diff --git a/consume.ipynb b/consume.ipynb
new file mode 100644
index 0000000..f244f6b
--- /dev/null
+++ b/consume.ipynb
@@ -0,0 +1,1428 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## COBIFY - Best Fuel"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- Let's import the libraries we'll use and the csv file with the data."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import seaborn as sns\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "import statsmodels.formula.api as smf"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "consume_data = pd.read_csv('/home/julian/Cursos/Ironhack/Career/reto_data/consume_data.csv')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Exploratory Data Analysis"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<class 'pandas.core.frame.DataFrame'>\n",
+      "RangeIndex: 376 entries, 0 to 375\n",
+      "Data columns (total 11 columns):\n",
+      " #   Column        Non-Null Count  Dtype  \n",
+      "---  ------        --------------  -----  \n",
+      " 0   consume_id    376 non-null    int64  \n",
+      " 1   distance      376 non-null    float64\n",
+      " 2   consume       376 non-null    float64\n",
+      " 3   speed         376 non-null    int64  \n",
+      " 4   temp_inside   376 non-null    float64\n",
+      " 5   temp_outside  376 non-null    int64  \n",
+      " 6   gas_type      376 non-null    object \n",
+      " 7   AC            376 non-null    int64  \n",
+      " 8   rain          376 non-null    int64  \n",
+      " 9   sun           376 non-null    int64  \n",
+      " 10  snow          376 non-null    int64  \n",
+      "dtypes: float64(3), int64(7), object(1)\n",
+      "memory usage: 32.4+ KB\n"
+     ]
+    }
+   ],
+   "source": [
+    "consume_data.info()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "consume_id  distance  consume  speed  temp_inside  temp_outside  gas_type  AC     rain   sun    snow \n",
+       "False       False     False    False  False        False         False     False  False  False  False    376\n",
+       "dtype: int64"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "consume_data.isna().value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "SP98    219\n",
+       "E10     157\n",
+       "Name: gas_type, dtype: int64"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "consume_data.gas_type.value_counts()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- We can see that all variabels are numerical, except for gas_type, which we'll transform into a dummy variable. On the other hand, we don't have null values.\n",
+    "- We'll also drop consume_id column as is an index column.\n",
+    "- 58% of cars use SP98 gas."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "consume_data = pd.get_dummies(consume_data, columns=['gas_type'], drop_first=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "consume_data.drop(columns=['consume_id'], inplace=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "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>distance</th>\n",
+       "      <th>consume</th>\n",
+       "      <th>speed</th>\n",
+       "      <th>temp_inside</th>\n",
+       "      <th>temp_outside</th>\n",
+       "      <th>AC</th>\n",
+       "      <th>rain</th>\n",
+       "      <th>sun</th>\n",
+       "      <th>snow</th>\n",
+       "      <th>gas_type_SP98</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>28.0</td>\n",
+       "      <td>5.0</td>\n",
+       "      <td>26</td>\n",
+       "      <td>21.5</td>\n",
+       "      <td>12</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>12.0</td>\n",
+       "      <td>4.2</td>\n",
+       "      <td>30</td>\n",
+       "      <td>21.5</td>\n",
+       "      <td>13</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>11.2</td>\n",
+       "      <td>5.5</td>\n",
+       "      <td>38</td>\n",
+       "      <td>21.5</td>\n",
+       "      <td>15</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   distance  consume  speed  temp_inside  temp_outside  AC  rain  sun  snow  \\\n",
+       "0      28.0      5.0     26         21.5            12   0     0    0     0   \n",
+       "1      12.0      4.2     30         21.5            13   0     0    0     0   \n",
+       "2      11.2      5.5     38         21.5            15   0     0    0     0   \n",
+       "\n",
+       "   gas_type_SP98  \n",
+       "0              0  \n",
+       "1              0  \n",
+       "2              0  "
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "consume_data.head(3)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Minimum</th>\n",
+       "      <th>Maximum</th>\n",
+       "      <th>Mean</th>\n",
+       "      <th>Median</th>\n",
+       "      <th>Mode</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>Consume</th>\n",
+       "      <td>3.3</td>\n",
+       "      <td>12.2</td>\n",
+       "      <td>4.925266</td>\n",
+       "      <td>4.7</td>\n",
+       "      <td>4.5</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         Minimum  Maximum      Mean  Median  Mode\n",
+       "Consume      3.3     12.2  4.925266     4.7   4.5"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "consume = {'Minimum': consume_data.consume.min(),\n",
+    "          'Maximum': consume_data.consume.max(),\n",
+    "          'Mean': consume_data.consume.mean(),\n",
+    "          'Median': consume_data.consume.median(),\n",
+    "          'Mode': consume_data.consume.mode()}\n",
+    "consume_descr = pd.DataFrame(consume) \n",
+    "consume_descr = consume_descr.rename(index={0:'Consume'})\n",
+    "consume_descr"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<AxesSubplot:xlabel='consume', ylabel='Count'>"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "sns.histplot(x='consume', data=consume_data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<AxesSubplot:xlabel='consume'>"
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "sns.boxplot(x=consume_data.consume)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- From the histogram and boxplot, we can observe that consume variable has some outliers, which we'll have to take into account when doing the regression analysis"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "corr_matrix = consume_data.corr()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "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>distance</th>\n",
+       "      <th>consume</th>\n",
+       "      <th>speed</th>\n",
+       "      <th>temp_inside</th>\n",
+       "      <th>temp_outside</th>\n",
+       "      <th>AC</th>\n",
+       "      <th>rain</th>\n",
+       "      <th>sun</th>\n",
+       "      <th>snow</th>\n",
+       "      <th>gas_type_SP98</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>distance</th>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>-0.131379</td>\n",
+       "      <td>0.565505</td>\n",
+       "      <td>0.075305</td>\n",
+       "      <td>0.086774</td>\n",
+       "      <td>0.019030</td>\n",
+       "      <td>-0.030784</td>\n",
+       "      <td>0.078627</td>\n",
+       "      <td>-0.034428</td>\n",
+       "      <td>-0.047877</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>consume</th>\n",
+       "      <td>-0.131379</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>-0.234537</td>\n",
+       "      <td>-0.161991</td>\n",
+       "      <td>-0.318758</td>\n",
+       "      <td>0.089472</td>\n",
+       "      <td>0.251756</td>\n",
+       "      <td>-0.169173</td>\n",
+       "      <td>0.072308</td>\n",
+       "      <td>-0.020928</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>speed</th>\n",
+       "      <td>0.565505</td>\n",
+       "      <td>-0.234537</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.059725</td>\n",
+       "      <td>0.019327</td>\n",
+       "      <td>-0.024485</td>\n",
+       "      <td>-0.004173</td>\n",
+       "      <td>0.077181</td>\n",
+       "      <td>0.032500</td>\n",
+       "      <td>-0.099495</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>temp_inside</th>\n",
+       "      <td>0.075305</td>\n",
+       "      <td>-0.161991</td>\n",
+       "      <td>0.059725</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.361308</td>\n",
+       "      <td>0.313197</td>\n",
+       "      <td>-0.038269</td>\n",
+       "      <td>0.246120</td>\n",
+       "      <td>0.097156</td>\n",
+       "      <td>0.010340</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>temp_outside</th>\n",
+       "      <td>0.086774</td>\n",
+       "      <td>-0.318758</td>\n",
+       "      <td>0.019327</td>\n",
+       "      <td>0.361308</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.180096</td>\n",
+       "      <td>-0.184763</td>\n",
+       "      <td>0.349988</td>\n",
+       "      <td>-0.162948</td>\n",
+       "      <td>0.156990</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>AC</th>\n",
+       "      <td>0.019030</td>\n",
+       "      <td>0.089472</td>\n",
+       "      <td>-0.024485</td>\n",
+       "      <td>0.313197</td>\n",
+       "      <td>0.180096</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.242197</td>\n",
+       "      <td>0.085928</td>\n",
+       "      <td>0.063170</td>\n",
+       "      <td>0.096943</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>rain</th>\n",
+       "      <td>-0.030784</td>\n",
+       "      <td>0.251756</td>\n",
+       "      <td>-0.004173</td>\n",
+       "      <td>-0.038269</td>\n",
+       "      <td>-0.184763</td>\n",
+       "      <td>0.242197</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>-0.111916</td>\n",
+       "      <td>0.277739</td>\n",
+       "      <td>0.052789</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>sun</th>\n",
+       "      <td>0.078627</td>\n",
+       "      <td>-0.169173</td>\n",
+       "      <td>0.077181</td>\n",
+       "      <td>0.246120</td>\n",
+       "      <td>0.349988</td>\n",
+       "      <td>0.085928</td>\n",
+       "      <td>-0.111916</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>-0.031083</td>\n",
+       "      <td>0.038120</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>snow</th>\n",
+       "      <td>-0.034428</td>\n",
+       "      <td>0.072308</td>\n",
+       "      <td>0.032500</td>\n",
+       "      <td>0.097156</td>\n",
+       "      <td>-0.162948</td>\n",
+       "      <td>0.063170</td>\n",
+       "      <td>0.277739</td>\n",
+       "      <td>-0.031083</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.035231</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>gas_type_SP98</th>\n",
+       "      <td>-0.047877</td>\n",
+       "      <td>-0.020928</td>\n",
+       "      <td>-0.099495</td>\n",
+       "      <td>0.010340</td>\n",
+       "      <td>0.156990</td>\n",
+       "      <td>0.096943</td>\n",
+       "      <td>0.052789</td>\n",
+       "      <td>0.038120</td>\n",
+       "      <td>0.035231</td>\n",
+       "      <td>1.000000</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "               distance   consume     speed  temp_inside  temp_outside  \\\n",
+       "distance       1.000000 -0.131379  0.565505     0.075305      0.086774   \n",
+       "consume       -0.131379  1.000000 -0.234537    -0.161991     -0.318758   \n",
+       "speed          0.565505 -0.234537  1.000000     0.059725      0.019327   \n",
+       "temp_inside    0.075305 -0.161991  0.059725     1.000000      0.361308   \n",
+       "temp_outside   0.086774 -0.318758  0.019327     0.361308      1.000000   \n",
+       "AC             0.019030  0.089472 -0.024485     0.313197      0.180096   \n",
+       "rain          -0.030784  0.251756 -0.004173    -0.038269     -0.184763   \n",
+       "sun            0.078627 -0.169173  0.077181     0.246120      0.349988   \n",
+       "snow          -0.034428  0.072308  0.032500     0.097156     -0.162948   \n",
+       "gas_type_SP98 -0.047877 -0.020928 -0.099495     0.010340      0.156990   \n",
+       "\n",
+       "                     AC      rain       sun      snow  gas_type_SP98  \n",
+       "distance       0.019030 -0.030784  0.078627 -0.034428      -0.047877  \n",
+       "consume        0.089472  0.251756 -0.169173  0.072308      -0.020928  \n",
+       "speed         -0.024485 -0.004173  0.077181  0.032500      -0.099495  \n",
+       "temp_inside    0.313197 -0.038269  0.246120  0.097156       0.010340  \n",
+       "temp_outside   0.180096 -0.184763  0.349988 -0.162948       0.156990  \n",
+       "AC             1.000000  0.242197  0.085928  0.063170       0.096943  \n",
+       "rain           0.242197  1.000000 -0.111916  0.277739       0.052789  \n",
+       "sun            0.085928 -0.111916  1.000000 -0.031083       0.038120  \n",
+       "snow           0.063170  0.277739 -0.031083  1.000000       0.035231  \n",
+       "gas_type_SP98  0.096943  0.052789  0.038120  0.035231       1.000000  "
+      ]
+     },
+     "execution_count": 14,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "corr_matrix"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "mascara = np.triu(np.ones_like(corr_matrix, dtype=bool))\n",
+    "color_map = sns.diverging_palette(0, 10, as_cmap=True) \n",
+    "\n",
+    "sns.heatmap(corr_matrix, \n",
+    "            mask=mascara,\n",
+    "            cmap=\"viridis\",\n",
+    "            vmax=1,\n",
+    "            vmin=-1,\n",
+    "            square=True, \n",
+    "            linewidth=.5, \n",
+    "            cbar_kws={\"shrink\": .5} \n",
+    ");"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- We can see through the correlations heatmap that `consume` is mostly correlated with `temp_outside` (negatively), `rain` (positively), `speed` (negatively), `sun` (negatively), and `temp_inside` (negatively).\n",
+    "- We can also observe that `speed` and `distance` are correlated, and `temp_inside` wit `AC`. Both are not surprise.\n",
+    "- Next we'll plot some of these variables against the consume."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "sns.scatterplot(x=consume_data.temp_outside, y=consume_data.consume, hue=consume_data.gas_type_SP98);"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- We can see how lower outside temperatures accounts for higher consumption levels. "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x360 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fig, axs = plt.subplots(nrows=1, ncols=2, figsize=(15, 5))\n",
+    "sns.scatterplot(data=consume_data, x=\"temp_inside\", y=\"consume\", hue=\"gas_type_SP98\", ax=axs[0])\n",
+    "sns.scatterplot(data=consume_data, x=\"temp_inside\", y=\"consume\", hue=\"AC\", ax=axs[1]);"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- We can see that for lower inside temperature values, there are higher consumption values. On the other hand, we can see that most of the cars had the AC off, but those with the AC on, accounts for higher consumption values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>consume</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>gas_type_SP98</th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>4.950955</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>4.906849</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                consume\n",
+       "gas_type_SP98          \n",
+       "0              4.950955\n",
+       "1              4.906849"
+      ]
+     },
+     "execution_count": 36,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "gtype_gr = consume_data.groupby('gas_type_SP98').agg({'consume': 'mean'})\n",
+    "gtype_gr"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>consume</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>AC</th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>4.897391</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>5.235484</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "     consume\n",
+       "AC          \n",
+       "0   4.897391\n",
+       "1   5.235484"
+      ]
+     },
+     "execution_count": 19,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "ac_group = consume_data.groupby('AC').agg({'consume': 'mean'})\n",
+    "ac_group"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "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>consume</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>rain</th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>4.827576</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>5.626087</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       consume\n",
+       "rain          \n",
+       "0     4.827576\n",
+       "1     5.626087"
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "rain_gr = consume_data.groupby('rain').agg({'consume': 'mean'})\n",
+    "rain_gr"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- From the above data frames we can see that the consumption is higher in average when: the AC is on, is raining\n",
+    "- But also the consumption is higher in average when the car is using the ethanol type of gas. Although is higher in a small proportion, it's considerable given that most of the cars in the data set use non-ethanol gas. "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### REGRESSION ANALYSIS"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- We'll model the `consume` variable as a linear regression over the other variables to check which of them has statistical significance in order to explain the consume of the car."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "linear_model = smf.ols(\"consume ~ distance + speed + temp_inside + temp_outside + AC + rain + sun + snow + gas_type_SP98\", data=consume_data).fit()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>         <td>consume</td>     <th>  R-squared:         </th> <td>   0.208</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.188</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   10.68</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 17 Oct 2021</td> <th>  Prob (F-statistic):</th> <td>9.76e-15</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>20:45:17</td>     <th>  Log-Likelihood:    </th> <td> -504.20</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>   376</td>      <th>  AIC:               </th> <td>   1028.</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>   366</td>      <th>  BIC:               </th> <td>   1068.</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>     9</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>Intercept</th>     <td>    7.7409</td> <td>    1.186</td> <td>    6.526</td> <td> 0.000</td> <td>    5.408</td> <td>   10.073</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>distance</th>      <td>    0.0018</td> <td>    0.003</td> <td>    0.683</td> <td> 0.495</td> <td>   -0.003</td> <td>    0.007</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>speed</th>         <td>   -0.0184</td> <td>    0.004</td> <td>   -4.259</td> <td> 0.000</td> <td>   -0.027</td> <td>   -0.010</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>temp_inside</th>   <td>   -0.0776</td> <td>    0.055</td> <td>   -1.414</td> <td> 0.158</td> <td>   -0.185</td> <td>    0.030</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>temp_outside</th>  <td>   -0.0394</td> <td>    0.008</td> <td>   -4.873</td> <td> 0.000</td> <td>   -0.055</td> <td>   -0.023</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>AC</th>            <td>    0.4471</td> <td>    0.195</td> <td>    2.298</td> <td> 0.022</td> <td>    0.065</td> <td>    0.830</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>rain</th>          <td>    0.5432</td> <td>    0.162</td> <td>    3.344</td> <td> 0.001</td> <td>    0.224</td> <td>    0.863</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>sun</th>           <td>   -0.1210</td> <td>    0.190</td> <td>   -0.636</td> <td> 0.525</td> <td>   -0.495</td> <td>    0.253</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>snow</th>          <td>   -0.1020</td> <td>    0.503</td> <td>   -0.203</td> <td> 0.839</td> <td>   -1.091</td> <td>    0.887</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>gas_type_SP98</th> <td>   -0.0408</td> <td>    0.101</td> <td>   -0.406</td> <td> 0.685</td> <td>   -0.239</td> <td>    0.157</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>248.482</td> <th>  Durbin-Watson:     </th> <td>   2.096</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td>2919.818</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td> 2.651</td>  <th>  Prob(JB):          </th> <td>    0.00</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td>15.580</td>  <th>  Cond. No.          </th> <td>1.37e+03</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, 1.37e+03. 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:                consume   R-squared:                       0.208\n",
+       "Model:                            OLS   Adj. R-squared:                  0.188\n",
+       "Method:                 Least Squares   F-statistic:                     10.68\n",
+       "Date:                Sun, 17 Oct 2021   Prob (F-statistic):           9.76e-15\n",
+       "Time:                        20:45:17   Log-Likelihood:                -504.20\n",
+       "No. Observations:                 376   AIC:                             1028.\n",
+       "Df Residuals:                     366   BIC:                             1068.\n",
+       "Df Model:                           9                                         \n",
+       "Covariance Type:            nonrobust                                         \n",
+       "=================================================================================\n",
+       "                    coef    std err          t      P>|t|      [0.025      0.975]\n",
+       "---------------------------------------------------------------------------------\n",
+       "Intercept         7.7409      1.186      6.526      0.000       5.408      10.073\n",
+       "distance          0.0018      0.003      0.683      0.495      -0.003       0.007\n",
+       "speed            -0.0184      0.004     -4.259      0.000      -0.027      -0.010\n",
+       "temp_inside      -0.0776      0.055     -1.414      0.158      -0.185       0.030\n",
+       "temp_outside     -0.0394      0.008     -4.873      0.000      -0.055      -0.023\n",
+       "AC                0.4471      0.195      2.298      0.022       0.065       0.830\n",
+       "rain              0.5432      0.162      3.344      0.001       0.224       0.863\n",
+       "sun              -0.1210      0.190     -0.636      0.525      -0.495       0.253\n",
+       "snow             -0.1020      0.503     -0.203      0.839      -1.091       0.887\n",
+       "gas_type_SP98    -0.0408      0.101     -0.406      0.685      -0.239       0.157\n",
+       "==============================================================================\n",
+       "Omnibus:                      248.482   Durbin-Watson:                   2.096\n",
+       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):             2919.818\n",
+       "Skew:                           2.651   Prob(JB):                         0.00\n",
+       "Kurtosis:                      15.580   Cond. No.                     1.37e+03\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, 1.37e+03. This might indicate that there are\n",
+       "strong multicollinearity or other numerical problems.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 23,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "linear_model.summary()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- From the summary we can see that a linear model is not very good to model the consumption, as we got a R2 of 0.208. \n",
+    "- We can see from the p-values that `distance` and `temp_inside` are not significant, which is no suprise given the correlation they have with `speed` and `AC` respectively.\n",
+    "- We can also see that `sun`, `snow` and `gas_type` are non-significant.\n",
+    "- We'll perform another regression but this time without `distance`. We'll log-transform the target variable to see if we can reduce the potential influence of outliers and standarize the data set."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "consume_data['log_consume'] = np.log(consume_data.consume)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "sns.histplot(consume_data.log_consume);"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "scaler = StandardScaler()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [],
+   "source": [
+    "consume_data_scaled=pd.DataFrame(scaler.fit_transform(consume_data), columns=consume_data.columns)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "linear_model_2 = smf.ols(\"log_consume ~ speed  + temp_outside + AC + rain + sun + snow + gas_type_SP98\", data=consume_data).fit()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>       <td>log_consume</td>   <th>  R-squared:         </th> <td>   0.229</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.214</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   15.62</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 17 Oct 2021</td> <th>  Prob (F-statistic):</th> <td>6.11e-18</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>20:45:18</td>     <th>  Log-Likelihood:    </th> <td>  163.29</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>   376</td>      <th>  AIC:               </th> <td>  -310.6</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>   368</td>      <th>  BIC:               </th> <td>  -279.1</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>     7</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>Intercept</th>     <td>    1.7727</td> <td>    0.032</td> <td>   56.115</td> <td> 0.000</td> <td>    1.711</td> <td>    1.835</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>speed</th>         <td>   -0.0026</td> <td>    0.001</td> <td>   -4.271</td> <td> 0.000</td> <td>   -0.004</td> <td>   -0.001</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>temp_outside</th>  <td>   -0.0085</td> <td>    0.001</td> <td>   -6.511</td> <td> 0.000</td> <td>   -0.011</td> <td>   -0.006</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>AC</th>            <td>    0.0664</td> <td>    0.032</td> <td>    2.101</td> <td> 0.036</td> <td>    0.004</td> <td>    0.128</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>rain</th>          <td>    0.0958</td> <td>    0.027</td> <td>    3.500</td> <td> 0.001</td> <td>    0.042</td> <td>    0.150</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>sun</th>           <td>   -0.0320</td> <td>    0.032</td> <td>   -1.005</td> <td> 0.316</td> <td>   -0.095</td> <td>    0.031</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>snow</th>          <td>   -0.0436</td> <td>    0.084</td> <td>   -0.521</td> <td> 0.602</td> <td>   -0.208</td> <td>    0.121</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>gas_type_SP98</th> <td>   -0.0095</td> <td>    0.017</td> <td>   -0.563</td> <td> 0.574</td> <td>   -0.043</td> <td>    0.024</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>111.195</td> <th>  Durbin-Watson:     </th> <td>   2.112</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td> 366.740</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td> 1.317</td>  <th>  Prob(JB):          </th> <td>2.31e-80</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 7.059</td>  <th>  Cond. No.          </th> <td>    468.</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:            log_consume   R-squared:                       0.229\n",
+       "Model:                            OLS   Adj. R-squared:                  0.214\n",
+       "Method:                 Least Squares   F-statistic:                     15.62\n",
+       "Date:                Sun, 17 Oct 2021   Prob (F-statistic):           6.11e-18\n",
+       "Time:                        20:45:18   Log-Likelihood:                 163.29\n",
+       "No. Observations:                 376   AIC:                            -310.6\n",
+       "Df Residuals:                     368   BIC:                            -279.1\n",
+       "Df Model:                           7                                         \n",
+       "Covariance Type:            nonrobust                                         \n",
+       "=================================================================================\n",
+       "                    coef    std err          t      P>|t|      [0.025      0.975]\n",
+       "---------------------------------------------------------------------------------\n",
+       "Intercept         1.7727      0.032     56.115      0.000       1.711       1.835\n",
+       "speed            -0.0026      0.001     -4.271      0.000      -0.004      -0.001\n",
+       "temp_outside     -0.0085      0.001     -6.511      0.000      -0.011      -0.006\n",
+       "AC                0.0664      0.032      2.101      0.036       0.004       0.128\n",
+       "rain              0.0958      0.027      3.500      0.001       0.042       0.150\n",
+       "sun              -0.0320      0.032     -1.005      0.316      -0.095       0.031\n",
+       "snow             -0.0436      0.084     -0.521      0.602      -0.208       0.121\n",
+       "gas_type_SP98    -0.0095      0.017     -0.563      0.574      -0.043       0.024\n",
+       "==============================================================================\n",
+       "Omnibus:                      111.195   Durbin-Watson:                   2.112\n",
+       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              366.740\n",
+       "Skew:                           1.317   Prob(JB):                     2.31e-80\n",
+       "Kurtosis:                       7.059   Cond. No.                         468.\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 29,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "linear_model_2.summary()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- This time we can see that we got a R2 of 0.229, a slightly improvement,   \n",
+    "- We can see from the p-values that `sun`, `snow` and `gas_type` are still non-significant."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### From the regression analisis, we can say that the variables whic are significant to predict the mean of consumption are `rain`, `AC`, `temp_outside`  and `speed`. Which goes in line with the correlation analysis."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### CONSUME COST"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- We'll enrich the data with the price of the gas"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def price(x):\n",
+    "    E10_price = 1.379\n",
+    "    SP98_price = 1.459\n",
+    "    if x == 0:\n",
+    "        return E10_price\n",
+    "    else:\n",
+    "        return SP98_price"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "consume_data['price'] = consume_data.gas_type_SP98.apply(price)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "consume_data['cost_per_distance'] =(((consume_data.consume)/100) * (consume_data.distance)) * consume_data.price "
+   ]
+  },
+  {
+   "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>cost_per_distance</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>gas_type_SP98</th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1.396125</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>1.305041</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "               cost_per_distance\n",
+       "gas_type_SP98                   \n",
+       "0                       1.396125\n",
+       "1                       1.305041"
+      ]
+     },
+     "execution_count": 40,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "price_gr = consume_data.groupby('gas_type_SP98').agg({'cost_per_distance': 'mean'})\n",
+    "price_gr"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- From the above data frame we can see that the cars with ethanol gas-type, have a higher cost consumption, even if the price of ethanol is lower."
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "ironcon",
+   "language": "python",
+   "name": "ironcon"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.11"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}