Random_Forest_Regression_Practice.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forest Regression\n",
    "\n",
    "A [random forest](https://en.wikipedia.org/wiki/Random_forest) is a meta estimator that fits a number of classifying [decision trees](https://en.wikipedia.org/wiki/Decision_tree_learning) on various sub-samples of the dataset and use averaging to improve the predictive accuracy and control over-fitting. The sub-sample size is always the same as the original input sample size but the samples are drawn with replacement (can be changed by user).\n",
    "\n",
    "Generally, Decision Tree and Random Forest models are used for classification task. However, the idea of Random Forest as a regularizing meta-estimator over single decision tree is best demonstrated by applying them to regresion problems. This way it can be shown that, **in the presence of random noise, single decision tree is prone to overfitting and learn spurious correlations while a properly constructed Random Forest model is more immune to such overfitting.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create some syntehtic data using scikit-learn's built-in regression generator"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 543,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 544,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Import make_regression method to generate artificial data samples \n",
    "from sklearn.datasets import make_regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### What is *make_regression* method?\n",
    "\n",
    "It is a convinient method/function from scikit-learn stable to generate a random regression problem. The input set can either be well conditioned (by default) or have a low rank-fat tail singular profile.\n",
    "\n",
    "The output is generated by applying a (potentially biased) random linear regression model with `n_informative` nonzero regressors to the previously generated input and some gaussian centered noise with some adjustable scale."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 545,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "n_samples = 100 # Number of samples\n",
    "n_features = 6 # Number of features\n",
    "n_informative = 3 # Number of informative features i.e. actual features which influence the output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 546,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X, y,coef = make_regression(n_samples=n_samples, n_features=n_features, n_informative=n_informative,\n",
    "                       random_state=None, shuffle=False,noise=20,coef=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Make a data frame and create basic visualizations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Data Frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 547,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>X5</th>\n",
       "      <th>X6</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.126157</td>\n",
       "      <td>1.537836</td>\n",
       "      <td>0.806673</td>\n",
       "      <td>-1.491675</td>\n",
       "      <td>0.299527</td>\n",
       "      <td>-1.474775</td>\n",
       "      <td>163.192326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-1.069024</td>\n",
       "      <td>0.054889</td>\n",
       "      <td>1.039015</td>\n",
       "      <td>-1.735270</td>\n",
       "      <td>1.821929</td>\n",
       "      <td>-1.468514</td>\n",
       "      <td>27.475965</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.768465</td>\n",
       "      <td>0.987083</td>\n",
       "      <td>1.031437</td>\n",
       "      <td>0.188459</td>\n",
       "      <td>0.777789</td>\n",
       "      <td>0.479504</td>\n",
       "      <td>119.033181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.656240</td>\n",
       "      <td>1.082382</td>\n",
       "      <td>-1.189696</td>\n",
       "      <td>0.407922</td>\n",
       "      <td>-0.365730</td>\n",
       "      <td>0.229083</td>\n",
       "      <td>-40.529855</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.014315</td>\n",
       "      <td>-0.393556</td>\n",
       "      <td>0.957417</td>\n",
       "      <td>-0.281082</td>\n",
       "      <td>1.336032</td>\n",
       "      <td>-1.126818</td>\n",
       "      <td>51.106814</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>-0.573993</td>\n",
       "      <td>-0.325107</td>\n",
       "      <td>1.364623</td>\n",
       "      <td>-0.250283</td>\n",
       "      <td>-0.188160</td>\n",
       "      <td>-1.172770</td>\n",
       "      <td>79.695119</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.157546</td>\n",
       "      <td>-1.190279</td>\n",
       "      <td>-1.062111</td>\n",
       "      <td>0.967536</td>\n",
       "      <td>-0.412225</td>\n",
       "      <td>1.300293</td>\n",
       "      <td>-152.172329</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.401044</td>\n",
       "      <td>-0.327829</td>\n",
       "      <td>1.166431</td>\n",
       "      <td>1.737604</td>\n",
       "      <td>-1.131857</td>\n",
       "      <td>1.081729</td>\n",
       "      <td>83.956168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.357468</td>\n",
       "      <td>1.440760</td>\n",
       "      <td>-0.421540</td>\n",
       "      <td>-0.379527</td>\n",
       "      <td>0.317447</td>\n",
       "      <td>0.841242</td>\n",
       "      <td>118.143163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>-1.690400</td>\n",
       "      <td>-0.298059</td>\n",
       "      <td>1.984504</td>\n",
       "      <td>-0.578445</td>\n",
       "      <td>-0.115120</td>\n",
       "      <td>-1.034285</td>\n",
       "      <td>32.072988</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         X1        X2        X3        X4        X5        X6           y\n",
       "0  0.126157  1.537836  0.806673 -1.491675  0.299527 -1.474775  163.192326\n",
       "1 -1.069024  0.054889  1.039015 -1.735270  1.821929 -1.468514   27.475965\n",
       "2 -0.768465  0.987083  1.031437  0.188459  0.777789  0.479504  119.033181\n",
       "3 -0.656240  1.082382 -1.189696  0.407922 -0.365730  0.229083  -40.529855\n",
       "4 -0.014315 -0.393556  0.957417 -0.281082  1.336032 -1.126818   51.106814\n",
       "5 -0.573993 -0.325107  1.364623 -0.250283 -0.188160 -1.172770   79.695119\n",
       "6  0.157546 -1.190279 -1.062111  0.967536 -0.412225  1.300293 -152.172329\n",
       "7  0.401044 -0.327829  1.166431  1.737604 -1.131857  1.081729   83.956168\n",
       "8  1.357468  1.440760 -0.421540 -0.379527  0.317447  0.841242  118.143163\n",
       "9 -1.690400 -0.298059  1.984504 -0.578445 -0.115120 -1.034285   32.072988"
      ]
     },
     "execution_count": 547,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1 = pd.DataFrame(data=X,columns=['X'+str(i) for i in range(1,n_features+1)])\n",
    "df2=pd.DataFrame(data=y,columns=['y'])\n",
    "df=pd.concat([df1,df2],axis=1)\n",
    "df.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Scatter plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 548,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEGCAYAAACZ0MnKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d130a6a080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEGCAYAAACZ0MnKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEjSEEEKYJkFDCCGEaf8/QZeZd2Cm0UYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d12f7220f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEGCAYAAACZ0MnKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nf8LAOmjZpFDvAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d12f7bd3c8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEGCAYAAACZ0MnKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCCF0+/8EU9M3nzo90wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d130fc82e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEGCAYAAACZ0MnKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d12eefd6a0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEGCAYAAACZ0MnKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3kHA7eMLwMsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d12f0c0908>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with plt.style.context(('seaborn-dark')):\n",
    "    for i,col in enumerate(df.columns[:-1]):\n",
    "        plt.figure(figsize=(6,4))\n",
    "        plt.grid(True)\n",
    "        plt.xlabel('Feature:'+col,fontsize=12)\n",
    "        plt.ylabel('Output: y',fontsize=12)\n",
    "        plt.scatter(df[col],df['y'],c='red',s=50,alpha=0.6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### It is clear from the scatter plots that some of the features influence the output while the others don't. This is the result of choosing a particular *n_informative* in the *make_regression* method"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Histograms of the feature space"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 549,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAawAAAEdCAYAAABQXlN8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAGvVJREFUeJzt3X1QVXXix/EPwkoEyiXDi0/Apjg+hLZZumaamtG6qQhm\nZO66mTouacWULWBrbY6CmjGa66I7jm2z2kYqjbrNqFQ+pWC704M2NA2tI/hwJXwAuQmVwu+Pfl1F\nAa8C95wvvl8zd+J+z7n3fM4R/XTuPQ9+5eXltQIAwObaWB0AAABvUFgAACNQWAAAI1BYAAAjUFgA\nACNQWAAAI1BYAAAj+KSwsrKyNGLECHXr1k3du3dXUlKSCgsL68yTnJwsh8NR5zFq1ChfxAMAGCDA\nFwv5+OOPNW3aNN19992qra1VRkaGxo8frwMHDigsLMwz3/Dhw7V69WrP87Zt2/oiHgDAAD4prNzc\n3DrPV69ercjISBUUFGj06NGe8cDAQDmdTl9EAgAYxpLvsNxut2pqauRwOOqM5+fnq0ePHhowYICe\nffZZlZWVWREPAGBDflZcS/DJJ5/U//73P+3atUv+/v6SpE2bNikoKEhRUVEqKSnRggULVFNTo127\ndikwMLDe9ykqKvJlbABAC4qJiWl0us8La+7cucrNzdW2bdsUHR3d4Hwul0uxsbFau3atxo0b57uA\nFisqKrrmH9rNgm1xCdviErbFJTfbtvDJd1g/S09PV25urrZu3dpoWUlSp06d1LlzZx0+fNg34QAA\ntuazwkpNTdV7772nrVu3qmfPntec/9SpU3K5XByEAQCQ5KPCmjNnjnJycrRu3To5HA6VlpZKkoKD\ngxUSEiK3261FixZp3LhxcjqdKikp0fz58xUeHq4xY8b4IiIAwOZ8Ulhr1qyRJMXHx9cZT01NVXp6\nuvz9/VVYWKh33nlHFRUVcjqdGjp0qN588021a9fOFxEBADbnk8IqLy9vdHpQUNBV52oBAHA5riUI\nADAChQUAMIJPD2sHcP1SPkhRxbkKhRaHWh1Fy0YtszoCbmLsYQEAjEBhAQCMQGEBAIxAYQEAjEBh\nAQCMQGEBAIxAYQEAjEBhAQCMQGEBAIxAYQEAjEBhAQCMQGEBAIxAYQEAjEBhAQCMQGEBAIxAYQEA\njEBhAQCMQGEBAIxAYQEAjEBhAQCMQGEBAIxAYQEAjEBhAQCMQGEBAIxAYQEAjEBhAQCMQGEBAIxA\nYQEAjEBhAQCMQGEBAIxAYQEAjEBhAQCM4JPCysrK0ogRI9StWzd1795dSUlJKiwsrDNPbW2tMjMz\n1atXL0VEROiRRx7RV1995Yt4AAAD+KSwPv74Y02bNk3bt2/Xli1bFBAQoPHjx+vs2bOeeZYvX66V\nK1dq8eLF+uijjxQeHq6EhARVVlb6IiIAwOYCfLGQ3NzcOs9Xr16tyMhIFRQUaPTo0aqtrVV2drZS\nUlIUHx8vScrOzlZMTIw2btyoqVOn+iImAMDGLPkOy+12q6amRg6HQ5JUXFys0tJSjRw50jNPUFCQ\n7rvvPh04cMCKiAAAm/HJHtaV0tLSFBsbq4EDB0qSSktLJUnh4eF15gsPD5fL5WrwfYqKiloupIVa\n63rdCLaFVHGuos5/rWSXPw+75LCD1rQtYmJiGp3u88KaO3euCgoKtG3bNvn7+zfpva61ciYqKipq\nlet1I9gWPwktDlXFuQqFtg+1Ooot/jz4vbjkZtsWPv1IMD09XZs2bdKWLVsUHR3tGXc6nZKksrKy\nOvOXlZWpY8eOvowIALApnxVWamqqp6x69uxZZ1pUVJScTqd27tzpGauurlZ+fr4GDRrkq4gAABvz\nyUeCc+bMUU5OjtatWyeHw+H5zio4OFghISHy8/NTcnKysrKyFBMTox49emjp0qUKDg7Wo48+6ouI\nAACb80lhrVmzRpI8h6z/LDU1Venp6ZKk5557TlVVVXrxxRdVXl6uAQMGKDc3V+3atfNFRACAzfmk\nsMrLy685j5+fn9LT0z0FBgDA5biWIADACBQWAMAIlpw4DJgg5YMUqyMAuAx7WAAAI1BYAAAjUFgA\nACNQWAAAI1BYAAAjUFgAACNQWAAAI1BYAAAjUFgAACNQWAAAI1BYAAAjUFgAACNQWAAAI1BYAAAj\nUFgAACNQWAAAI1BYAAAjUFgAACNQWAAAI1BYAAAjUFgAACNQWAAAI1BYAAAjUFgAACNQWAAAI1BY\nAAAjUFgAACNQWAAAI1BYAAAjUFgAACNQWAAAI1BYAAAjeFVYhw4daukcAAA0yqvCGj9+vIYMGaIV\nK1bo5MmTN7Sgffv26fHHH1fv3r3lcDi0fv36OtOTk5PlcDjqPEaNGnVDywIAtD5eFdbXX3+tuXPn\n6r///a8GDBighIQE5eTk6Pz5814v6LvvvlOfPn20aNEiBQUF1TvP8OHD9fXXX3seGzZs8Pr9AQCt\nm1eFFRAQoEceeURvvfWWCgsLlZCQoOXLl6tnz56aOXOmCgoKrvkecXFxevnllxUfH682bepfbGBg\noJxOp+cRFhZ2fWsDAGi1Aq5nZrfbrffff1+bNm3SiRMnlJiYqK5du2rGjBl6+OGHtXTp0iaFyc/P\nV48ePRQaGqohQ4Zo3rx5Cg8Pb3D+oqKiJi3Prlrret0IK7dFxbkKy5ZdHzvkscvvpl1y2EFr2hYx\nMTGNTveqsLZv366cnBx98MEHGjRokKZMmaJHHnlEt9xyiyRpxowZuvPOO5tUWKNGjdLYsWMVFRWl\nkpISLViwQOPGjdOuXbsUGBhY72uutXImKioqapXrdSOs3hahxaGWLftKFecqFNre+jx2+N20+vfC\nTm62beFVYb366quaNGmSMjIyFBERcdX0sLAwZWZmNinIhAkTPD/37dtXd911l2JjY7V9+3aNGzeu\nSe8NADCfV4W1f//+a84zZcqUJoe5XKdOndS5c2cdPny4Wd8XAGAm2544fOrUKblcLjmdTqujAABs\n4LoOumgKt9vt2VuqqanRsWPHdPDgQYWFhSksLEyLFi3SuHHj5HQ6VVJSovnz5ys8PFxjxozxVUQA\ngI35bA/rs88+07BhwzRs2DBVVVUpMzNTw4YNU0ZGhvz9/VVYWKgnnnhC99xzj5KTk9WjRw/t2LFD\n7dq181VEAICN+WwPa+jQoSovL29wem5urq+iAAAM5PUe1qZNm64a27hxY7OGAQCgIV4XVlZW1lVj\nr7/+erOGAQCgIV4X1r59+64ay8/Pb9YwAAA0xKvCWrFiRb3jf/3rX5s1DAAADfGqsJYsWVLveFOv\nHQgAgLcaPUpw9+7dkqSLFy9qz549qq2t9UwrLi5WSEhIy6YDAOD/NVpYzzzzjCSpurpas2fP9oz7\n+fnJ6XQ2uOcFAEBza7SwDh48KEmaOXOmVq9e7ZNAAADUx6vvsCgrAIDVvLrSRd++feXn51fvtC+/\n/LJZAwEAUB+vCuvKPazS0lKtWrVKiYmJLRIKAIAreVVY999/f71jEyZMUHJycrOHAgDgSjd8tfbA\nwECVlJQ0ZxYAABrk1R7WwoUL6zyvqqpSXl6eRo0a1SKhAAC4kleFdfz48TrPg4ODNWvWLCUlJbVI\nKAAAruRVYf3tb39r6RwAADTK6xs47t69W5s2bdLJkycVERGhCRMm6IEHHmjJbAAAeHh9tfZp06Yp\nLCxMcXFxuu222zR9+vQGr+IOAEBz8/ojwS1btqhPnz6esaSkJCUkJHiuNwgAQEvy+rD2O+64o87z\n6OjoBq9+AQBAc/NqDystLU3PPPOM0tLS1LlzZx0/flxLlixRenq6ampqPPO1aXPDp3UBHikfpEiS\nKs5VKLQ41OI0uNzPfzZWstPvxbJRy6yOcFPxqrBSUn76Jd24caP8/Pw898XasGGDUlJSVFtbKz8/\nP505c6blkgIAbmpeFdYXX3zR0jkAAGiUV5/hbd68WZGRkVc9tmzZUuc5AAAtxavCaujOwkuXLm3W\nMAAANKTRjwR3794tSbp48aL27Nnj+e5KkoqLixUSEtKy6QAA+H+NFtbP51hVV1dr9uzZnnE/Pz85\nnc4G97wAAGhujRbWwYMHJUkzZ8686iaOAAD4klffYVFWAACreXVYe9++fRu8qsWXX37ZrIEAAKiP\nV4V15R5WaWmpVq1apcTExBYJBQDAlbwqrPvvv7/esQkTJig5ObnZQwEAcKUbvvhfYGCgSkpKmjML\nAAAN8moPa+HChXWeV1VVKS8vT6NGjWqRUAAAXMmrwjp+/Hid58HBwZo1a5aSkpJaJBQAAFfy+gaO\nTbVv3z6tWLFCX3zxhVwul1auXKnJkyd7ptfW1mrRokV66623VF5ergEDBmjp0qXq3bt3k5cNADDf\nNb/DunDhgtavX68ZM2YoMTFRM2bM0Lp16/Tjjz9e14K+++479enTR4sWLVJQUNBV05cvX66VK1dq\n8eLF+uijjxQeHq6EhARVVlZe13IAAK1To4VVUVGhuLg4vfLKKwoICFD//v0VEBCgV199VXFxcaqo\nqPB6QXFxcXr55ZcVHx9/1Y0ea2trlZ2drZSUFMXHx6tPnz7Kzs6W2+3Wxo0bb2zNAACtSqMfCc6f\nP1+33367tm7dquDgYM+42+3WU089pfnz5+v1119vcoji4mKVlpZq5MiRnrGgoCDdd999OnDggKZO\nndrkZQAAzNZoYb3//vvKy8urU1aSFBISotdee01xcXHNUlilpaWSpPDw8Drj4eHhcrlcDb6uqKio\nycu2o9a6Xt6qOFdR7883O7bFJXbZFnb4u2qHDM0lJiam0emNFta5c+fUuXPneqd16dLF8u+XrrVy\nJioqKmqV63U9QotDJf30j1Jo+1CL09gD2+ISO20Lq/+u3mz/XjT6HVZ0dLT27NlT77Tdu3crOjq6\nWUI4nU5JUllZWZ3xsrIydezYsVmWAQAwW6OFNWvWLP3xj3/U5s2bVVNTI0mqqanR5s2b9fTTT+vp\np59ulhBRUVFyOp3auXOnZ6y6ulr5+fkaNGhQsywDAGC2Rj8SnDx5ss6cOaNZs2Zp+vTp6tChg06f\nPq3AwED96U9/0u9+9zuvF+R2u3X48GFJP5XesWPHdPDgQYWFhalbt25KTk5WVlaWYmJi1KNHDy1d\nulTBwcF69NFHm7aGAIBW4ZonDj/zzDN68skn9cknn+j06dPq0KGD7r33XrVv3/66FvTZZ59p7Nix\nnueZmZnKzMzUpEmTlJ2dreeee05VVVV68cUXPScO5+bmql27dte/VgCAVserK120a9dODz74YJMW\nNHToUJWXlzc43c/PT+np6UpPT2/ScgAArdMNX60dAABforAAAEagsAAARqCwAABGoLAAAEagsAAA\nRqCwAABGoLAAAEagsAAARqCwAABGoLAAAEagsAAARqCwAABGoLAAAEagsAAARqCwAABGoLAAAEag\nsAAARqCwAABGoLAAAEagsAAARqCwAABGoLAAAEagsAAARqCwAABGoLAAAEagsAAARqCwAABGoLAA\nAEagsAAARgiwOgCklA9SPD9XnKtQaHGohWkAwJ7YwwIAGIHCAgAYgcICABiBwgIAGME2hZWZmSmH\nw1Hn0bNnT6tjAQBswlZHCcbExOjf//6357m/v7+FaQAAdmKrwgoICJDT6bQ6BgDAhmzzkaAkHTly\nRL169VK/fv301FNP6ciRI1ZHAgDYhF95eXmt1SEkKS8vT263WzExMTp16pRee+01FRUVqaCgQLfd\ndlu9rykqKvJxypaRcSjD6ggADDY3dq7VEZpFTExMo9Nt85HgQw89VOf5vffeq/79++vtt9/W7Nmz\n633NtVbOFJdf2aLiXIVC23OlC4ltcTm2xSVsi0t+3hat5d/Ca7HVR4KXCw4OVq9evXT48GGrowAA\nbMC2hVVdXa2ioiIOwgAASLLRR4J//vOf9Zvf/EZdu3b1fId1/vx5TZo0yepoAAAbsE1hnThxQtOn\nT9fp06d1++2365577lFeXp4iIyOtjgYAsAHbFNbatWutjgAAsDHbfocFAMDlKCwAgBEoLACAESgs\nAIARKCwAgBEoLACAESgsAIARKCwAgBEoLACAESgsAIARKCwAgBEoLACAEWxz8VsrpHyQYnUEAICX\n2MMCABiBwgIAGIHCAgAYgcICABiBwgIAGIHCAgAYgcICABiBwgIAGIHCAgAYgcICABiBwgIAGIHC\nAgAYgcICABiBwgIAGIHCAgAYgcICABiBwgIAGOGmvuMwALQGdrl7+rJRy1r0/dnDAgAYgcICABiB\nwgIAGIHCAgAYgcICABjBdoW1Zs0a9evXT06nUw888ID2799vdSQAgA3YqrByc3OVlpamF154QXv2\n7NHAgQM1ceJEHT161OpoAACL+ZWXl9daHeJnDz74oPr27as33njDM3b33XcrPj5er7zyioXJAABW\ns80e1g8//KDPP/9cI0eOrDM+cuRIHThwwKJUAAC7sE1hnT59WhcvXlR4eHid8fDwcH377bcWpQIA\n2IVtCgsAgMbYprA6dOggf39/lZWV1RkvKytTx44dLUoFALAL2xRW27Ztddddd2nnzp11xnfu3KlB\ngwZZlAoAYBe2ulr7rFmzNHPmTA0YMECDBg3S2rVrdfLkSU2dOtXqaAAAi9lmD0uSEhMTlZmZqdde\ne01Dhw5VQUGB3n33XUVGRlodzafOnj2rF198Uffee68iIiLUt29fPf/88zpz5ozV0Szxj3/8Q2PG\njFFkZKQcDoeKi4utjuQznEj/k3379unxxx9X79695XA4tH79eqsjWSYrK0sjRoxQt27d1L17dyUl\nJamwsNDqWD5hq8KSpOnTp+vQoUP69ttvtXv3bg0ZMsTqSD7ncrnkcrn06quvav/+/Vq9erX279+v\nadOmWR3NEufPn9fIkSOVlpZmdRSf4kT6S7777jv16dNHixYtUlBQkNVxLPXxxx9r2rRp2r59u7Zs\n2aKAgACNHz9eZ8+etTpai7PVicNo2I4dO5SUlKTi4mK1b9/e6jiW+OyzzzRixAh98cUXioqKsjpO\ni+NE+vp16dJFS5Ys0eTJk62OYgtut1uRkZFav369Ro8ebXWcFmW7PSzUr7KyUoGBgbr11lutjgIf\n4ER6eMvtdqumpkYOh8PqKC2OwjJAeXm5Fi5cqClTpiggwFbHyaCFcCI9vJWWlqbY2FgNHDjQ6igt\njsLyoQULFsjhcDT62Lt3b53XuN1uTZo0SZ06ddL8+fMtSt78bmRbAKhr7ty5Kigo0D//+U/5+/tb\nHafF8b/rPpScnKzHHnus0Xm6du3q+dntdmvixImSpJycHN1yyy0tms+Xrndb3Gw4kR7Xkp6ertzc\nXG3dulXR0dFWx/EJCsuHOnTooA4dOng1b2VlpSZOnKja2lpt3LhRISEhLZzOt65nW9yMLj+Rfvz4\n8Z7xnTt3aty4cRYmgx2kpqbqvffe09atW9WzZ0+r4/gMhWVDlZWVSkxMVGVlpdavX6/z58/r/Pnz\nkqSwsDC1bdvW4oS+VVpaqtLSUn3zzTeSpK+//loVFRXq1q2bwsLCLE7XcjiR/hK3263Dhw9Lkmpq\nanTs2DEdPHhQYWFh6tatm8XpfGvOnDnKycnRunXr5HA4VFpaKkkKDg5udf9jeyUOa7ehvXv3auzY\nsfVO27p1q4YOHerjRNbKzMzU4sWLrxpfuXJlqz+0ec2aNVq+fLlKS0vVu3dvZWRk3JTnJjb0d2LS\npEnKzs62IJF1GjoaMDU1Venp6T5O41sUFgDACBwlCAAwAoUFADAChQUAMAKFBQAwAoUFADAChQUA\nMAKFBQAwAoUFXENsbKwiIiLUpUsXz8Plct3w++3du1d9+vRpxoTX9tJLLykhIaHOWFpampKSkiT9\ndDuTKVOmKDY2lgsPw7YoLMAL77zzjo4fP+55dOrUybIsFy5cuO7XvPTSSzpy5IjWrVsnSfrkk0/0\nr3/9S1lZWZ55fv3rX+vvf/+7nE5ns2UFmhOFBdyg//znP4qLi1NkZKSGDBlSZ69k3bp1GjhwoLp2\n7ar+/fvrzTfflPTTrd4nTpwol8tVZ28tOTlZCxYs8Lz+yr2w2NhYLVu2TPfdd586d+6sCxcuyOVy\n6fe//726d++ufv36adWqVQ1mvfXWW7V8+XLNmzdPJSUlmj17tv7yl7+oS5cukn662O7TTz+twYMH\n3xS3qYCZKCzgBpw4cUKPPfaY5syZoyNHjmjBggWaMmWKTp06JemnGy3m5OTo6NGjWrlypebOnavP\nP/9cwcHB2rBhgzp16nTde2sbN27Uu+++q+LiYrVp00aPP/647rzzTn311VfasmWLsrOz9eGHH0qS\n8vPzFRkZWef1w4YNU3x8vIYPH66OHTvqySefbNZtArQ0CgvwwuTJkxUZGanIyEg98cQTevfdd/XQ\nQw8pLi5Obdq00YgRI/SrX/1KO3bskCQ9/PDD+uUvfyk/Pz/df//9GjFihPLz85uUYebMmeratauC\ngoL06aef6vTp00pNTVXbtm0VHR2tP/zhD9q0aZMkafDgwSopKbnqPQYPHqwzZ85o4sSJ8vPza1Ie\nwNe4vQjghfXr12v48OGe5y+88II2b96sbdu2ecYuXLjguZJ+Xl6eFi9erG+++UY1NTWqqqpq8oEW\nl9/Q8ujRo3K5XHX2ompqajR48OAGX3/mzBnNmzdPycnJysjIUHx8fINX/gbsiMICbkCXLl2UlJSk\nN95446pp33//vaZMmaJVq1bpt7/9rX7xi1/oiSee8Eyvb88mODjYc88zSfr222+vmufy13Xp0kVR\nUVH69NNPvc6clpamBx98UJmZmTp58qTmzZunFStWeP16wGp8JAjcgMcee0zbtm3Thx9+qIsXL6q6\nulp79+7V8ePH9cMPP+j7779Xhw4dFBAQoLy8PO3cudPz2o4dO+rMmTOqqKjwjMXGxiovL09nz55V\naWnpNe/xNGDAAIWEhGjZsmWqqqrSxYsXVVhY2GCB7dixQ7t27VJGRoYkacmSJXr//fe1Z88ezzzf\nf/+9qqurJUk//vijqqurVVvL3YdgHxQWcAO6du2qt99+W6+//rq6d++uvn37asWKFaqpqVG7du20\nePFiTZ06VVFRUdqwYYNGjx7teW3Pnj01YcIE3XXXXYqMjJTL5VJSUpLuvPNO9evXTwkJCVedM3Ul\nf39/5eTk6NChQ+rfv7/uuOMOPfvsszp37pwkaf/+/Z4jACsrK/X8889r8eLFnjs0h4eHa8GCBUpJ\nSVFVVZUk6Z577lFERIROnDihxMRERURE1Ps9GGAVbuAIADACe1gAACNQWAAAI1BYAAAjUFgAACNQ\nWAAAI1BYAAAjUFgAACNQWAAAI/wfokusVRRQIggAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d12ef0d8d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAawAAAEaCAYAAABNW2PEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAoUFADAFCgsAYAoUFgDAFCgsAIAp/B/cJrs6r4HoIAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d12eac0a58>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbkAAAEaCAYAAACM3CloAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d12f2edcc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAawAAAEaCAYAAABNW2PEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGOH/APuGTM4SmlMPAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d12f7af898>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAawAAAEaCAYAAABNW2PEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAGuhJREFUeJzt3XtQ1XXi//EXSZKBAktcDEQ2xLyE2po63i8ZrVkqGiHW\naqaN4SWZ0hVtnd0cFe+ruYrOuO3urOwuheyKOZOXwkveqknDxqbBdQQviCiCoJir8Pujr+cnIniS\nw/l83vh8zDDDeX8+53Ne5z3Ai885n/P5eJSUlFQJAACbe8jqAAAAOIPCAgAYgcICABiBwgIAGIHC\nAgAYgcICABiBwgIAGIHCAgAYgcJqJHJzc62OYCvMR3XMR03MSXUmzAeFBQAwAoUFADAChQUAMAKF\nBQAwAoUFADAChQUAMAKFBQAwAoUFADCCp9UBANQtaWdSvbdRerlUvnm+9d7OysEr670N4H6xhwUA\nMAKFBQAwAoUFADAChQUAMAKFBQAwAoUFADAChQUAMAKFBQAwAoUFADAChQUAMAKFBQAwAoUFADAC\nhQUAMAKFBQAwAoUFADCCWwprxYoVGjhwoFq1aqXIyEjFx8fr2LFj1dapqqpSSkqK2rVrp5CQEA0d\nOlTff/+9O+IBAAzglsL64osvNGHCBG3btk1ZWVny9PTUiBEjdOnSJcc6q1at0po1a7R48WJ9/vnn\nCgwMVGxsrMrKytwREQBgc2654nBmZma12+vXr1d4eLgOHjyoIUOGqKqqSqmpqUpKStLw4cMlSamp\nqYqKilJGRobGjx/vjpgAABuz5D2s8vJyVVZWys/PT5KUl5enwsJCDRo0yLFOs2bN1KtXLx06dMiK\niAAAm3HLHtadkpOTFR0dre7du0uSCgsLJUmBgYHV1gsMDFRBQUGt28nNzW24kAZiPqprLPNRernU\nNttpLHN6S2N7PvVl9XxERUXVudzthTVnzhwdPHhQn376qZo0aVKvbd3ryT1IcnNzmY/bNKb58M3z\nrfc2Si+XyrdF/bfTWOZUalw/I65gwny49SXB2bNna9OmTcrKylJERIRjPDg4WJJUVFRUbf2ioiIF\nBQW5MyIAwKbcVlizZs1ylFXbtm2rLWvdurWCg4OVnZ3tGLt27ZoOHDigHj16uCsiAMDG3PKS4IwZ\nM5Senq6NGzfKz8/P8Z6Vt7e3fHx85OHhocTERK1YsUJRUVFq06aNli1bJm9vb7388svuiAgAsDm3\nFNaGDRskyXHI+i2zZs3S7NmzJUnTp09XRUWFZs6cqZKSEnXt2lWZmZlq3ry5OyICAGzOLYVVUlJy\nz3U8PDw0e/ZsR4EBAHA7ziUIADAChQUAMAKFBQAwAoUFADAChQUAMAKFBQAwAoUFADAChQUAMAKF\nBQAwAoUFADAChQUAMAKFBQAwAoUFADAChQUAMAKFBQAwAoUFADAChQUAMAKFBQAwAoUFADAChQUA\nMAKFBQAwAoUFADAChQUAMAKFBQAwgqfVAYA7Je1Mqvc2Si+XyjfPt17bWDl4Zb1zAHAd9rAAAEag\nsAAARqCwAABGoLAAAEagsAAARqCwAABGoLAAAEagsAAARqCwAABG4EwXAJzmirOQuAJnIXkwsYcF\nADAChQUAMAKFBQAwAoUFADCC2wpr3759Gj16tNq3by8/Pz+lpaVVW56YmCg/P79qX4MHD3ZXPACA\nzbntKMErV66oQ4cOSkhI0FtvvXXXdQYMGKD169c7bjdt2tRd8QAANue2woqJiVFMTIwkafLkyXdd\nx8vLS8HBwe6KBAAwiK3ewzpw4IDatGmjrl276u2331ZRUZHVkQAANmGbDw4PHjxYL730klq3bq38\n/HzNnz9fw4YN065du+Tl5WV1PACAxWxTWKNGjXJ837FjR3Xp0kXR0dHatm2bhg0bdtf75Obmuiue\nERrLfJReLrXFduwyn3aZDzsZnzneNRs6Wr+7z4me45ocNmH1z3xUVFSdy21TWHdq2bKlHn/8cZ04\ncaLWde715B4kubm5jWY+fPN8672N0sul8m1Rv+3YZT7tMh+NTWP6GXEFE/6G2Oo9rNtduHBBBQUF\nHIQBAJDkxj2s8vJyx95SZWWlTp8+rZycHPn7+8vf31+LFi3SsGHDFBwcrPz8fM2bN0+BgYF68cUX\n3RURAGBjbtvDOnz4sPr166d+/fqpoqJCKSkp6tevnxYuXKgmTZro2LFjGjNmjJ555hklJiaqTZs2\n2r59u5o3b+6uiAAAG3PbHlbfvn1VUlJS6/LMzEx3RQEAGMi272EBAHA7pwrr6NF6HvsJAEA9OVVY\nI0aMUO/evbV69WqdO3euoTMBAFCDU4X1ww8/aM6cOfr666/VtWtXxcbGKj09XVevXm3ofAAASHKy\nsDw9PTV06FD97W9/07FjxxQbG6tVq1apbdu2mjRpkg4ePNjQOQEAD7ifddBFeXm5tm7dqk2bNuns\n2bMaOXKkIiMj9eabb2rGjBkNlREAAOcOa9+2bZvS09O1c+dO9ejRQ2PHjtXQoUP1yCOPSJLefPNN\nPfXUU1q2bFmDhgUAPLicKqz3339fCQkJWrhwoUJCQmos9/f3V0pKisvDAQBwi1OFtX///nuuM3bs\n2HqHAQCgNnxwGABgBAoLAGAECgsAYAQKCwBgBKcLa9OmTTXGMjIyXBoGAIDaOF1YK1asqDG2fPly\nl4YBAKA2ThfWvn37aowdOHDApWEAAKiNU4W1evXqu47/6U9/cmkYAABq41RhLVmy5K7jnIoJAOAu\ndZ7pYvfu3ZKkmzdvas+ePaqqqnIsy8vLk4+PT8OmAwDg/9RZWNOmTZMkXbt2TVOnTnWMe3h4KDg4\nuNY9LwAAXK3OwsrJyZEkTZo0SevXr3dLIAAA7sap97AoKwCA1Zw6W3vHjh3l4eFx12XfffedSwMB\nAHA3ThXWnXtYhYWFWrdunUaOHNkgoQAAuJNThdWnT5+7jo0aNUqJiYkuDwUAwJ3u++S3Xl5eys/P\nd2UWAABq5dQe1oIFC6rdrqio0I4dOzR48OAGCQUAwJ2cKqwzZ85Uu+3t7a0pU6YoPj6+QUIBAHAn\npwpr7dq1DZ0DAIA6OVVY0k+nadq0aZPOnTunkJAQjRo1Sv3792/IbAAAODh9tvYJEybI399fMTEx\n+sUvfqGJEyfWehZ3AABczemXBLOystShQwfHWHx8vGJjYx3nGwQAoCE5fVj7E088Ue12RERErWe/\nAADA1ZwqrOTkZE2bNk3//e9/VVFRoePHj2v69OmaPXu2KisrHV8AADQUp14STEpKkiRlZGTIw8PD\ncV2sjz/+WElJSaqqqpKHh4eKi4sbLikA4IHmVGF9++23DZ0DAIA6OfWS4ObNmxUeHl7jKysrq9pt\nAAAailOFVduVhZctW+bSMAAA1KbOlwR3794tSbp586b27NnjeO9KkvLy8uTj49Ow6QAA+D91Ftat\nz1hdu3ZNU6dOdYx7eHgoODi41j0vAABcrc6XBHNycpSTk6O4uDjH9zk5Ofr222+1fft2vfDCC04/\n0L59+zR69Gi1b99efn5+SktLq7a8qqpKKSkpateunUJCQjR06FB9//339/esAACNjlPvYd15xeH7\nceXKFXXo0EGLFi1Ss2bNaixftWqV1qxZo8WLF+vzzz9XYGCgYmNjVVZWVu/HBgCYz6nD2jt27Fjr\nWS2+++47px4oJiZGMTExkqTJkydXW1ZVVaXU1FQlJSVp+PDhkqTU1FRFRUUpIyND48ePd+oxAACN\nl1OFdeceVmFhodatW6eRI0e6JEReXp4KCws1aNAgx1izZs3Uq1cvHTp0iMICADhXWH369Lnr2KhR\no5SYmFjvEIWFhZKkwMDAauOBgYEqKCio9X65ubn1fuzGpLHMR+nlUltsZ3xm4/pHyVXz2pjUd04a\ny+/cLVY/n6ioqDqXO309rDt5eXkpPz//fu/uEvd6cg+S3NzcRjMfvnm+9d5G6eVS+bao/3YaC+aj\nJlfMSWP5nZPM+BviVGEtWLCg2u2Kigrt2LFDgwcPdkmI4OBgSVJRUZFatWrlGC8qKlJQUJBLHgMA\nYDanjhI8c+ZMta8ff/xRU6ZMUWpqqktCtG7dWsHBwcrOznaMXbt2TQcOHFCPHj1c8hgAALM5fQHH\n+iovL9eJEyckSZWVlTp9+rRycnLk7++vVq1aKTExUStWrFBUVJTatGmjZcuWydvbWy+//HK9HxsA\nYL57FtaNGzeUnp6uXbt26eLFiwoICFD//v0VHx+vhx9+2OkHOnz4sF566SXH7ZSUFKWkpCghIUGp\nqamaPn26KioqNHPmTJWUlKhr167KzMxU8+bN7++ZAQAaFY+SkpKq2haWlpYqNjZW+fn5eu655xQS\nEqJz585p586dCgsL03/+8x/5+vJGrh2Y8Iaps5J2JtV7GxxkUB3zUZMr5mTl4JUuSmM9E/6G1LmH\nNW/ePD322GPasmWLvL29HePl5eV64403NG/ePC1fvrzBQwIAUOdBF1u3btXy5curlZUk+fj4aOnS\npfrkk08aNBwAALfUWViXL1/W448/ftdloaGhnOcPAOA2dRZWRESE9uzZc9dlu3fvVkRERENkAgCg\nhjoLa8qUKXrrrbe0efNmVVZWSvrpkPTNmzdr8uTJNU5iCwBAQ6nzoItXX31VxcXFmjJliiZOnKiA\ngABdvHhRXl5e+u1vf6vXXnvNXTkBAA+4e34Oa9q0aXr99df15ZdfOj6H1a1bN7Vo0cId+QAAkOTk\nmS6aN2+uZ599tqGzAABQK6fOJQgAgNUoLACAESgsAIARKCwAgBEoLACAESgsAIARKCwAgBEoLACA\nESgsAIARKCwAgBEoLACAESgsAIARKCwAgBEoLACAESgsAIARKCwAgBGcuoAjAKCmpJ1JVkeQJK0c\nvNLqCG7BHhYAwAgUFgDACBQWAMAIFBYAwAgUFgDACBQWAMAIFBYAwAgUFgDACBQWAMAIFBYAwAgU\nFgDACBQWAMAIFBYAwAgUFgDACLYprJSUFPn5+VX7atu2rdWxAAA2YavrYUVFRemTTz5x3G7SpImF\naQAAdmKrwvL09FRwcLDVMQAANmSblwQl6eTJk2rXrp06deqkN954QydPnrQ6EgDAJmxTWM8884zW\nrl2rjIwMffDBByosLFRMTIyKi4utjgYAsAGPkpKSKqtD3M2VK1fUuXNnJSUlaerUqXddJzc3182p\nGreFRxdaHQHAfZgTPcfqCC4RFRVV53JbvYd1O29vb7Vr104nTpyodZ17PbkHSW5ubr3nwzfP10Vp\nrFd6uVS+LRrP86kv5qOmxjQnrvhb6Iq/IQ3NNi8J3unatWvKzc3lIAwAgCQb7WH97ne/069//WuF\nhYXpwoULWrp0qa5evaqEhASrowEAbMA2hXX27FlNnDhRFy9e1GOPPaZnnnlGO3bsUHh4uNXRAAA2\nYJvC+vDDD62OAACwMdu+hwUAwO0oLACAESgsAIARKCwAgBEoLACAESgsAIARKCwAgBEoLACAESgs\nAIARKCwAgBEoLACAESgsAIARKCwAgBEoLACAEWxzeRErJO1MsjqCy5ReLm1Ul7gHgDuxhwUAMAKF\nBQAwAoUFADAChQUAMAKFBQAwAoUFADAChQUAMAKFBQAwAoUFADDCA32mCwBoDFxx1h5XnC1n5eCV\n9c5RF/awAABGoLAAAEagsAAARqCwAABGoLAAAEagsAAARqCwAABGoLAAAEagsAAARqCwAABGoLAA\nAEagsAAARqCwAABGoLAAAEagsAAARrBdYW3YsEGdOnVScHCw+vfvr/3791sdCQBgA7YqrMzMTCUn\nJ+vdd9/Vnj171L17d8XFxenUqVNWRwMAWMxWhbVmzRqNGTNG48aN05NPPqmlS5cqODhYH374odXR\nAAAW87Q6wC3Xr1/XkSNHNG3atGrjgwYN0qFDhxrkMRv6cs4AANexzR7WxYsXdfPmTQUGBlYbDwwM\n1Pnz5y1KBQCwC9sUFgAAdbFNYQUEBKhJkyYqKiqqNl5UVKSgoCCLUgEA7MI2hdW0aVN16dJF2dnZ\n1cazs7PVo0cPi1IBAOzCNgddSNKUKVM0adIkde3aVT169NCHH36oc+fOafz48VZHAwBYzDZ7WJI0\ncuRIpaSkaOnSperbt68OHjyojz76SOHh4bXe5+2331aXLl0UEhKiyMhIJSQk6IcffnBjavu4dOmS\nZs6cqW7duikkJEQdO3bUO++8o+LiYqujWeavf/2rXnzxRYWHh8vPz095eXlWR3I7Poz//+3bt0+j\nR49W+/bt5efnp7S0NKsjWWbFihUaOHCgWrVqpcjISMXHx+vYsWNWx6qTrQpLkiZOnKijR4/q/Pnz\n2r17t3r37l3n+k8//bTWrl2rQ4cOadOmTaqqqtKIESP0v//9z02J7aOgoEAFBQV6//33tX//fq1f\nv1779+/XhAkTrI5mmatXr2rQoEFKTk62Oool+DB+dVeuXFGHDh20aNEiNWvWzOo4lvriiy80YcIE\nbdu2TVlZWfL09NSIESN06dIlq6PVyqOkpKTK6hCu9N1336lPnz766quvFBUVZXUcy23fvl3x8fHK\ny8tTixYtrI5jmcOHD2vgwIH69ttv1bp1a6vjuM2zzz6rjh076oMPPnCM/epXv9Lw4cP1+9//3sJk\n1gsNDdWSJUv06quvWh3FFsrLyxUeHq60tDQNGTLE6jh3Zbs9rPq4cuWK0tLSFBYWVufLiA+SsrIy\neXl56dFHH7U6Ctzs1ofxBw0aVG28IT+MD3OVl5ersrJSfn5+VkepVaMorA0bNig0NFShoaHauXOn\nsrKy5OXlZXUsy5WUlGjBggUaO3asPD1tdXwN3IAP4+PnSE5OVnR0tLp37251lFrZsrDmz58vPz+/\nOr/27t3rWD8uLk579uzR1q1bFRkZqXHjxunq1asWPgPX+rnzIf3031JCQoJatmypefPmWZS8YdzP\nfACo3Zw5c3Tw4EH9/e9/V5MmTayOUytb/tudmJioV155pc51wsLCHN/7+vrK19dXkZGR6tatmyIi\nIpSVlaXRo0c3dFS3+LnzUV5erri4OElSenq6HnnkkQbN524/dz4eVHwYH86YPXu2MjMztWXLFkVE\nRFgdp062LKyAgAAFBATc132rqqpUVVWl69evuziVdX7OfJSVlSkuLk5VVVXKyMiQj49PA6dzv/r8\nfDxIbv8w/ogRIxzj2dnZGjZsmIXJYBezZs3Sv//9b23ZskVt27a1Os492bKwnHXixAllZWVpwIAB\nCggI0NmzZ/XHP/5RTZs21fPPP291PLcrKyvTyJEjVVZWprS0NF29etXx0qi/v7+aNm1qcUL3Kyws\nVGFhoY4fPy5J+uGHH1RaWqpWrVrJ39/f4nQNjw/jV1deXq4TJ05IkiorK3X69Gnl5OTI399frVq1\nsjide82YMUPp6enauHGj/Pz8VFhYKEny9va27T+6Rh/Wfvr0aSUlJenIkSMqLS1VUFCQevXqpZkz\nZxrx34Kr7d27Vy+99NJdl23ZskV9+/Z1cyLrpaSkaPHixTXG16xZ88AczrxhwwatWrVKhYWFat++\nvRYuXHjPzzc2VrX9jiQkJCg1NdWCRNap7WjAWbNmafbs2W5O4xyjCwsA8OCw5VGCAADcicICABiB\nwgIAGIHCAgAYgcICABiBwgIAGIHCAgAYgcIC7iE6OlohISGOKwKEhoaqoKDgvre3d+9edejQwYUJ\n7+29995TbGxstbHk5GTFx8dLkvLy8uTn51ftOS5ZssStGYF7MfrUTIC7/Otf/9KAAQOsjiFJunHj\nxs++XMx7772n3r17a+PGjXrttdf05Zdf6p///Kf2799fbb28vDwuRQPbYg8LuE9fffWVYmJiFB4e\nrt69e1e7pMnGjRvVvXt3hYWFqXPnzvrLX/4i6aeLjMbFxamgoKDa3lpiYqLmz5/vuP+de2HR0dFa\nuXKlevXqpccff1w3btxQQUGBfvOb3ygyMlKdOnXSunXras366KOPatWqVZo7d67y8/M1depU/eEP\nf1BoaGgDzAzQMCgs4D6cPXtWr7zyimbMmKGTJ09q/vz5Gjt2rC5cuCDpp4skpqen69SpU1qzZo3m\nzJmjI0eOyNvbWx9//LFatmypM2fO6MyZM2rZsqVTj5mRkaGPPvpIeXl5euihhzR69Gg99dRT+v77\n75WVlaXU1FR99tlnkqQDBw7UuOp2v379NHz4cA0YMEBBQUF6/fXXazxGdHS0OnTooMmTJ+vixYv1\nmyTAxSgswAmvvvqqwsPDFR4erjFjxuijjz7Sc889p5iYGD300EMaOHCgnn76aW3fvl2S9Pzzz+uX\nv/ylPDw81KdPHw0cOFAHDhyoV4ZJkyYpLCxMzZo10zfffKOLFy9q1qxZatq0qSIiIjRu3Dht2rRJ\nktSzZ0/l5+fX2EbPnj1VXFysuLg4eXh4OMYDAgKUnZ2to0ePateuXSovL9ebb75Zr7yAq/FiNeCE\ntLS0au9hvfvuu9q8ebM+/fRTx9iNGzccZ8TfsWOHFi9erOPHj6uyslIVFRX1PtDi9otSnjp1SgUF\nBdX2oiorK9WzZ89a719cXKy5c+cqMTFRCxcu1PDhwx1n7Pbx8dHTTz8tSQoKCtLSpUv15JNPqqys\nTM2bN69XbsBVKCzgPoSGhio+Pl4ffPBBjWU//vijxo4dq3Xr1umFF17Qww8/rDFjxjiW375nc4u3\nt7fj2mWSdP78+Rrr3H6/0NBQtW7dWt98843TmZOTk/Xss88qJSVF586d09y5c7V69eq7rnvrsSor\nK53ePtDQeEkQuA+vvPKKPv30U3322We6efOmrl27pr179+rMmTO6fv26fvzxRwUEBMjT01M7duxQ\ndna2475BQUEqLi5WaWmpYyw6Olo7duzQpUuXVFhYeM9rM3Xt2lU+Pj5auXKlKioqdPPmTR07dqzW\nAtu+fbt27dqlhQsXSpKWLFmirVu3as+ePZKkr7/+Wrm5uaqsrFRxcbFmzZqlPn36yNfXt75TBbgM\nhQXch7CwMP3jH//Q8uXLFRkZqY4dO2r16tWqrKxU8+bNtXjxYo0fP16tW7fWxx9/rCFDhjju27Zt\nW40aNUpdunRReHi4CgoKFB8fr6eeekqdOnVSbGxsjc9M3alJkyZKT0/X0aNH1blzZz3xxBN6++23\ndfnyZUnS/v37HUcAlpWV6Z133tHixYsdV1kODAzU/PnzlZSUpIqKCp08eVKjRo1SWFiYevbsqaZN\nm+rPf/5zA80ecH+4gCMAwAjsYQEAjEBhAQCMQGEBAIxAYQEAjEBhAQCMQGEBAIxAYQEAjEBhAQCM\nQGEBAIzw/wArHabfIz0KMwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d12f713908>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbkAAAEaCAYAAACM3CloAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d130ad97b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with plt.style.context(('fivethirtyeight')):\n",
    "    for i,col in enumerate(df.columns[:-1]):\n",
    "        plt.figure(figsize=(6,4))\n",
    "        plt.grid(True)\n",
    "        plt.xlabel('Feature:'+col,fontsize=12)\n",
    "        plt.ylabel('Output: y',fontsize=12)\n",
    "        plt.hist(df[col],alpha=0.6,facecolor='g')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## How will a Decision Tree regressor do?\n",
    "\n",
    "Every run will generate different result but on most occassions, **the single decision tree regressor is likely to learn spurious features** i.e. will assign small importance to features which are not true regressors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 550,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 551,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(criterion='mse', max_depth=5, max_features=None,\n",
       "           max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
       "           min_impurity_split=None, min_samples_leaf=1,\n",
       "           min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "           presort=False, random_state=None, splitter='best')"
      ]
     },
     "execution_count": 551,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree_model = tree.DecisionTreeRegressor(max_depth=5,random_state=None)\n",
    "tree_model.fit(X,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 566,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Relative importance of the features:  [ 0.06896493  0.35741588  0.54578154  0.0230699   0.          0.00476775]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAo4AAAGyCAYAAABqXkWrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngSMRERER+fIXIGx4i6NHYAIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d13085f080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"Relative importance of the features: \",tree_model.feature_importances_)\n",
    "with plt.style.context('dark_background'):\n",
    "    plt.figure(figsize=(10,7))\n",
    "    plt.grid(True)\n",
    "    plt.yticks(range(n_features+1,1,-1),df.columns[:-1],fontsize=20)\n",
    "    plt.xlabel(\"Relative (normalized) importance of parameters\",fontsize=15)\n",
    "    plt.ylabel(\"Features\\n\",fontsize=20)\n",
    "    plt.barh(range(n_features+1,1,-1),width=tree_model.feature_importances_,height=0.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Print the $R^2$ score of the Decision Tree regression model\n",
    "\n",
    "Even though the $R^2$ score is pretty high, the model is slightly flawed because it has assigned importance to regressors which have no true significance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 553,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regression coefficient: 0.95695111153\n"
     ]
    }
   ],
   "source": [
    "print(\"Regression coefficient:\",tree_model.score(X,y))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Forest Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 554,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 555,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=5,\n",
       "           max_features='auto', max_leaf_nodes=5,\n",
       "           min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "           min_samples_leaf=1, min_samples_split=2,\n",
       "           min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=1,\n",
       "           oob_score=False, random_state=None, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 555,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = RandomForestRegressor(max_depth=5, random_state=None,max_features='auto',max_leaf_nodes=5,n_estimators=100)\n",
    "model.fit(X, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Print the relative importance of the features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 565,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Relative importance of the features:  [ 0.03456204  0.46959355  0.49500136  0.          0.          0.00084305]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAo4AAAGyCAYAAABqXkWrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d12f672780>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"Relative importance of the features: \",model.feature_importances_)\n",
    "with plt.style.context('dark_background'):\n",
    "    plt.figure(figsize=(10,7))\n",
    "    plt.grid(True)\n",
    "    plt.yticks(range(n_features+1,1,-1),df.columns[:-1],fontsize=20)\n",
    "    plt.xlabel(\"Relative (normalized) importance of parameters\",fontsize=15)\n",
    "    plt.ylabel(\"Features\\n\",fontsize=20)\n",
    "    plt.barh(range(n_features+1,1,-1),width=model.feature_importances_,height=0.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Print the $R^2$ score of the Random Forest regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 557,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regression coefficient: 0.811794737752\n"
     ]
    }
   ],
   "source": [
    "print(\"Regression coefficient:\",model.score(X,y))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Benchmark to statsmodel (ordinary least-square solution by exact analytic method)\n",
    "\n",
    "[Statsmodel is a Python module](http://www.statsmodels.org/dev/index.html) that provides classes and functions for the estimation of many different statistical models, as well as for conducting statistical tests, and statistical data exploration. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 558,
   "metadata": {},
   "outputs": [],
   "source": [
    "import statsmodels.api as sm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 559,
   "metadata": {},
   "outputs": [],
   "source": [
    "Xs=sm.add_constant(X)\n",
    "stat_model = sm.OLS(y,Xs)\n",
    "stat_result = stat_model.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 560,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                      y   R-squared:                       0.971\n",
      "Model:                            OLS   Adj. R-squared:                  0.969\n",
      "Method:                 Least Squares   F-statistic:                     521.4\n",
      "Date:                Thu, 04 Jan 2018   Prob (F-statistic):           2.79e-69\n",
      "Time:                        23:56:18   Log-Likelihood:                -434.92\n",
      "No. Observations:                 100   AIC:                             883.8\n",
      "Df Residuals:                      93   BIC:                             902.1\n",
      "Df Model:                           6                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const          0.9153      2.056      0.445      0.657      -3.167       4.998\n",
      "x1            38.2702      1.900     20.142      0.000      34.497      42.043\n",
      "x2            69.6705      1.888     36.908      0.000      65.922      73.419\n",
      "x3            75.1667      2.057     36.546      0.000      71.082      79.251\n",
      "x4            -0.3881      2.073     -0.187      0.852      -4.505       3.729\n",
      "x5            -1.3036      2.038     -0.640      0.524      -5.352       2.744\n",
      "x6            -0.2271      2.112     -0.108      0.915      -4.420       3.966\n",
      "==============================================================================\n",
      "Omnibus:                        0.366   Durbin-Watson:                   1.855\n",
      "Prob(Omnibus):                  0.833   Jarque-Bera (JB):                0.531\n",
      "Skew:                           0.044   Prob(JB):                        0.767\n",
      "Kurtosis:                       2.654   Cond. No.                         1.47\n",
      "==============================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "print(stat_result.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Make arrays of regression coefficients estimated by the models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 561,
   "metadata": {},
   "outputs": [],
   "source": [
    "rf_coef=np.array(coef)\n",
    "stat_coef=np.array(stat_result.params[1:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Show the true regression coefficients (as returned by the generator function) and the OLS model's estimated coefficients side by side"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 562,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>X5</th>\n",
       "      <th>X6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>True Regressors</th>\n",
       "      <td>38.474632</td>\n",
       "      <td>71.735567</td>\n",
       "      <td>72.595476</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OLS method estimation</th>\n",
       "      <td>38.270216</td>\n",
       "      <td>69.670541</td>\n",
       "      <td>75.166745</td>\n",
       "      <td>-0.388141</td>\n",
       "      <td>-1.303634</td>\n",
       "      <td>-0.227058</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              X1         X2         X3        X4        X5  \\\n",
       "True Regressors        38.474632  71.735567  72.595476  0.000000  0.000000   \n",
       "OLS method estimation  38.270216  69.670541  75.166745 -0.388141 -1.303634   \n",
       "\n",
       "                             X6  \n",
       "True Regressors        0.000000  \n",
       "OLS method estimation -0.227058  "
      ]
     },
     "execution_count": 562,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_coef = pd.DataFrame(data=[rf_coef,stat_coef],columns=df.columns[:-1],index=['True Regressors', 'OLS method estimation'])\n",
    "df_coef"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Show the relative importance of regressors side by side\n",
    "\n",
    "For Random Forest Model, show the relative importance of features as determined by the meta-estimator. For the OLS model, show normalized t-statistic values.\n",
    "\n",
    "**It will be clear that although the RandomForest regressor identifies the important regressors correctly, it does not assign the same level of relative importance to them as done by OLS method t-statistic**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 563,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>X5</th>\n",
       "      <th>X6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>RF Regressor relative importance</th>\n",
       "      <td>0.034562</td>\n",
       "      <td>0.469594</td>\n",
       "      <td>0.495001</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000843</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OLS method normalized t-statistic</th>\n",
       "      <td>0.217370</td>\n",
       "      <td>0.398305</td>\n",
       "      <td>0.394407</td>\n",
       "      <td>-0.00202</td>\n",
       "      <td>-0.006902</td>\n",
       "      <td>-0.001160</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         X1        X2        X3       X4  \\\n",
       "RF Regressor relative importance   0.034562  0.469594  0.495001  0.00000   \n",
       "OLS method normalized t-statistic  0.217370  0.398305  0.394407 -0.00202   \n",
       "\n",
       "                                         X5        X6  \n",
       "RF Regressor relative importance   0.000000  0.000843  \n",
       "OLS method normalized t-statistic -0.006902 -0.001160  "
      ]
     },
     "execution_count": 563,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_importance = pd.DataFrame(data=[model.feature_importances_,stat_result.tvalues[1:]/sum(stat_result.tvalues[1:])],\n",
    "                             columns=df.columns[:-1],\n",
    "                             index=['RF Regressor relative importance', 'OLS method normalized t-statistic'])\n",
    "df_importance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Further explorations\n",
    "\n",
    "Play with the hyperparameters of the models e.g. tree depth, leaf quantity, splitting strategy, number of trees in the forest, etc. to see how the regression model behaves. Also, play with the nature and complexity of the sample generator by writing your own nonlinear function."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}