</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>5.279595e-01</td>\n",
       "      <td>3.642680e-01</td>\n",
       "      <td>0.002732</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.99</td>\n",
       "      <td>2.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehPower</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>6.294019e+00</td>\n",
       "      <td>1.780453e+00</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>5.00</td>\n",
       "      <td>6.00</td>\n",
       "      <td>7.00</td>\n",
       "      <td>11.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VehAge</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>7.027170e+00</td>\n",
       "      <td>5.510512e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.00</td>\n",
       "      <td>6.00</td>\n",
       "      <td>11.00</td>\n",
       "      <td>35.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DrivAge</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>4.510123e+01</td>\n",
       "      <td>1.372731e+01</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>34.00</td>\n",
       "      <td>44.00</td>\n",
       "      <td>54.00</td>\n",
       "      <td>80.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BonusMalus</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>5.989299e+01</td>\n",
       "      <td>1.568537e+01</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>50.00</td>\n",
       "      <td>50.00</td>\n",
       "      <td>66.00</td>\n",
       "      <td>163.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Density</th>\n",
       "      <td>654729.0</td>\n",
       "      <td>1.771515e+03</td>\n",
       "      <td>3.899493e+03</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>92.00</td>\n",
       "      <td>392.00</td>\n",
       "      <td>1629.00</td>\n",
       "      <td>27000.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               count          mean           std        min         25%  \\\n",
       "IDpol       654729.0  2.608857e+06  1.639082e+06   1.000000  1154843.00   \n",
       "ClaimNb     654729.0  5.318536e-02  2.378317e-01   0.000000        0.00   \n",
       "Exposure    654729.0  5.279595e-01  3.642680e-01   0.002732        0.17   \n",
       "VehPower    654729.0  6.294019e+00  1.780453e+00   4.000000        5.00   \n",
       "VehAge      654729.0  7.027170e+00  5.510512e+00   0.000000        2.00   \n",
       "DrivAge     654729.0  4.510123e+01  1.372731e+01  18.000000       34.00   \n",
       "BonusMalus  654729.0  5.989299e+01  1.568537e+01  50.000000       50.00   \n",
       "Density     654729.0  1.771515e+03  3.899493e+03   1.000000       92.00   \n",
       "\n",
       "                   50%         75%         max  \n",
       "IDpol       2269955.00  4041436.00  6114330.00  \n",
       "ClaimNb           0.00        0.00        6.00  \n",
       "Exposure          0.49        0.99        2.01  \n",
       "VehPower          6.00        7.00       11.00  \n",
       "VehAge            6.00       11.00       35.00  \n",
       "DrivAge          44.00       54.00       80.00  \n",
       "BonusMalus       50.00       66.00      163.00  \n",
       "Density         392.00     1629.00    27000.00  "
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Find and remove outliers\n",
    "\n",
    "# Function to find outliers on columns based on percentile\n",
    "def removeOutliers(df, colList, lowPercentile=0.05, highPercentile=0.95, verbose=False):\n",
    "    quant_df = df[colList].quantile([lowPercentile, highPercentile])\n",
    "    if verbose:\n",
    "        print(quant_df)\n",
    "    for name in list(df[colList].columns):\n",
    "        df = df[(df[name] >= quant_df.loc[lowPercentile, name]) & (df[name] <= quant_df.loc[highPercentile, name])]\n",
    "    return df\n",
    "\n",
    "# Claim Number\n",
    "X.drop(X[X.ClaimNb > 7].index, inplace=True)\n",
    "\n",
    "# Exposure\n",
    "X.loc[(X['Exposure'] > 1)] == 1\n",
    "\n",
    "# Vehicle Power\n",
    "cols = ['VehPower']\n",
    "X = removeOutliers(X, colList=cols , verbose=True)\n",
    "\n",
    "# Vehicle Age\n",
    "X.drop(X[X.VehAge > 35].index, inplace=True)\n",
    "\n",
    "\n",
    "# Driver's Age\n",
    "cols = ['DrivAge']\n",
    "X = removeOutliers(X, colList=cols, lowPercentile=0.00, highPercentile=0.99, verbose=True)\n",
    "\n",
    "# Bonus Malus\n",
    "X.drop(X[X.BonusMalus > ((max(X.BonusMalus)-100)/2)+100].index, inplace=True)\n",
    "\n",
    "X.describe().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Group Region\n",
    "\n",
    "X['Region'] = X['Region'].replace(['R22','R23','R25','R26','R41','R42','R43','R74','R83','R94','R54','R73'],'R21')\n",
    "\n",
    "# Bin the Driver's age\n",
    "binsInterval = [17, 30, 52, 75, 81] # Create bins interval based on distribution/quartiles\n",
    "intervalLabels = ['(18, 30]', '(31, 52]',  '(53, 75]', '(76, 80]']\n",
    "X['DrivAge_binned'] = pd.cut(X['DrivAge'], bins = binsInterval, labels=intervalLabels)\n",
    "\n",
    "#Group Vehicle Brand\n",
    "\n",
    "X['VehBrand'] = X['VehBrand'].replace(['B10','B11','B13'],'B14')\n",
    "X['VehBrand'] = X['VehBrand'].replace(['B4','B5'],'B6')\n",
    "\n",
    "#Group Vehicle Power\n",
    "\n",
    "X['VehPower'] = X['VehPower'].replace(['11','12','13','14'],'15')\n",
    "\n",
    "# Group Region (pvalue)\n",
    "X['Region'] = X['Region'].replace(['R53'],'R24')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "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>IDpol</th>\n",
       "      <th>ClaimNb</th>\n",
       "      <th>Exposure</th>\n",
       "      <th>Area_D</th>\n",
       "      <th>Area_B</th>\n",
       "      <th>Area_E</th>\n",
       "      <th>Area_C</th>\n",
       "      <th>Area_F</th>\n",
       "      <th>Area_A</th>\n",
       "      <th>VehPower</th>\n",
       "      <th>...</th>\n",
       "      <th>Region_R31</th>\n",
       "      <th>Region_R91</th>\n",
       "      <th>Region_R52</th>\n",
       "      <th>Region_R93</th>\n",
       "      <th>Region_R11</th>\n",
       "      <th>Region_R24</th>\n",
       "      <th>DrivAge_binned_(18, 30]</th>\n",
       "      <th>DrivAge_binned_(31, 52]</th>\n",
       "      <th>DrivAge_binned_(53, 75]</th>\n",
       "      <th>DrivAge_binned_(76, 80]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.10000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.77000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.75000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.09000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.84000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678008</th>\n",
       "      <td>6114326.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00274</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678009</th>\n",
       "      <td>6114327.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00274</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678010</th>\n",
       "      <td>6114328.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00274</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678011</th>\n",
       "      <td>6114329.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00274</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678012</th>\n",
       "      <td>6114330.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00274</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>654729 rows × 35 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            IDpol  ClaimNb  Exposure  Area_D  Area_B  Area_E  Area_C  Area_F  \\\n",
       "0             1.0        1   0.10000       1       0       0       0       0   \n",
       "1             3.0        1   0.77000       1       0       0       0       0   \n",
       "2             5.0        1   0.75000       0       1       0       0       0   \n",
       "3            10.0        1   0.09000       0       1       0       0       0   \n",
       "4            11.0        1   0.84000       0       1       0       0       0   \n",
       "...           ...      ...       ...     ...     ...     ...     ...     ...   \n",
       "678008  6114326.0        0   0.00274       0       0       1       0       0   \n",
       "678009  6114327.0        0   0.00274       0       0       1       0       0   \n",
       "678010  6114328.0        0   0.00274       1       0       0       0       0   \n",
       "678011  6114329.0        0   0.00274       0       1       0       0       0   \n",
       "678012  6114330.0        0   0.00274       0       1       0       0       0   \n",
       "\n",
       "        Area_A  VehPower  ...  Region_R31  Region_R91  Region_R52  Region_R93  \\\n",
       "0            0         5  ...           0           0           0           0   \n",
       "1            0         5  ...           0           0           0           0   \n",
       "2            0         6  ...           0           0           0           0   \n",
       "3            0         7  ...           0           0           0           0   \n",
       "4            0         7  ...           0           0           0           0   \n",
       "...        ...       ...  ...         ...         ...         ...         ...   \n",
       "678008       0         4  ...           0           0           0           1   \n",
       "678009       0         4  ...           0           0           0           0   \n",
       "678010       0         6  ...           0           0           0           0   \n",
       "678011       0         4  ...           0           0           0           0   \n",
       "678012       0         7  ...           0           0           0           0   \n",
       "\n",
       "        Region_R11  Region_R24  DrivAge_binned_(18, 30]  \\\n",
       "0                0           0                        0   \n",
       "1                0           0                        0   \n",
       "2                0           0                        0   \n",
       "3                0           0                        0   \n",
       "4                0           0                        0   \n",
       "...            ...         ...                      ...   \n",
       "678008           0           0                        0   \n",
       "678009           1           0                        0   \n",
       "678010           0           0                        0   \n",
       "678011           0           0                        0   \n",
       "678012           0           0                        1   \n",
       "\n",
       "        DrivAge_binned_(31, 52]  DrivAge_binned_(53, 75]  \\\n",
       "0                             0                        1   \n",
       "1                             0                        1   \n",
       "2                             1                        0   \n",
       "3                             1                        0   \n",
       "4                             1                        0   \n",
       "...                         ...                      ...   \n",
       "678008                        0                        1   \n",
       "678009                        1                        0   \n",
       "678010                        1                        0   \n",
       "678011                        0                        1   \n",
       "678012                        0                        0   \n",
       "\n",
       "        DrivAge_binned_(76, 80]  \n",
       "0                             0  \n",
       "1                             0  \n",
       "2                             0  \n",
       "3                             0  \n",
       "4                             0  \n",
       "...                         ...  \n",
       "678008                        0  \n",
       "678009                        0  \n",
       "678010                        0  \n",
       "678011                        0  \n",
       "678012                        0  \n",
       "\n",
       "[654729 rows x 35 columns]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Encode categorical values into dummy variables\n",
    "cols = ['Area', 'VehBrand', 'VehGas', 'Region','DrivAge_binned']\n",
    "ce_one_hot = ce.OneHotEncoder(cols = cols, use_cat_names=True)\n",
    "X = ce_one_hot.fit_transform(X)\n",
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Create the target\n",
    "y = X['ClaimNb']/X['Exposure']\n",
    "\n",
    "# Drop not used columns\n",
    "cols = ['IDpol', 'ClaimNb', 'Exposure','VehPower','VehAge','DrivAge','BonusMalus','Density']\n",
    "X1 = X.drop(columns=cols)\n",
    "\n",
    "for cols in cols:\n",
    "    X1[cols]=X[cols]\n",
    "    \n",
    "X.drop(\"ClaimNb\", axis=1, inplace=True)  #Remove Claim Nb which is the target variable\n",
    "X.drop(\"IDpol\", axis=1, inplace=True)    #Remove IDpolicy since logicaly doesn't explain the number of claims\n",
    "X.drop(\"Exposure\", axis=1, inplace=True) #Remove Exposure which is in the target variable"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modeling DT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split the dataset intro train and test\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, train_size = 0.75, test_size = 0.25, random_state = 500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialize model\n",
    "tree_regr = tree.DecisionTreeRegressor(random_state=123, max_depth=4, min_samples_leaf=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Learning Curve for DecisionTreeRegressor'}, xlabel='Training Instances', ylabel='Score'>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Plot the learning curve\n",
    "cv = 10\n",
    "sizes = np.linspace(0.1, 1.0, 5)\n",
    "visualizer = LearningCurve(estimator=tree_regr, cv=cv, scoring='r2', train_sizes=sizes, n_jobs=4, random_state=123)\n",
    "visualizer.fit(X, y)\n",
    "visualizer.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Simmilarly to the Linear Regression Model, the Learning curve obtained for the Decision Tree model follows the same evolution as training instances increase, i.e, the gap between training score and the cross validation score tends to decrease."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(max_depth=4, min_samples_leaf=5, random_state=123)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create and train the model\n",
    "tree_regr.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>Feature</th>\n",
       "      <th>Coefficient</th>\n",
       "      <th>AbsCoefficient</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>VehAge</td>\n",
       "      <td>0.450850</td>\n",
       "      <td>0.450850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>VehGas_Regular</td>\n",
       "      <td>0.203683</td>\n",
       "      <td>0.203683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>VehBrand_B12</td>\n",
       "      <td>0.188515</td>\n",
       "      <td>0.188515</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>VehPower</td>\n",
       "      <td>0.084458</td>\n",
       "      <td>0.084458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>BonusMalus</td>\n",
       "      <td>0.036929</td>\n",
       "      <td>0.036929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>DrivAge</td>\n",
       "      <td>0.010941</td>\n",
       "      <td>0.010941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Region_R91</td>\n",
       "      <td>0.008299</td>\n",
       "      <td>0.008299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Density</td>\n",
       "      <td>0.007290</td>\n",
       "      <td>0.007290</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>VehBrand_B6</td>\n",
       "      <td>0.006568</td>\n",
       "      <td>0.006568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Region_R11</td>\n",
       "      <td>0.002148</td>\n",
       "      <td>0.002148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Region_R52</td>\n",
       "      <td>0.000319</td>\n",
       "      <td>0.000319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>Region_R24</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>DrivAge_binned_(18, 30]</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>DrivAge_binned_(31, 52]</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Region_R21</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>DrivAge_binned_(53, 75]</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Region_R93</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>Region_R31</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Region_R72</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Area_D</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Region_R82</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>VehGas_Diesel</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Area_B</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>VehBrand_B1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>VehBrand_B14</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>VehBrand_B2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>VehBrand_B3</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Area_A</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Area_F</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Area_C</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Area_E</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>DrivAge_binned_(76, 80]</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    Feature  Coefficient  AbsCoefficient\n",
       "7                    VehAge     0.450850        0.450850\n",
       "16           VehGas_Regular     0.203683        0.203683\n",
       "10             VehBrand_B12     0.188515        0.188515\n",
       "6                  VehPower     0.084458        0.084458\n",
       "9                BonusMalus     0.036929        0.036929\n",
       "8                   DrivAge     0.010941        0.010941\n",
       "23               Region_R91     0.008299        0.008299\n",
       "18                  Density     0.007290        0.007290\n",
       "11              VehBrand_B6     0.006568        0.006568\n",
       "26               Region_R11     0.002148        0.002148\n",
       "24               Region_R52     0.000319        0.000319\n",
       "27               Region_R24     0.000000        0.000000\n",
       "28  DrivAge_binned_(18, 30]     0.000000        0.000000\n",
       "29  DrivAge_binned_(31, 52]     0.000000        0.000000\n",
       "20               Region_R21     0.000000        0.000000\n",
       "30  DrivAge_binned_(53, 75]     0.000000        0.000000\n",
       "25               Region_R93     0.000000        0.000000\n",
       "22               Region_R31     0.000000        0.000000\n",
       "21               Region_R72     0.000000        0.000000\n",
       "0                    Area_D     0.000000        0.000000\n",
       "19               Region_R82     0.000000        0.000000\n",
       "17            VehGas_Diesel     0.000000        0.000000\n",
       "1                    Area_B     0.000000        0.000000\n",
       "15              VehBrand_B1     0.000000        0.000000\n",
       "14             VehBrand_B14     0.000000        0.000000\n",
       "13              VehBrand_B2     0.000000        0.000000\n",
       "12              VehBrand_B3     0.000000        0.000000\n",
       "5                    Area_A     0.000000        0.000000\n",
       "4                    Area_F     0.000000        0.000000\n",
       "3                    Area_C     0.000000        0.000000\n",
       "2                    Area_E     0.000000        0.000000\n",
       "31  DrivAge_binned_(76, 80]     0.000000        0.000000"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Explore feature importance (Gini importance - normalized)\n",
    "formattedList = [float(format(member,'.6f')) for member in tree_regr.feature_importances_]\n",
    "formattedList2 = [abs(float(format(member,'.6f'))) for member in tree_regr.feature_importances_]\n",
    "data_tuples = list(zip(X.columns,formattedList,formattedList2))\n",
    "coeff_df = pd.DataFrame(data=data_tuples, columns=['Feature','Coefficient','AbsCoefficient'])\n",
    "coeff_df.reset_index(drop=True, inplace=True)\n",
    "coeff_df.sort_values(by=['AbsCoefficient'], inplace=True, ascending=False)\n",
    "coeff_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As one can observe from the obtained coeficients, all of them influence positively the number of claims per year. Also, it is possible to observe that some of them are zero, which means that those specific variables (for instance, Region 31 does not explain the number of claims even though Region 91 does."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\r\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\r\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\r\n",
       "<!-- Generated by graphviz version 3.0.0 (20220226.1711)\r\n",
       " -->\r\n",
       "<!-- Title: Tree Pages: 1 -->\r\n",
       "<svg width=\"2940pt\" height=\"477pt\"\r\n",
       " viewBox=\"0.00 0.00 2939.50 477.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\r\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 473)\">\r\n",
       "<title>Tree</title>\r\n",
       "<polygon fill=\"white\" stroke=\"transparent\" points=\"-4,4 -4,-473 2935.5,-473 2935.5,4 -4,4\"/>\r\n",
       "<!-- 0 -->\r\n",
       "<g id=\"node1\" class=\"node\">\r\n",
       "<title>0</title>\r\n",
       "<path fill=\"#fef8f5\" stroke=\"black\" d=\"M1537,-469C1537,-469 1394,-469 1394,-469 1388,-469 1382,-463 1382,-457 1382,-457 1382,-413 1382,-413 1382,-407 1388,-401 1394,-401 1394,-401 1537,-401 1537,-401 1543,-401 1549,-407 1549,-413 1549,-413 1549,-457 1549,-457 1549,-463 1543,-469 1537,-469\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1421.5\" y=\"-453.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">VehAge ≤ 0.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"1390\" y=\"-438.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 21.034</text>\r\n",
       "<text text-anchor=\"start\" x=\"1405.5\" y=\"-423.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 491046</text>\r\n",
       "<text text-anchor=\"start\" x=\"1422.5\" y=\"-408.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.263</text>\r\n",
       "</g>\r\n",
       "<!-- 1 -->\r\n",
       "<g id=\"node2\" class=\"node\">\r\n",
       "<title>1</title>\r\n",
       "<path fill=\"#f6d4bb\" stroke=\"black\" d=\"M1168.5,-365C1168.5,-365 1016.5,-365 1016.5,-365 1010.5,-365 1004.5,-359 1004.5,-353 1004.5,-353 1004.5,-309 1004.5,-309 1004.5,-303 1010.5,-297 1016.5,-297 1016.5,-297 1168.5,-297 1168.5,-297 1174.5,-297 1180.5,-303 1180.5,-309 1180.5,-309 1180.5,-353 1180.5,-353 1180.5,-359 1174.5,-365 1168.5,-365\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1026\" y=\"-349.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">VehBrand_B12 ≤ 0.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"1012.5\" y=\"-334.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 116.064</text>\r\n",
       "<text text-anchor=\"start\" x=\"1036.5\" y=\"-319.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 41270</text>\r\n",
       "<text text-anchor=\"start\" x=\"1049.5\" y=\"-304.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 1.145</text>\r\n",
       "</g>\r\n",
       "<!-- 0&#45;&gt;1 -->\r\n",
       "<g id=\"edge1\" class=\"edge\">\r\n",
       "<title>0&#45;&gt;1</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1381.79,-411.11C1325.13,-395.61 1250.05,-375.08 1190.58,-358.82\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1191.37,-355.41 1180.8,-356.15 1189.52,-362.16 1191.37,-355.41\"/>\r\n",
       "<text text-anchor=\"middle\" x=\"1193.19\" y=\"-374.16\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">True</text>\r\n",
       "</g>\r\n",
       "<!-- 16 -->\r\n",
       "<g id=\"node17\" class=\"node\">\r\n",
       "<title>16</title>\r\n",
       "<path fill=\"#fefcfa\" stroke=\"black\" d=\"M1908,-365C1908,-365 1765,-365 1765,-365 1759,-365 1753,-359 1753,-353 1753,-353 1753,-309 1753,-309 1753,-303 1759,-297 1765,-297 1765,-297 1908,-297 1908,-297 1914,-297 1920,-303 1920,-309 1920,-309 1920,-353 1920,-353 1920,-359 1914,-365 1908,-365\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1776\" y=\"-349.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">BonusMalus ≤ 95.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"1761\" y=\"-334.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 12.236</text>\r\n",
       "<text text-anchor=\"start\" x=\"1776.5\" y=\"-319.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 449776</text>\r\n",
       "<text text-anchor=\"start\" x=\"1793.5\" y=\"-304.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.182</text>\r\n",
       "</g>\r\n",
       "<!-- 0&#45;&gt;16 -->\r\n",
       "<g id=\"edge16\" class=\"edge\">\r\n",
       "<title>0&#45;&gt;16</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1549.22,-410.98C1606.82,-395.15 1683.37,-374.1 1742.97,-357.71\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1744.05,-361.05 1752.76,-355.02 1742.19,-354.3 1744.05,-361.05\"/>\r\n",
       "<text text-anchor=\"middle\" x=\"1740.4\" y=\"-373.05\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">False</text>\r\n",
       "</g>\r\n",
       "<!-- 2 -->\r\n",
       "<g id=\"node3\" class=\"node\">\r\n",
       "<title>2</title>\r\n",
       "<path fill=\"#fdf5ef\" stroke=\"black\" d=\"M610,-261C610,-261 467,-261 467,-261 461,-261 455,-255 455,-249 455,-249 455,-205 455,-205 455,-199 461,-193 467,-193 467,-193 610,-193 610,-193 616,-193 622,-199 622,-205 622,-205 622,-249 622,-249 622,-255 616,-261 610,-261\"/>\r\n",
       "<text text-anchor=\"start\" x=\"491\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">DrivAge ≤ 45.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"463\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 45.955</text>\r\n",
       "<text text-anchor=\"start\" x=\"482.5\" y=\"-215.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 14945</text>\r\n",
       "<text text-anchor=\"start\" x=\"495.5\" y=\"-200.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.352</text>\r\n",
       "</g>\r\n",
       "<!-- 1&#45;&gt;2 -->\r\n",
       "<g id=\"edge2\" class=\"edge\">\r\n",
       "<title>1&#45;&gt;2</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1004.41,-313.78C902.67,-295.05 735.99,-264.36 632.01,-245.22\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"632.62,-241.77 622.15,-243.4 631.35,-248.65 632.62,-241.77\"/>\r\n",
       "</g>\r\n",
       "<!-- 9 -->\r\n",
       "<g id=\"node10\" class=\"node\">\r\n",
       "<title>9</title>\r\n",
       "<path fill=\"#f2c19e\" stroke=\"black\" d=\"M1168.5,-261C1168.5,-261 1016.5,-261 1016.5,-261 1010.5,-261 1004.5,-255 1004.5,-249 1004.5,-249 1004.5,-205 1004.5,-205 1004.5,-199 1010.5,-193 1016.5,-193 1016.5,-193 1168.5,-193 1168.5,-193 1174.5,-193 1180.5,-199 1180.5,-205 1180.5,-205 1180.5,-249 1180.5,-249 1180.5,-255 1174.5,-261 1168.5,-261\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1020\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">VehGas_Regular ≤ 0.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"1012.5\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 155.306</text>\r\n",
       "<text text-anchor=\"start\" x=\"1036.5\" y=\"-215.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 26325</text>\r\n",
       "<text text-anchor=\"start\" x=\"1049.5\" y=\"-200.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 1.596</text>\r\n",
       "</g>\r\n",
       "<!-- 1&#45;&gt;9 -->\r\n",
       "<g id=\"edge9\" class=\"edge\">\r\n",
       "<title>1&#45;&gt;9</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1092.5,-296.88C1092.5,-288.78 1092.5,-279.98 1092.5,-271.47\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1096,-271.3 1092.5,-261.3 1089,-271.3 1096,-271.3\"/>\r\n",
       "</g>\r\n",
       "<!-- 3 -->\r\n",
       "<g id=\"node4\" class=\"node\">\r\n",
       "<title>3</title>\r\n",
       "<path fill=\"#fefcfb\" stroke=\"black\" d=\"M328,-157C328,-157 193,-157 193,-157 187,-157 181,-151 181,-145 181,-145 181,-101 181,-101 181,-95 187,-89 193,-89 193,-89 328,-89 328,-89 334,-89 340,-95 340,-101 340,-101 340,-145 340,-145 340,-151 334,-157 328,-157\"/>\r\n",
       "<text text-anchor=\"start\" x=\"202\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Region_R52 ≤ 0.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"189\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 3.155</text>\r\n",
       "<text text-anchor=\"start\" x=\"209\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 7013</text>\r\n",
       "<text text-anchor=\"start\" x=\"217.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.173</text>\r\n",
       "</g>\r\n",
       "<!-- 2&#45;&gt;3 -->\r\n",
       "<g id=\"edge3\" class=\"edge\">\r\n",
       "<title>2&#45;&gt;3</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M454.71,-195.26C421.68,-183.14 383.69,-169.2 350.11,-156.88\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"350.9,-153.44 340.3,-153.28 348.48,-160.01 350.9,-153.44\"/>\r\n",
       "</g>\r\n",
       "<!-- 6 -->\r\n",
       "<g id=\"node7\" class=\"node\">\r\n",
       "<title>6</title>\r\n",
       "<path fill=\"#fceee5\" stroke=\"black\" d=\"M610,-157C610,-157 467,-157 467,-157 461,-157 455,-151 455,-145 455,-145 455,-101 455,-101 455,-95 461,-89 467,-89 467,-89 610,-89 610,-89 616,-89 622,-95 622,-101 622,-101 622,-145 622,-145 622,-151 616,-157 610,-157\"/>\r\n",
       "<text text-anchor=\"start\" x=\"484.5\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Density ≤ 4122.0</text>\r\n",
       "<text text-anchor=\"start\" x=\"463\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 83.744</text>\r\n",
       "<text text-anchor=\"start\" x=\"487\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 7932</text>\r\n",
       "<text text-anchor=\"start\" x=\"499.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.51</text>\r\n",
       "</g>\r\n",
       "<!-- 2&#45;&gt;6 -->\r\n",
       "<g id=\"edge6\" class=\"edge\">\r\n",
       "<title>2&#45;&gt;6</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M538.5,-192.88C538.5,-184.78 538.5,-175.98 538.5,-167.47\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"542,-167.3 538.5,-157.3 535,-167.3 542,-167.3\"/>\r\n",
       "</g>\r\n",
       "<!-- 4 -->\r\n",
       "<g id=\"node5\" class=\"node\">\r\n",
       "<title>4</title>\r\n",
       "<path fill=\"#fffdfb\" stroke=\"black\" d=\"M147,-53C147,-53 12,-53 12,-53 6,-53 0,-47 0,-41 0,-41 0,-12 0,-12 0,-6 6,0 12,0 12,0 147,0 147,0 153,0 159,-6 159,-12 159,-12 159,-41 159,-41 159,-47 153,-53 147,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"8\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 1.756</text>\r\n",
       "<text text-anchor=\"start\" x=\"28\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 6669</text>\r\n",
       "<text text-anchor=\"start\" x=\"40.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.16</text>\r\n",
       "</g>\r\n",
       "<!-- 3&#45;&gt;4 -->\r\n",
       "<g id=\"edge4\" class=\"edge\">\r\n",
       "<title>3&#45;&gt;4</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M197.19,-88.95C177.75,-78.79 156.45,-67.67 137.38,-57.72\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"138.99,-54.61 128.51,-53.09 135.75,-60.82 138.99,-54.61\"/>\r\n",
       "</g>\r\n",
       "<!-- 5 -->\r\n",
       "<g id=\"node6\" class=\"node\">\r\n",
       "<title>5</title>\r\n",
       "<path fill=\"#fcf1e9\" stroke=\"black\" d=\"M332,-53C332,-53 189,-53 189,-53 183,-53 177,-47 177,-41 177,-41 177,-12 177,-12 177,-6 183,0 189,0 189,0 332,0 332,0 338,0 344,-6 344,-12 344,-12 344,-41 344,-41 344,-47 338,-53 332,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"185\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 30.204</text>\r\n",
       "<text text-anchor=\"start\" x=\"213\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 344</text>\r\n",
       "<text text-anchor=\"start\" x=\"217.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.436</text>\r\n",
       "</g>\r\n",
       "<!-- 3&#45;&gt;5 -->\r\n",
       "<g id=\"edge5\" class=\"edge\">\r\n",
       "<title>3&#45;&gt;5</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M260.5,-88.95C260.5,-80.72 260.5,-71.85 260.5,-63.48\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"264,-63.24 260.5,-53.24 257,-63.24 264,-63.24\"/>\r\n",
       "</g>\r\n",
       "<!-- 7 -->\r\n",
       "<g id=\"node8\" class=\"node\">\r\n",
       "<title>7</title>\r\n",
       "<path fill=\"#fcf2ea\" stroke=\"black\" d=\"M517,-53C517,-53 374,-53 374,-53 368,-53 362,-47 362,-41 362,-41 362,-12 362,-12 362,-6 368,0 374,0 374,0 517,0 517,0 523,0 529,-6 529,-12 529,-12 529,-41 529,-41 529,-47 523,-53 517,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"370\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 52.637</text>\r\n",
       "<text text-anchor=\"start\" x=\"394\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 7189</text>\r\n",
       "<text text-anchor=\"start\" x=\"402.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.424</text>\r\n",
       "</g>\r\n",
       "<!-- 6&#45;&gt;7 -->\r\n",
       "<g id=\"edge7\" class=\"edge\">\r\n",
       "<title>6&#45;&gt;7</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M505.97,-88.95C496.88,-79.71 487,-69.67 477.92,-60.44\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"480.34,-57.91 470.83,-53.24 475.35,-62.82 480.34,-57.91\"/>\r\n",
       "</g>\r\n",
       "<!-- 8 -->\r\n",
       "<g id=\"node9\" class=\"node\">\r\n",
       "<title>8</title>\r\n",
       "<path fill=\"#f4ccae\" stroke=\"black\" d=\"M702,-53C702,-53 559,-53 559,-53 553,-53 547,-47 547,-41 547,-41 547,-12 547,-12 547,-6 553,0 559,0 559,0 702,0 702,0 708,0 714,-6 714,-12 714,-12 714,-41 714,-41 714,-47 708,-53 702,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"555\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 383.95</text>\r\n",
       "<text text-anchor=\"start\" x=\"583\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 743</text>\r\n",
       "<text text-anchor=\"start\" x=\"587.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 1.344</text>\r\n",
       "</g>\r\n",
       "<!-- 6&#45;&gt;8 -->\r\n",
       "<g id=\"edge8\" class=\"edge\">\r\n",
       "<title>6&#45;&gt;8</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M570.68,-88.95C579.67,-79.71 589.44,-69.67 598.43,-60.44\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"600.98,-62.84 605.44,-53.24 595.96,-57.96 600.98,-62.84\"/>\r\n",
       "</g>\r\n",
       "<!-- 10 -->\r\n",
       "<g id=\"node11\" class=\"node\">\r\n",
       "<title>10</title>\r\n",
       "<path fill=\"#f9e2d1\" stroke=\"black\" d=\"M1069,-157C1069,-157 926,-157 926,-157 920,-157 914,-151 914,-145 914,-145 914,-101 914,-101 914,-95 920,-89 926,-89 926,-89 1069,-89 1069,-89 1075,-89 1081,-95 1081,-101 1081,-101 1081,-145 1081,-145 1081,-151 1075,-157 1069,-157\"/>\r\n",
       "<text text-anchor=\"start\" x=\"946.5\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">VehPower ≤ 4.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"922\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 114.62</text>\r\n",
       "<text text-anchor=\"start\" x=\"941.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 13105</text>\r\n",
       "<text text-anchor=\"start\" x=\"954.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.814</text>\r\n",
       "</g>\r\n",
       "<!-- 9&#45;&gt;10 -->\r\n",
       "<g id=\"edge10\" class=\"edge\">\r\n",
       "<title>9&#45;&gt;10</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1061.66,-192.88C1053.28,-183.89 1044.11,-174.04 1035.39,-164.68\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1037.89,-162.23 1028.51,-157.3 1032.77,-167 1037.89,-162.23\"/>\r\n",
       "</g>\r\n",
       "<!-- 13 -->\r\n",
       "<g id=\"node14\" class=\"node\">\r\n",
       "<title>13</title>\r\n",
       "<path fill=\"#eca16b\" stroke=\"black\" d=\"M1263.5,-157C1263.5,-157 1111.5,-157 1111.5,-157 1105.5,-157 1099.5,-151 1099.5,-145 1099.5,-145 1099.5,-101 1099.5,-101 1099.5,-95 1105.5,-89 1111.5,-89 1111.5,-89 1263.5,-89 1263.5,-89 1269.5,-89 1275.5,-95 1275.5,-101 1275.5,-101 1275.5,-145 1275.5,-145 1275.5,-151 1269.5,-157 1263.5,-157\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1136.5\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">VehPower ≤ 4.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"1107.5\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 194.434</text>\r\n",
       "<text text-anchor=\"start\" x=\"1131.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 13220</text>\r\n",
       "<text text-anchor=\"start\" x=\"1148.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 2.37</text>\r\n",
       "</g>\r\n",
       "<!-- 9&#45;&gt;13 -->\r\n",
       "<g id=\"edge13\" class=\"edge\">\r\n",
       "<title>9&#45;&gt;13</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1123.34,-192.88C1131.72,-183.89 1140.89,-174.04 1149.61,-164.68\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1152.23,-167 1156.49,-157.3 1147.11,-162.23 1152.23,-167\"/>\r\n",
       "</g>\r\n",
       "<!-- 11 -->\r\n",
       "<g id=\"node12\" class=\"node\">\r\n",
       "<title>11</title>\r\n",
       "<path fill=\"#fffffe\" stroke=\"black\" d=\"M879,-53C879,-53 744,-53 744,-53 738,-53 732,-47 732,-41 732,-41 732,-12 732,-12 732,-6 738,0 744,0 744,0 879,0 879,0 885,0 891,-6 891,-12 891,-12 891,-41 891,-41 891,-47 885,-53 879,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"740\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 5.075</text>\r\n",
       "<text text-anchor=\"start\" x=\"760\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 2428</text>\r\n",
       "<text text-anchor=\"start\" x=\"768.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.116</text>\r\n",
       "</g>\r\n",
       "<!-- 10&#45;&gt;11 -->\r\n",
       "<g id=\"edge11\" class=\"edge\">\r\n",
       "<title>10&#45;&gt;11</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M932.45,-88.95C912.46,-78.79 890.57,-67.67 870.98,-57.72\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"872.36,-54.5 861.86,-53.09 869.19,-60.74 872.36,-54.5\"/>\r\n",
       "</g>\r\n",
       "<!-- 12 -->\r\n",
       "<g id=\"node13\" class=\"node\">\r\n",
       "<title>12</title>\r\n",
       "<path fill=\"#f8dbc6\" stroke=\"black\" d=\"M1073.5,-53C1073.5,-53 921.5,-53 921.5,-53 915.5,-53 909.5,-47 909.5,-41 909.5,-41 909.5,-12 909.5,-12 909.5,-6 915.5,0 921.5,0 921.5,0 1073.5,0 1073.5,0 1079.5,0 1085.5,-6 1085.5,-12 1085.5,-12 1085.5,-41 1085.5,-41 1085.5,-47 1079.5,-53 1073.5,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"917.5\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 139.395</text>\r\n",
       "<text text-anchor=\"start\" x=\"941.5\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 10677</text>\r\n",
       "<text text-anchor=\"start\" x=\"954.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.973</text>\r\n",
       "</g>\r\n",
       "<!-- 10&#45;&gt;12 -->\r\n",
       "<g id=\"edge12\" class=\"edge\">\r\n",
       "<title>10&#45;&gt;12</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M997.5,-88.95C997.5,-80.72 997.5,-71.85 997.5,-63.48\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1001,-63.24 997.5,-53.24 994,-63.24 1001,-63.24\"/>\r\n",
       "</g>\r\n",
       "<!-- 14 -->\r\n",
       "<g id=\"node15\" class=\"node\">\r\n",
       "<title>14</title>\r\n",
       "<path fill=\"#f2c29f\" stroke=\"black\" d=\"M1259,-53C1259,-53 1116,-53 1116,-53 1110,-53 1104,-47 1104,-41 1104,-41 1104,-12 1104,-12 1104,-6 1110,0 1116,0 1116,0 1259,0 1259,0 1265,0 1271,-6 1271,-12 1271,-12 1271,-41 1271,-41 1271,-47 1265,-53 1259,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1112\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 96.724</text>\r\n",
       "<text text-anchor=\"start\" x=\"1136\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 5011</text>\r\n",
       "<text text-anchor=\"start\" x=\"1144.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 1.571</text>\r\n",
       "</g>\r\n",
       "<!-- 13&#45;&gt;14 -->\r\n",
       "<g id=\"edge14\" class=\"edge\">\r\n",
       "<title>13&#45;&gt;14</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1187.5,-88.95C1187.5,-80.72 1187.5,-71.85 1187.5,-63.48\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1191,-63.24 1187.5,-53.24 1184,-63.24 1191,-63.24\"/>\r\n",
       "</g>\r\n",
       "<!-- 15 -->\r\n",
       "<g id=\"node16\" class=\"node\">\r\n",
       "<title>15</title>\r\n",
       "<path fill=\"#e78d4b\" stroke=\"black\" d=\"M1453.5,-53C1453.5,-53 1301.5,-53 1301.5,-53 1295.5,-53 1289.5,-47 1289.5,-41 1289.5,-41 1289.5,-12 1289.5,-12 1289.5,-6 1295.5,0 1301.5,0 1301.5,0 1453.5,0 1453.5,0 1459.5,0 1465.5,-6 1465.5,-12 1465.5,-12 1465.5,-41 1465.5,-41 1465.5,-47 1459.5,-53 1453.5,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1297.5\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 253.452</text>\r\n",
       "<text text-anchor=\"start\" x=\"1326\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 8209</text>\r\n",
       "<text text-anchor=\"start\" x=\"1334.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 2.858</text>\r\n",
       "</g>\r\n",
       "<!-- 13&#45;&gt;15 -->\r\n",
       "<g id=\"edge15\" class=\"edge\">\r\n",
       "<title>13&#45;&gt;15</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1253.95,-88.95C1274.46,-78.75 1296.93,-67.57 1317.01,-57.59\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1318.66,-60.67 1326.06,-53.09 1315.54,-54.41 1318.66,-60.67\"/>\r\n",
       "</g>\r\n",
       "<!-- 17 -->\r\n",
       "<g id=\"node18\" class=\"node\">\r\n",
       "<title>17</title>\r\n",
       "<path fill=\"#fefcfb\" stroke=\"black\" d=\"M1908,-261C1908,-261 1765,-261 1765,-261 1759,-261 1753,-255 1753,-249 1753,-249 1753,-205 1753,-205 1753,-199 1759,-193 1765,-193 1765,-193 1908,-193 1908,-193 1914,-193 1920,-199 1920,-205 1920,-205 1920,-249 1920,-249 1920,-255 1914,-261 1908,-261\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1776\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">BonusMalus ≤ 57.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"1761\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 10.827</text>\r\n",
       "<text text-anchor=\"start\" x=\"1776.5\" y=\"-215.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 431288</text>\r\n",
       "<text text-anchor=\"start\" x=\"1793.5\" y=\"-200.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.167</text>\r\n",
       "</g>\r\n",
       "<!-- 16&#45;&gt;17 -->\r\n",
       "<g id=\"edge17\" class=\"edge\">\r\n",
       "<title>16&#45;&gt;17</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1836.5,-296.88C1836.5,-288.78 1836.5,-279.98 1836.5,-271.47\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1840,-271.3 1836.5,-261.3 1833,-271.3 1840,-271.3\"/>\r\n",
       "</g>\r\n",
       "<!-- 24 -->\r\n",
       "<g id=\"node25\" class=\"node\">\r\n",
       "<title>24</title>\r\n",
       "<path fill=\"#fbede3\" stroke=\"black\" d=\"M2449,-261C2449,-261 2306,-261 2306,-261 2300,-261 2294,-255 2294,-249 2294,-249 2294,-205 2294,-205 2294,-199 2300,-193 2306,-193 2306,-193 2449,-193 2449,-193 2455,-193 2461,-199 2461,-205 2461,-205 2461,-249 2461,-249 2461,-255 2455,-261 2449,-261\"/>\r\n",
       "<text text-anchor=\"start\" x=\"2315\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">VehBrand_B6 ≤ 0.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"2302\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 44.996</text>\r\n",
       "<text text-anchor=\"start\" x=\"2321.5\" y=\"-215.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 18488</text>\r\n",
       "<text text-anchor=\"start\" x=\"2334.5\" y=\"-200.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.528</text>\r\n",
       "</g>\r\n",
       "<!-- 16&#45;&gt;24 -->\r\n",
       "<g id=\"edge24\" class=\"edge\">\r\n",
       "<title>16&#45;&gt;24</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1920.14,-314.23C2018.45,-295.69 2181.16,-265.02 2283.68,-245.69\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"2284.53,-249.09 2293.71,-243.8 2283.24,-242.21 2284.53,-249.09\"/>\r\n",
       "</g>\r\n",
       "<!-- 18 -->\r\n",
       "<g id=\"node19\" class=\"node\">\r\n",
       "<title>18</title>\r\n",
       "<path fill=\"#fffdfd\" stroke=\"black\" d=\"M1814,-157C1814,-157 1679,-157 1679,-157 1673,-157 1667,-151 1667,-145 1667,-145 1667,-101 1667,-101 1667,-95 1673,-89 1679,-89 1679,-89 1814,-89 1814,-89 1820,-89 1826,-95 1826,-101 1826,-101 1826,-145 1826,-145 1826,-151 1820,-157 1814,-157\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1698.5\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">VehAge ≤ 10.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"1675\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 8.515</text>\r\n",
       "<text text-anchor=\"start\" x=\"1686.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 296523</text>\r\n",
       "<text text-anchor=\"start\" x=\"1703.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.142</text>\r\n",
       "</g>\r\n",
       "<!-- 17&#45;&gt;18 -->\r\n",
       "<g id=\"edge18\" class=\"edge\">\r\n",
       "<title>17&#45;&gt;18</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1807.28,-192.88C1799.42,-183.98 1790.83,-174.24 1782.64,-164.96\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1785.12,-162.48 1775.88,-157.3 1779.87,-167.11 1785.12,-162.48\"/>\r\n",
       "</g>\r\n",
       "<!-- 21 -->\r\n",
       "<g id=\"node22\" class=\"node\">\r\n",
       "<title>21</title>\r\n",
       "<path fill=\"#fefaf7\" stroke=\"black\" d=\"M1999,-157C1999,-157 1856,-157 1856,-157 1850,-157 1844,-151 1844,-145 1844,-145 1844,-101 1844,-101 1844,-95 1850,-89 1856,-89 1856,-89 1999,-89 1999,-89 2005,-89 2011,-95 2011,-101 2011,-101 2011,-145 2011,-145 2011,-151 2005,-157 1999,-157\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1869\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Region_R11 ≤ 0.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"1852\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 15.909</text>\r\n",
       "<text text-anchor=\"start\" x=\"1867.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 134765</text>\r\n",
       "<text text-anchor=\"start\" x=\"1884.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.221</text>\r\n",
       "</g>\r\n",
       "<!-- 17&#45;&gt;21 -->\r\n",
       "<g id=\"edge21\" class=\"edge\">\r\n",
       "<title>17&#45;&gt;21</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1866.05,-192.88C1873.99,-183.98 1882.68,-174.24 1890.96,-164.96\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1893.75,-167.09 1897.79,-157.3 1888.52,-162.43 1893.75,-167.09\"/>\r\n",
       "</g>\r\n",
       "<!-- 19 -->\r\n",
       "<g id=\"node20\" class=\"node\">\r\n",
       "<title>19</title>\r\n",
       "<path fill=\"#fffdfc\" stroke=\"black\" d=\"M1639,-53C1639,-53 1496,-53 1496,-53 1490,-53 1484,-47 1484,-41 1484,-41 1484,-12 1484,-12 1484,-6 1490,0 1496,0 1496,0 1639,0 1639,0 1645,0 1651,-6 1651,-12 1651,-12 1651,-41 1651,-41 1651,-47 1645,-53 1639,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1492\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 10.459</text>\r\n",
       "<text text-anchor=\"start\" x=\"1507.5\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 212386</text>\r\n",
       "<text text-anchor=\"start\" x=\"1524.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.157</text>\r\n",
       "</g>\r\n",
       "<!-- 18&#45;&gt;19 -->\r\n",
       "<g id=\"edge19\" class=\"edge\">\r\n",
       "<title>18&#45;&gt;19</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1683.89,-88.95C1664.75,-78.84 1643.79,-67.78 1625,-57.86\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1626.44,-54.66 1615.97,-53.09 1623.18,-60.85 1626.44,-54.66\"/>\r\n",
       "</g>\r\n",
       "<!-- 20 -->\r\n",
       "<g id=\"node21\" class=\"node\">\r\n",
       "<title>20</title>\r\n",
       "<path fill=\"#ffffff\" stroke=\"black\" d=\"M1816,-53C1816,-53 1681,-53 1681,-53 1675,-53 1669,-47 1669,-41 1669,-41 1669,-12 1669,-12 1669,-6 1675,0 1681,0 1681,0 1816,0 1816,0 1822,0 1828,-6 1828,-12 1828,-12 1828,-41 1828,-41 1828,-47 1822,-53 1816,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1677\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 3.603</text>\r\n",
       "<text text-anchor=\"start\" x=\"1692.5\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 84137</text>\r\n",
       "<text text-anchor=\"start\" x=\"1705.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.105</text>\r\n",
       "</g>\r\n",
       "<!-- 18&#45;&gt;20 -->\r\n",
       "<g id=\"edge20\" class=\"edge\">\r\n",
       "<title>18&#45;&gt;20</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1747.2,-88.95C1747.37,-80.72 1747.56,-71.85 1747.74,-63.48\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1751.24,-63.31 1747.96,-53.24 1744.24,-63.16 1751.24,-63.31\"/>\r\n",
       "</g>\r\n",
       "<!-- 22 -->\r\n",
       "<g id=\"node23\" class=\"node\">\r\n",
       "<title>22</title>\r\n",
       "<path fill=\"#fefbf8\" stroke=\"black\" d=\"M1993,-53C1993,-53 1858,-53 1858,-53 1852,-53 1846,-47 1846,-41 1846,-41 1846,-12 1846,-12 1846,-6 1852,0 1858,0 1858,0 1993,0 1993,0 1999,0 2005,-6 2005,-12 2005,-12 2005,-41 2005,-41 2005,-47 1999,-53 1993,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"1854\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 9.797</text>\r\n",
       "<text text-anchor=\"start\" x=\"1865.5\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 119653</text>\r\n",
       "<text text-anchor=\"start\" x=\"1882.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.208</text>\r\n",
       "</g>\r\n",
       "<!-- 21&#45;&gt;22 -->\r\n",
       "<g id=\"edge22\" class=\"edge\">\r\n",
       "<title>21&#45;&gt;22</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1926.8,-88.95C1926.63,-80.72 1926.44,-71.85 1926.26,-63.48\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"1929.76,-63.16 1926.04,-53.24 1922.76,-63.31 1929.76,-63.16\"/>\r\n",
       "</g>\r\n",
       "<!-- 23 -->\r\n",
       "<g id=\"node24\" class=\"node\">\r\n",
       "<title>23</title>\r\n",
       "<path fill=\"#fdf6f1\" stroke=\"black\" d=\"M2178,-53C2178,-53 2035,-53 2035,-53 2029,-53 2023,-47 2023,-41 2023,-41 2023,-12 2023,-12 2023,-6 2029,0 2035,0 2035,0 2178,0 2178,0 2184,0 2190,-6 2190,-12 2190,-12 2190,-41 2190,-41 2190,-47 2184,-53 2178,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"2031\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 64.293</text>\r\n",
       "<text text-anchor=\"start\" x=\"2050.5\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 15112</text>\r\n",
       "<text text-anchor=\"start\" x=\"2067.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.32</text>\r\n",
       "</g>\r\n",
       "<!-- 21&#45;&gt;23 -->\r\n",
       "<g id=\"edge23\" class=\"edge\">\r\n",
       "<title>21&#45;&gt;23</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M1990.11,-88.95C2009.25,-78.84 2030.21,-67.78 2049,-57.86\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"2050.82,-60.85 2058.03,-53.09 2047.56,-54.66 2050.82,-60.85\"/>\r\n",
       "</g>\r\n",
       "<!-- 25 -->\r\n",
       "<g id=\"node26\" class=\"node\">\r\n",
       "<title>25</title>\r\n",
       "<path fill=\"#fcf0e8\" stroke=\"black\" d=\"M2449,-157C2449,-157 2306,-157 2306,-157 2300,-157 2294,-151 2294,-145 2294,-145 2294,-101 2294,-101 2294,-95 2300,-89 2306,-89 2306,-89 2449,-89 2449,-89 2455,-89 2461,-95 2461,-101 2461,-101 2461,-145 2461,-145 2461,-151 2455,-157 2449,-157\"/>\r\n",
       "<text text-anchor=\"start\" x=\"2330\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">DrivAge ≤ 23.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"2302\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 22.414</text>\r\n",
       "<text text-anchor=\"start\" x=\"2321.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 15514</text>\r\n",
       "<text text-anchor=\"start\" x=\"2334.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.455</text>\r\n",
       "</g>\r\n",
       "<!-- 24&#45;&gt;25 -->\r\n",
       "<g id=\"edge25\" class=\"edge\">\r\n",
       "<title>24&#45;&gt;25</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M2377.5,-192.88C2377.5,-184.78 2377.5,-175.98 2377.5,-167.47\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"2381,-167.3 2377.5,-157.3 2374,-167.3 2381,-167.3\"/>\r\n",
       "</g>\r\n",
       "<!-- 28 -->\r\n",
       "<g id=\"node29\" class=\"node\">\r\n",
       "<title>28</title>\r\n",
       "<path fill=\"#f8decb\" stroke=\"black\" d=\"M2729.5,-157C2729.5,-157 2577.5,-157 2577.5,-157 2571.5,-157 2565.5,-151 2565.5,-145 2565.5,-145 2565.5,-101 2565.5,-101 2565.5,-95 2571.5,-89 2577.5,-89 2577.5,-89 2729.5,-89 2729.5,-89 2735.5,-89 2741.5,-95 2741.5,-101 2741.5,-101 2741.5,-145 2741.5,-145 2741.5,-151 2735.5,-157 2729.5,-157\"/>\r\n",
       "<text text-anchor=\"start\" x=\"2595\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Region_R91 ≤ 0.5</text>\r\n",
       "<text text-anchor=\"start\" x=\"2573.5\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 162.623</text>\r\n",
       "<text text-anchor=\"start\" x=\"2602\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 2974</text>\r\n",
       "<text text-anchor=\"start\" x=\"2610.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.909</text>\r\n",
       "</g>\r\n",
       "<!-- 24&#45;&gt;28 -->\r\n",
       "<g id=\"edge28\" class=\"edge\">\r\n",
       "<title>24&#45;&gt;28</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M2461.06,-195.12C2490.97,-184.07 2524.94,-171.51 2555.92,-160.06\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"2557.27,-163.29 2565.44,-156.55 2554.84,-156.73 2557.27,-163.29\"/>\r\n",
       "</g>\r\n",
       "<!-- 26 -->\r\n",
       "<g id=\"node27\" class=\"node\">\r\n",
       "<title>26</title>\r\n",
       "<path fill=\"#fae6d8\" stroke=\"black\" d=\"M2363,-53C2363,-53 2220,-53 2220,-53 2214,-53 2208,-47 2208,-41 2208,-41 2208,-12 2208,-12 2208,-6 2214,0 2220,0 2220,0 2363,0 2363,0 2369,0 2375,-6 2375,-12 2375,-12 2375,-41 2375,-41 2375,-47 2369,-53 2363,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"2216\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 56.805</text>\r\n",
       "<text text-anchor=\"start\" x=\"2240\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 4695</text>\r\n",
       "<text text-anchor=\"start\" x=\"2248.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.709</text>\r\n",
       "</g>\r\n",
       "<!-- 25&#45;&gt;26 -->\r\n",
       "<g id=\"edge26\" class=\"edge\">\r\n",
       "<title>25&#45;&gt;26</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M2347.42,-88.95C2339.1,-79.8 2330.06,-69.87 2321.73,-60.71\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"2324.24,-58.28 2314.92,-53.24 2319.06,-62.99 2324.24,-58.28\"/>\r\n",
       "</g>\r\n",
       "<!-- 27 -->\r\n",
       "<g id=\"node28\" class=\"node\">\r\n",
       "<title>27</title>\r\n",
       "<path fill=\"#fdf5ef\" stroke=\"black\" d=\"M2540,-53C2540,-53 2405,-53 2405,-53 2399,-53 2393,-47 2393,-41 2393,-41 2393,-12 2393,-12 2393,-6 2399,0 2405,0 2405,0 2540,0 2540,0 2546,0 2552,-6 2552,-12 2552,-12 2552,-41 2552,-41 2552,-47 2546,-53 2540,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"2401\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 7.449</text>\r\n",
       "<text text-anchor=\"start\" x=\"2416.5\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 10819</text>\r\n",
       "<text text-anchor=\"start\" x=\"2429.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.345</text>\r\n",
       "</g>\r\n",
       "<!-- 25&#45;&gt;27 -->\r\n",
       "<g id=\"edge27\" class=\"edge\">\r\n",
       "<title>25&#45;&gt;27</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M2410.73,-88.95C2420.01,-79.71 2430.11,-69.67 2439.39,-60.44\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"2442.01,-62.77 2446.63,-53.24 2437.07,-57.81 2442.01,-62.77\"/>\r\n",
       "</g>\r\n",
       "<!-- 29 -->\r\n",
       "<g id=\"node30\" class=\"node\">\r\n",
       "<title>29</title>\r\n",
       "<path fill=\"#f9e2d1\" stroke=\"black\" d=\"M2725,-53C2725,-53 2582,-53 2582,-53 2576,-53 2570,-47 2570,-41 2570,-41 2570,-12 2570,-12 2570,-6 2576,0 2582,0 2582,0 2725,0 2725,0 2731,0 2737,-6 2737,-12 2737,-12 2737,-41 2737,-41 2737,-47 2731,-53 2725,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"2578\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 122.89</text>\r\n",
       "<text text-anchor=\"start\" x=\"2602\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 2849</text>\r\n",
       "<text text-anchor=\"start\" x=\"2610.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 0.811</text>\r\n",
       "</g>\r\n",
       "<!-- 28&#45;&gt;29 -->\r\n",
       "<g id=\"edge29\" class=\"edge\">\r\n",
       "<title>28&#45;&gt;29</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M2653.5,-88.95C2653.5,-80.72 2653.5,-71.85 2653.5,-63.48\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"2657,-63.24 2653.5,-53.24 2650,-63.24 2657,-63.24\"/>\r\n",
       "</g>\r\n",
       "<!-- 30 -->\r\n",
       "<g id=\"node31\" class=\"node\">\r\n",
       "<title>30</title>\r\n",
       "<path fill=\"#e58139\" stroke=\"black\" d=\"M2919.5,-53C2919.5,-53 2767.5,-53 2767.5,-53 2761.5,-53 2755.5,-47 2755.5,-41 2755.5,-41 2755.5,-12 2755.5,-12 2755.5,-6 2761.5,0 2767.5,0 2767.5,0 2919.5,0 2919.5,0 2925.5,0 2931.5,-6 2931.5,-12 2931.5,-12 2931.5,-41 2931.5,-41 2931.5,-47 2925.5,-53 2919.5,-53\"/>\r\n",
       "<text text-anchor=\"start\" x=\"2763.5\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">squared_error = 1063.01</text>\r\n",
       "<text text-anchor=\"start\" x=\"2796\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 125</text>\r\n",
       "<text text-anchor=\"start\" x=\"2800.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = 3.139</text>\r\n",
       "</g>\r\n",
       "<!-- 28&#45;&gt;30 -->\r\n",
       "<g id=\"edge30\" class=\"edge\">\r\n",
       "<title>28&#45;&gt;30</title>\r\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M2719.95,-88.95C2740.46,-78.75 2762.93,-67.57 2783.01,-57.59\"/>\r\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"2784.66,-60.67 2792.06,-53.09 2781.54,-54.41 2784.66,-60.67\"/>\r\n",
       "</g>\r\n",
       "</g>\r\n",
       "</svg>\r\n"
      ],
      "text/plain": [
       "<graphviz.files.Source at 0x1acc27099d0>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Plot the tree\n",
    "dot_data = tree.export_graphviz(tree_regr, out_file=None, \n",
    "                      feature_names=list(X_train.columns),  \n",
    "                      class_names='Returns',  \n",
    "                      filled=True, rounded=True,  \n",
    "                      special_characters=True)  \n",
    "graph = graphviz.Source(dot_data)  \n",
    "graph "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluation DT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "# predict y for X_train and X_test\n",
    "y_pred_train = tree_regr.predict(X_train) \n",
    "y_pred_test = tree_regr.predict(X_test) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to create dataframe with metrics\n",
    "def performanceMetricsDF(metricsObj, yTrain, yPredTrain, yTest, yPredTest):\n",
    "  measures_list = ['MAE','RMSE', 'R^2','MAPE (%)','MAX Error']\n",
    "  train_results = [metricsObj.mean_absolute_error(yTrain, yPredTrain),\n",
    "                np.sqrt(metricsObj.mean_squared_error(yTrain, yPredTrain)),\n",
    "                metricsObj.r2_score(yTrain, yPredTrain),\n",
    "                metricsObj.mean_absolute_percentage_error(yTrain, yPredTrain),\n",
    "                metricsObj.max_error(yTrain, yPredTrain)]\n",
    "  test_results = [metricsObj.mean_absolute_error(yTest, yPredTest),\n",
    "                np.sqrt(metricsObj.mean_squared_error(yTest, yPredTest)),\n",
    "                metricsObj.r2_score(yTest, yPredTest),\n",
    "                  metricsObj.mean_absolute_percentage_error(yTest, yPredTest),\n",
    "                metricsObj.max_error(yTest, yPredTest)]\n",
    "  resultsDF = pd.DataFrame({'Measure': measures_list, 'Train': train_results, 'Test':test_results})\n",
    "  return(resultsDF)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "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>Measure</th>\n",
       "      <th>Train</th>\n",
       "      <th>Test</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MAE</td>\n",
       "      <td>4.841439e-01</td>\n",
       "      <td>4.951278e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>RMSE</td>\n",
       "      <td>4.568885e+00</td>\n",
       "      <td>4.757416e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>R^2</td>\n",
       "      <td>7.570143e-03</td>\n",
       "      <td>8.245013e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>MAPE (%)</td>\n",
       "      <td>1.084499e+15</td>\n",
       "      <td>1.088088e+15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>MAX Error</td>\n",
       "      <td>7.316798e+02</td>\n",
       "      <td>7.291425e+02</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Measure         Train          Test\n",
       "0        MAE  4.841439e-01  4.951278e-01\n",
       "1       RMSE  4.568885e+00  4.757416e+00\n",
       "2        R^2  7.570143e-03  8.245013e-03\n",
       "3   MAPE (%)  1.084499e+15  1.088088e+15\n",
       "4  MAX Error  7.316798e+02  7.291425e+02"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show performance results\n",
    "resultsDF = performanceMetricsDF(metrics, y_train, y_pred_train, y_test, y_pred_test)\n",
    "resultsDF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\lotar\\anaconda3\\lib\\site-packages\\sklearn\\base.py:450: UserWarning: X does not have valid feature names, but DecisionTreeRegressor was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Residuals for DecisionTreeRegressor Model'}, xlabel='Predicted Value', ylabel='Residuals'>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Visualize the residuals\n",
    "residuals_vis = ResidualsPlot(tree_regr)\n",
    "residuals_vis.fit(X_train, y_train)  # Fit the training data to the visualizer\n",
    "residuals_vis.score(X_test, y_test)  # Evaluate the model on the test data\n",
    "residuals_vis.show() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\lotar\\anaconda3\\lib\\site-packages\\sklearn\\base.py:450: UserWarning: X does not have valid feature names, but DecisionTreeRegressor was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Prediction Error for DecisionTreeRegressor'}, xlabel='$y$', ylabel='$\\\\hat{y}$'>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Visualize the prediction error\n",
    "predError_vis = PredictionError(tree_regr)\n",
    "predError_vis.fit(X_train, y_train)  # Fit the training data to the visualizer\n",
    "predError_vis.score(X_test, y_test)  # Evaluate the model on the test data\n",
    "predError_vis.show() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similarly to what was observed using the Linear Regression Model, for the Decision Tree model the results obtained in the training set are similar to the ones obtained using the test set. Evethough, onde again, all obtained measures show us that the model predictive power can be considered weak. \n",
    "\n",
    "Initialy, one could infer that the Decision Tree Model should return more robust results. Although, this was not the case. One possible reason for such is the fact that the initial dataset is extremely inbalanced (many more observations reporting zero claims than higher than zero). As well as that, a higher number of observations, meaning, a larger dataset, could lead to better results."
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "d7a18bc08bf6b314cad8b0dd8f53415ad78d1015cc806d14e4873c235fb4e191"
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}