Updated_Data.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "bef68ea0",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "9e142c72",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('/Users/marcostellez/code/abefarkas/Thalassa_Regime_Classifier/raw_data/data_set.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "dd0d3200",
   "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>Unnamed: 0</th>\n",
       "      <th>primary_key</th>\n",
       "      <th>bp1</th>\n",
       "      <th>bs1</th>\n",
       "      <th>bp2</th>\n",
       "      <th>bs2</th>\n",
       "      <th>bp3</th>\n",
       "      <th>bs3</th>\n",
       "      <th>bp4</th>\n",
       "      <th>bs4</th>\n",
       "      <th>...</th>\n",
       "      <th>ap16</th>\n",
       "      <th>as16</th>\n",
       "      <th>ap17</th>\n",
       "      <th>as17</th>\n",
       "      <th>ap18</th>\n",
       "      <th>as18</th>\n",
       "      <th>ap19</th>\n",
       "      <th>as19</th>\n",
       "      <th>ap20</th>\n",
       "      <th>as20</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2022-05-19 00:00:15</td>\n",
       "      <td>28720.075362</td>\n",
       "      <td>3.772080</td>\n",
       "      <td>28719.533333</td>\n",
       "      <td>1.405377</td>\n",
       "      <td>28719.011594</td>\n",
       "      <td>0.626551</td>\n",
       "      <td>28718.371014</td>\n",
       "      <td>0.505087</td>\n",
       "      <td>...</td>\n",
       "      <td>28724.006522</td>\n",
       "      <td>0.418891</td>\n",
       "      <td>28724.167391</td>\n",
       "      <td>0.487558</td>\n",
       "      <td>28724.361594</td>\n",
       "      <td>0.706101</td>\n",
       "      <td>28724.511594</td>\n",
       "      <td>0.707326</td>\n",
       "      <td>28724.657971</td>\n",
       "      <td>1.104145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2022-05-19 00:00:45</td>\n",
       "      <td>28755.960345</td>\n",
       "      <td>1.878828</td>\n",
       "      <td>28755.570690</td>\n",
       "      <td>0.335776</td>\n",
       "      <td>28755.050000</td>\n",
       "      <td>0.573517</td>\n",
       "      <td>28754.684483</td>\n",
       "      <td>0.737259</td>\n",
       "      <td>...</td>\n",
       "      <td>28760.520690</td>\n",
       "      <td>7.383086</td>\n",
       "      <td>28760.674138</td>\n",
       "      <td>4.193586</td>\n",
       "      <td>28760.872414</td>\n",
       "      <td>0.620983</td>\n",
       "      <td>28761.067241</td>\n",
       "      <td>0.773897</td>\n",
       "      <td>28761.250000</td>\n",
       "      <td>0.854017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2022-05-19 00:01:15</td>\n",
       "      <td>28782.703333</td>\n",
       "      <td>1.771600</td>\n",
       "      <td>28782.131667</td>\n",
       "      <td>0.536867</td>\n",
       "      <td>28781.485000</td>\n",
       "      <td>0.402600</td>\n",
       "      <td>28781.140000</td>\n",
       "      <td>0.382467</td>\n",
       "      <td>...</td>\n",
       "      <td>28787.931667</td>\n",
       "      <td>0.427633</td>\n",
       "      <td>28788.138333</td>\n",
       "      <td>0.580383</td>\n",
       "      <td>28788.331667</td>\n",
       "      <td>0.421400</td>\n",
       "      <td>28788.506667</td>\n",
       "      <td>0.376617</td>\n",
       "      <td>28788.750000</td>\n",
       "      <td>0.425300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>2022-05-19 00:01:45</td>\n",
       "      <td>28789.470000</td>\n",
       "      <td>1.767750</td>\n",
       "      <td>28788.766667</td>\n",
       "      <td>0.343217</td>\n",
       "      <td>28788.278333</td>\n",
       "      <td>0.243650</td>\n",
       "      <td>28788.046667</td>\n",
       "      <td>0.325300</td>\n",
       "      <td>...</td>\n",
       "      <td>28793.874833</td>\n",
       "      <td>0.486350</td>\n",
       "      <td>28794.093333</td>\n",
       "      <td>0.346800</td>\n",
       "      <td>28794.290000</td>\n",
       "      <td>1.161250</td>\n",
       "      <td>28794.483000</td>\n",
       "      <td>0.536067</td>\n",
       "      <td>28794.623333</td>\n",
       "      <td>0.395067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>2022-05-19 00:02:15</td>\n",
       "      <td>28788.382759</td>\n",
       "      <td>1.634448</td>\n",
       "      <td>28787.839655</td>\n",
       "      <td>0.342845</td>\n",
       "      <td>28787.370690</td>\n",
       "      <td>0.396414</td>\n",
       "      <td>28787.039655</td>\n",
       "      <td>0.232483</td>\n",
       "      <td>...</td>\n",
       "      <td>28792.984483</td>\n",
       "      <td>0.383293</td>\n",
       "      <td>28793.143103</td>\n",
       "      <td>0.595362</td>\n",
       "      <td>28793.387931</td>\n",
       "      <td>0.417172</td>\n",
       "      <td>28793.587931</td>\n",
       "      <td>0.473741</td>\n",
       "      <td>28793.763793</td>\n",
       "      <td>0.417776</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 82 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0          primary_key           bp1       bs1           bp2  \\\n",
       "0           0  2022-05-19 00:00:15  28720.075362  3.772080  28719.533333   \n",
       "1           1  2022-05-19 00:00:45  28755.960345  1.878828  28755.570690   \n",
       "2           2  2022-05-19 00:01:15  28782.703333  1.771600  28782.131667   \n",
       "3           3  2022-05-19 00:01:45  28789.470000  1.767750  28788.766667   \n",
       "4           4  2022-05-19 00:02:15  28788.382759  1.634448  28787.839655   \n",
       "\n",
       "        bs2           bp3       bs3           bp4       bs4  ...  \\\n",
       "0  1.405377  28719.011594  0.626551  28718.371014  0.505087  ...   \n",
       "1  0.335776  28755.050000  0.573517  28754.684483  0.737259  ...   \n",
       "2  0.536867  28781.485000  0.402600  28781.140000  0.382467  ...   \n",
       "3  0.343217  28788.278333  0.243650  28788.046667  0.325300  ...   \n",
       "4  0.342845  28787.370690  0.396414  28787.039655  0.232483  ...   \n",
       "\n",
       "           ap16      as16          ap17      as17          ap18      as18  \\\n",
       "0  28724.006522  0.418891  28724.167391  0.487558  28724.361594  0.706101   \n",
       "1  28760.520690  7.383086  28760.674138  4.193586  28760.872414  0.620983   \n",
       "2  28787.931667  0.427633  28788.138333  0.580383  28788.331667  0.421400   \n",
       "3  28793.874833  0.486350  28794.093333  0.346800  28794.290000  1.161250   \n",
       "4  28792.984483  0.383293  28793.143103  0.595362  28793.387931  0.417172   \n",
       "\n",
       "           ap19      as19          ap20      as20  \n",
       "0  28724.511594  0.707326  28724.657971  1.104145  \n",
       "1  28761.067241  0.773897  28761.250000  0.854017  \n",
       "2  28788.506667  0.376617  28788.750000  0.425300  \n",
       "3  28794.483000  0.536067  28794.623333  0.395067  \n",
       "4  28793.587931  0.473741  28793.763793  0.417776  \n",
       "\n",
       "[5 rows x 82 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4681bbdf",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Weighted Average Price\n",
    "data['WAP'] = (data['bp1']*data['bs1']\n",
    "               +data['bp2']*data['bs2']\n",
    "               +data['ap1']*data['as1']\n",
    "               +data['ap2']*data['as2'])/(data['bs1']+\n",
    "                                         data['bs2']+\n",
    "                                         data['as1']+\n",
    "                                         data['as2'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "5904fcb6",
   "metadata": {},
   "outputs": [],
   "source": [
    "data['spread'] = ((data['ap1']/data['bp1']) - 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "11373b1f",
   "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>Unnamed: 0</th>\n",
       "      <th>primary_key</th>\n",
       "      <th>bp1</th>\n",
       "      <th>bs1</th>\n",
       "      <th>bp2</th>\n",
       "      <th>bs2</th>\n",
       "      <th>bp3</th>\n",
       "      <th>bs3</th>\n",
       "      <th>bp4</th>\n",
       "      <th>bs4</th>\n",
       "      <th>...</th>\n",
       "      <th>ap17</th>\n",
       "      <th>as17</th>\n",
       "      <th>ap18</th>\n",
       "      <th>as18</th>\n",
       "      <th>ap19</th>\n",
       "      <th>as19</th>\n",
       "      <th>ap20</th>\n",
       "      <th>as20</th>\n",
       "      <th>WAP</th>\n",
       "      <th>spread</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2022-05-19 00:00:15</td>\n",
       "      <td>28720.075362</td>\n",
       "      <td>3.772080</td>\n",
       "      <td>28719.533333</td>\n",
       "      <td>1.405377</td>\n",
       "      <td>28719.011594</td>\n",
       "      <td>0.626551</td>\n",
       "      <td>28718.371014</td>\n",
       "      <td>0.505087</td>\n",
       "      <td>...</td>\n",
       "      <td>28724.167391</td>\n",
       "      <td>0.487558</td>\n",
       "      <td>28724.361594</td>\n",
       "      <td>0.706101</td>\n",
       "      <td>28724.511594</td>\n",
       "      <td>0.707326</td>\n",
       "      <td>28724.657971</td>\n",
       "      <td>1.104145</td>\n",
       "      <td>28720.131688</td>\n",
       "      <td>0.000015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2022-05-19 00:00:45</td>\n",
       "      <td>28755.960345</td>\n",
       "      <td>1.878828</td>\n",
       "      <td>28755.570690</td>\n",
       "      <td>0.335776</td>\n",
       "      <td>28755.050000</td>\n",
       "      <td>0.573517</td>\n",
       "      <td>28754.684483</td>\n",
       "      <td>0.737259</td>\n",
       "      <td>...</td>\n",
       "      <td>28760.674138</td>\n",
       "      <td>4.193586</td>\n",
       "      <td>28760.872414</td>\n",
       "      <td>0.620983</td>\n",
       "      <td>28761.067241</td>\n",
       "      <td>0.773897</td>\n",
       "      <td>28761.250000</td>\n",
       "      <td>0.854017</td>\n",
       "      <td>28756.179152</td>\n",
       "      <td>0.000010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2022-05-19 00:01:15</td>\n",
       "      <td>28782.703333</td>\n",
       "      <td>1.771600</td>\n",
       "      <td>28782.131667</td>\n",
       "      <td>0.536867</td>\n",
       "      <td>28781.485000</td>\n",
       "      <td>0.402600</td>\n",
       "      <td>28781.140000</td>\n",
       "      <td>0.382467</td>\n",
       "      <td>...</td>\n",
       "      <td>28788.138333</td>\n",
       "      <td>0.580383</td>\n",
       "      <td>28788.331667</td>\n",
       "      <td>0.421400</td>\n",
       "      <td>28788.506667</td>\n",
       "      <td>0.376617</td>\n",
       "      <td>28788.750000</td>\n",
       "      <td>0.425300</td>\n",
       "      <td>28782.777221</td>\n",
       "      <td>0.000005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>2022-05-19 00:01:45</td>\n",
       "      <td>28789.470000</td>\n",
       "      <td>1.767750</td>\n",
       "      <td>28788.766667</td>\n",
       "      <td>0.343217</td>\n",
       "      <td>28788.278333</td>\n",
       "      <td>0.243650</td>\n",
       "      <td>28788.046667</td>\n",
       "      <td>0.325300</td>\n",
       "      <td>...</td>\n",
       "      <td>28794.093333</td>\n",
       "      <td>0.346800</td>\n",
       "      <td>28794.290000</td>\n",
       "      <td>1.161250</td>\n",
       "      <td>28794.483000</td>\n",
       "      <td>0.536067</td>\n",
       "      <td>28794.623333</td>\n",
       "      <td>0.395067</td>\n",
       "      <td>28789.546017</td>\n",
       "      <td>0.000005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>2022-05-19 00:02:15</td>\n",
       "      <td>28788.382759</td>\n",
       "      <td>1.634448</td>\n",
       "      <td>28787.839655</td>\n",
       "      <td>0.342845</td>\n",
       "      <td>28787.370690</td>\n",
       "      <td>0.396414</td>\n",
       "      <td>28787.039655</td>\n",
       "      <td>0.232483</td>\n",
       "      <td>...</td>\n",
       "      <td>28793.143103</td>\n",
       "      <td>0.595362</td>\n",
       "      <td>28793.387931</td>\n",
       "      <td>0.417172</td>\n",
       "      <td>28793.587931</td>\n",
       "      <td>0.473741</td>\n",
       "      <td>28793.763793</td>\n",
       "      <td>0.417776</td>\n",
       "      <td>28788.543348</td>\n",
       "      <td>0.000008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2875</th>\n",
       "      <td>2875</td>\n",
       "      <td>2022-05-19 23:57:45</td>\n",
       "      <td>30338.635000</td>\n",
       "      <td>3.969417</td>\n",
       "      <td>30338.193333</td>\n",
       "      <td>0.415267</td>\n",
       "      <td>30337.791667</td>\n",
       "      <td>0.198983</td>\n",
       "      <td>30337.486667</td>\n",
       "      <td>0.348533</td>\n",
       "      <td>...</td>\n",
       "      <td>30341.966667</td>\n",
       "      <td>0.526467</td>\n",
       "      <td>30342.106667</td>\n",
       "      <td>1.050233</td>\n",
       "      <td>30342.235000</td>\n",
       "      <td>1.085950</td>\n",
       "      <td>30342.378333</td>\n",
       "      <td>0.718317</td>\n",
       "      <td>30338.642871</td>\n",
       "      <td>0.000003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2876</th>\n",
       "      <td>2876</td>\n",
       "      <td>2022-05-19 23:58:15</td>\n",
       "      <td>30326.763333</td>\n",
       "      <td>1.418667</td>\n",
       "      <td>30326.355000</td>\n",
       "      <td>0.436217</td>\n",
       "      <td>30325.738333</td>\n",
       "      <td>0.299033</td>\n",
       "      <td>30325.345000</td>\n",
       "      <td>0.070417</td>\n",
       "      <td>...</td>\n",
       "      <td>30330.665000</td>\n",
       "      <td>0.743667</td>\n",
       "      <td>30330.831667</td>\n",
       "      <td>1.051067</td>\n",
       "      <td>30331.013333</td>\n",
       "      <td>0.843717</td>\n",
       "      <td>30331.190000</td>\n",
       "      <td>1.326483</td>\n",
       "      <td>30326.837494</td>\n",
       "      <td>0.000003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2877</th>\n",
       "      <td>2877</td>\n",
       "      <td>2022-05-19 23:58:45</td>\n",
       "      <td>30319.180000</td>\n",
       "      <td>2.653744</td>\n",
       "      <td>30318.591111</td>\n",
       "      <td>0.620722</td>\n",
       "      <td>30318.086667</td>\n",
       "      <td>0.219122</td>\n",
       "      <td>30317.767222</td>\n",
       "      <td>0.326067</td>\n",
       "      <td>...</td>\n",
       "      <td>30322.896111</td>\n",
       "      <td>0.453300</td>\n",
       "      <td>30323.008333</td>\n",
       "      <td>0.712428</td>\n",
       "      <td>30323.123333</td>\n",
       "      <td>0.800967</td>\n",
       "      <td>30323.241667</td>\n",
       "      <td>0.675256</td>\n",
       "      <td>30319.209877</td>\n",
       "      <td>0.000003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2878</th>\n",
       "      <td>2878</td>\n",
       "      <td>2022-05-19 23:59:15</td>\n",
       "      <td>30319.861667</td>\n",
       "      <td>2.928600</td>\n",
       "      <td>30319.371667</td>\n",
       "      <td>0.253267</td>\n",
       "      <td>30319.170000</td>\n",
       "      <td>1.180767</td>\n",
       "      <td>30318.805000</td>\n",
       "      <td>1.711983</td>\n",
       "      <td>...</td>\n",
       "      <td>30323.665000</td>\n",
       "      <td>0.432567</td>\n",
       "      <td>30323.838333</td>\n",
       "      <td>0.591883</td>\n",
       "      <td>30324.008333</td>\n",
       "      <td>0.344667</td>\n",
       "      <td>30324.175000</td>\n",
       "      <td>0.469600</td>\n",
       "      <td>30319.925778</td>\n",
       "      <td>0.000003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2879</th>\n",
       "      <td>2879</td>\n",
       "      <td>2022-05-19 23:59:45</td>\n",
       "      <td>30320.560000</td>\n",
       "      <td>2.786183</td>\n",
       "      <td>30320.020000</td>\n",
       "      <td>0.357083</td>\n",
       "      <td>30319.876667</td>\n",
       "      <td>0.306200</td>\n",
       "      <td>30319.665000</td>\n",
       "      <td>0.771800</td>\n",
       "      <td>...</td>\n",
       "      <td>30323.726667</td>\n",
       "      <td>0.327533</td>\n",
       "      <td>30323.946667</td>\n",
       "      <td>0.375933</td>\n",
       "      <td>30324.131667</td>\n",
       "      <td>0.539817</td>\n",
       "      <td>30324.335000</td>\n",
       "      <td>0.620150</td>\n",
       "      <td>30320.631974</td>\n",
       "      <td>0.000003</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2880 rows × 84 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Unnamed: 0          primary_key           bp1       bs1           bp2  \\\n",
       "0              0  2022-05-19 00:00:15  28720.075362  3.772080  28719.533333   \n",
       "1              1  2022-05-19 00:00:45  28755.960345  1.878828  28755.570690   \n",
       "2              2  2022-05-19 00:01:15  28782.703333  1.771600  28782.131667   \n",
       "3              3  2022-05-19 00:01:45  28789.470000  1.767750  28788.766667   \n",
       "4              4  2022-05-19 00:02:15  28788.382759  1.634448  28787.839655   \n",
       "...          ...                  ...           ...       ...           ...   \n",
       "2875        2875  2022-05-19 23:57:45  30338.635000  3.969417  30338.193333   \n",
       "2876        2876  2022-05-19 23:58:15  30326.763333  1.418667  30326.355000   \n",
       "2877        2877  2022-05-19 23:58:45  30319.180000  2.653744  30318.591111   \n",
       "2878        2878  2022-05-19 23:59:15  30319.861667  2.928600  30319.371667   \n",
       "2879        2879  2022-05-19 23:59:45  30320.560000  2.786183  30320.020000   \n",
       "\n",
       "           bs2           bp3       bs3           bp4       bs4  ...  \\\n",
       "0     1.405377  28719.011594  0.626551  28718.371014  0.505087  ...   \n",
       "1     0.335776  28755.050000  0.573517  28754.684483  0.737259  ...   \n",
       "2     0.536867  28781.485000  0.402600  28781.140000  0.382467  ...   \n",
       "3     0.343217  28788.278333  0.243650  28788.046667  0.325300  ...   \n",
       "4     0.342845  28787.370690  0.396414  28787.039655  0.232483  ...   \n",
       "...        ...           ...       ...           ...       ...  ...   \n",
       "2875  0.415267  30337.791667  0.198983  30337.486667  0.348533  ...   \n",
       "2876  0.436217  30325.738333  0.299033  30325.345000  0.070417  ...   \n",
       "2877  0.620722  30318.086667  0.219122  30317.767222  0.326067  ...   \n",
       "2878  0.253267  30319.170000  1.180767  30318.805000  1.711983  ...   \n",
       "2879  0.357083  30319.876667  0.306200  30319.665000  0.771800  ...   \n",
       "\n",
       "              ap17      as17          ap18      as18          ap19      as19  \\\n",
       "0     28724.167391  0.487558  28724.361594  0.706101  28724.511594  0.707326   \n",
       "1     28760.674138  4.193586  28760.872414  0.620983  28761.067241  0.773897   \n",
       "2     28788.138333  0.580383  28788.331667  0.421400  28788.506667  0.376617   \n",
       "3     28794.093333  0.346800  28794.290000  1.161250  28794.483000  0.536067   \n",
       "4     28793.143103  0.595362  28793.387931  0.417172  28793.587931  0.473741   \n",
       "...            ...       ...           ...       ...           ...       ...   \n",
       "2875  30341.966667  0.526467  30342.106667  1.050233  30342.235000  1.085950   \n",
       "2876  30330.665000  0.743667  30330.831667  1.051067  30331.013333  0.843717   \n",
       "2877  30322.896111  0.453300  30323.008333  0.712428  30323.123333  0.800967   \n",
       "2878  30323.665000  0.432567  30323.838333  0.591883  30324.008333  0.344667   \n",
       "2879  30323.726667  0.327533  30323.946667  0.375933  30324.131667  0.539817   \n",
       "\n",
       "              ap20      as20           WAP    spread  \n",
       "0     28724.657971  1.104145  28720.131688  0.000015  \n",
       "1     28761.250000  0.854017  28756.179152  0.000010  \n",
       "2     28788.750000  0.425300  28782.777221  0.000005  \n",
       "3     28794.623333  0.395067  28789.546017  0.000005  \n",
       "4     28793.763793  0.417776  28788.543348  0.000008  \n",
       "...            ...       ...           ...       ...  \n",
       "2875  30342.378333  0.718317  30338.642871  0.000003  \n",
       "2876  30331.190000  1.326483  30326.837494  0.000003  \n",
       "2877  30323.241667  0.675256  30319.209877  0.000003  \n",
       "2878  30324.175000  0.469600  30319.925778  0.000003  \n",
       "2879  30324.335000  0.620150  30320.631974  0.000003  \n",
       "\n",
       "[2880 rows x 84 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "daafc24c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def log_price(list_stock_prices):\n",
    "    return np.log(list_stock_prices)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "6f12a7b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "data.insert(0, 'log_price', log_price(data['WAP']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "15b4b2a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "data['log_returns'] = data.log_price.diff()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3bcfc3b5",
   "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>log_price</th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>primary_key</th>\n",
       "      <th>bp1</th>\n",
       "      <th>bs1</th>\n",
       "      <th>bp2</th>\n",
       "      <th>bs2</th>\n",
       "      <th>bp3</th>\n",
       "      <th>bs3</th>\n",
       "      <th>bp4</th>\n",
       "      <th>...</th>\n",
       "      <th>as17</th>\n",
       "      <th>ap18</th>\n",
       "      <th>as18</th>\n",
       "      <th>ap19</th>\n",
       "      <th>as19</th>\n",
       "      <th>ap20</th>\n",
       "      <th>as20</th>\n",
       "      <th>WAP</th>\n",
       "      <th>spread</th>\n",
       "      <th>log_returns</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10.265354</td>\n",
       "      <td>0</td>\n",
       "      <td>2022-05-19 00:00:15</td>\n",
       "      <td>28720.075362</td>\n",
       "      <td>3.772080</td>\n",
       "      <td>28719.533333</td>\n",
       "      <td>1.405377</td>\n",
       "      <td>28719.011594</td>\n",
       "      <td>0.626551</td>\n",
       "      <td>28718.371014</td>\n",
       "      <td>...</td>\n",
       "      <td>0.487558</td>\n",
       "      <td>28724.361594</td>\n",
       "      <td>0.706101</td>\n",
       "      <td>28724.511594</td>\n",
       "      <td>0.707326</td>\n",
       "      <td>28724.657971</td>\n",
       "      <td>1.104145</td>\n",
       "      <td>28720.131688</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.266608</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-05-19 00:00:45</td>\n",
       "      <td>28755.960345</td>\n",
       "      <td>1.878828</td>\n",
       "      <td>28755.570690</td>\n",
       "      <td>0.335776</td>\n",
       "      <td>28755.050000</td>\n",
       "      <td>0.573517</td>\n",
       "      <td>28754.684483</td>\n",
       "      <td>...</td>\n",
       "      <td>4.193586</td>\n",
       "      <td>28760.872414</td>\n",
       "      <td>0.620983</td>\n",
       "      <td>28761.067241</td>\n",
       "      <td>0.773897</td>\n",
       "      <td>28761.250000</td>\n",
       "      <td>0.854017</td>\n",
       "      <td>28756.179152</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>0.001254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10.267532</td>\n",
       "      <td>2</td>\n",
       "      <td>2022-05-19 00:01:15</td>\n",
       "      <td>28782.703333</td>\n",
       "      <td>1.771600</td>\n",
       "      <td>28782.131667</td>\n",
       "      <td>0.536867</td>\n",
       "      <td>28781.485000</td>\n",
       "      <td>0.402600</td>\n",
       "      <td>28781.140000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.580383</td>\n",
       "      <td>28788.331667</td>\n",
       "      <td>0.421400</td>\n",
       "      <td>28788.506667</td>\n",
       "      <td>0.376617</td>\n",
       "      <td>28788.750000</td>\n",
       "      <td>0.425300</td>\n",
       "      <td>28782.777221</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10.267768</td>\n",
       "      <td>3</td>\n",
       "      <td>2022-05-19 00:01:45</td>\n",
       "      <td>28789.470000</td>\n",
       "      <td>1.767750</td>\n",
       "      <td>28788.766667</td>\n",
       "      <td>0.343217</td>\n",
       "      <td>28788.278333</td>\n",
       "      <td>0.243650</td>\n",
       "      <td>28788.046667</td>\n",
       "      <td>...</td>\n",
       "      <td>0.346800</td>\n",
       "      <td>28794.290000</td>\n",
       "      <td>1.161250</td>\n",
       "      <td>28794.483000</td>\n",
       "      <td>0.536067</td>\n",
       "      <td>28794.623333</td>\n",
       "      <td>0.395067</td>\n",
       "      <td>28789.546017</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10.267733</td>\n",
       "      <td>4</td>\n",
       "      <td>2022-05-19 00:02:15</td>\n",
       "      <td>28788.382759</td>\n",
       "      <td>1.634448</td>\n",
       "      <td>28787.839655</td>\n",
       "      <td>0.342845</td>\n",
       "      <td>28787.370690</td>\n",
       "      <td>0.396414</td>\n",
       "      <td>28787.039655</td>\n",
       "      <td>...</td>\n",
       "      <td>0.595362</td>\n",
       "      <td>28793.387931</td>\n",
       "      <td>0.417172</td>\n",
       "      <td>28793.587931</td>\n",
       "      <td>0.473741</td>\n",
       "      <td>28793.763793</td>\n",
       "      <td>0.417776</td>\n",
       "      <td>28788.543348</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>-0.000035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2875</th>\n",
       "      <td>10.320178</td>\n",
       "      <td>2875</td>\n",
       "      <td>2022-05-19 23:57:45</td>\n",
       "      <td>30338.635000</td>\n",
       "      <td>3.969417</td>\n",
       "      <td>30338.193333</td>\n",
       "      <td>0.415267</td>\n",
       "      <td>30337.791667</td>\n",
       "      <td>0.198983</td>\n",
       "      <td>30337.486667</td>\n",
       "      <td>...</td>\n",
       "      <td>0.526467</td>\n",
       "      <td>30342.106667</td>\n",
       "      <td>1.050233</td>\n",
       "      <td>30342.235000</td>\n",
       "      <td>1.085950</td>\n",
       "      <td>30342.378333</td>\n",
       "      <td>0.718317</td>\n",
       "      <td>30338.642871</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000406</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2876</th>\n",
       "      <td>10.319788</td>\n",
       "      <td>2876</td>\n",
       "      <td>2022-05-19 23:58:15</td>\n",
       "      <td>30326.763333</td>\n",
       "      <td>1.418667</td>\n",
       "      <td>30326.355000</td>\n",
       "      <td>0.436217</td>\n",
       "      <td>30325.738333</td>\n",
       "      <td>0.299033</td>\n",
       "      <td>30325.345000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.743667</td>\n",
       "      <td>30330.831667</td>\n",
       "      <td>1.051067</td>\n",
       "      <td>30331.013333</td>\n",
       "      <td>0.843717</td>\n",
       "      <td>30331.190000</td>\n",
       "      <td>1.326483</td>\n",
       "      <td>30326.837494</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-0.000389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2877</th>\n",
       "      <td>10.319537</td>\n",
       "      <td>2877</td>\n",
       "      <td>2022-05-19 23:58:45</td>\n",
       "      <td>30319.180000</td>\n",
       "      <td>2.653744</td>\n",
       "      <td>30318.591111</td>\n",
       "      <td>0.620722</td>\n",
       "      <td>30318.086667</td>\n",
       "      <td>0.219122</td>\n",
       "      <td>30317.767222</td>\n",
       "      <td>...</td>\n",
       "      <td>0.453300</td>\n",
       "      <td>30323.008333</td>\n",
       "      <td>0.712428</td>\n",
       "      <td>30323.123333</td>\n",
       "      <td>0.800967</td>\n",
       "      <td>30323.241667</td>\n",
       "      <td>0.675256</td>\n",
       "      <td>30319.209877</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-0.000252</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2878</th>\n",
       "      <td>10.319560</td>\n",
       "      <td>2878</td>\n",
       "      <td>2022-05-19 23:59:15</td>\n",
       "      <td>30319.861667</td>\n",
       "      <td>2.928600</td>\n",
       "      <td>30319.371667</td>\n",
       "      <td>0.253267</td>\n",
       "      <td>30319.170000</td>\n",
       "      <td>1.180767</td>\n",
       "      <td>30318.805000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.432567</td>\n",
       "      <td>30323.838333</td>\n",
       "      <td>0.591883</td>\n",
       "      <td>30324.008333</td>\n",
       "      <td>0.344667</td>\n",
       "      <td>30324.175000</td>\n",
       "      <td>0.469600</td>\n",
       "      <td>30319.925778</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2879</th>\n",
       "      <td>10.319584</td>\n",
       "      <td>2879</td>\n",
       "      <td>2022-05-19 23:59:45</td>\n",
       "      <td>30320.560000</td>\n",
       "      <td>2.786183</td>\n",
       "      <td>30320.020000</td>\n",
       "      <td>0.357083</td>\n",
       "      <td>30319.876667</td>\n",
       "      <td>0.306200</td>\n",
       "      <td>30319.665000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.327533</td>\n",
       "      <td>30323.946667</td>\n",
       "      <td>0.375933</td>\n",
       "      <td>30324.131667</td>\n",
       "      <td>0.539817</td>\n",
       "      <td>30324.335000</td>\n",
       "      <td>0.620150</td>\n",
       "      <td>30320.631974</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000023</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2880 rows × 86 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      log_price  Unnamed: 0          primary_key           bp1       bs1  \\\n",
       "0     10.265354           0  2022-05-19 00:00:15  28720.075362  3.772080   \n",
       "1     10.266608           1  2022-05-19 00:00:45  28755.960345  1.878828   \n",
       "2     10.267532           2  2022-05-19 00:01:15  28782.703333  1.771600   \n",
       "3     10.267768           3  2022-05-19 00:01:45  28789.470000  1.767750   \n",
       "4     10.267733           4  2022-05-19 00:02:15  28788.382759  1.634448   \n",
       "...         ...         ...                  ...           ...       ...   \n",
       "2875  10.320178        2875  2022-05-19 23:57:45  30338.635000  3.969417   \n",
       "2876  10.319788        2876  2022-05-19 23:58:15  30326.763333  1.418667   \n",
       "2877  10.319537        2877  2022-05-19 23:58:45  30319.180000  2.653744   \n",
       "2878  10.319560        2878  2022-05-19 23:59:15  30319.861667  2.928600   \n",
       "2879  10.319584        2879  2022-05-19 23:59:45  30320.560000  2.786183   \n",
       "\n",
       "               bp2       bs2           bp3       bs3           bp4  ...  \\\n",
       "0     28719.533333  1.405377  28719.011594  0.626551  28718.371014  ...   \n",
       "1     28755.570690  0.335776  28755.050000  0.573517  28754.684483  ...   \n",
       "2     28782.131667  0.536867  28781.485000  0.402600  28781.140000  ...   \n",
       "3     28788.766667  0.343217  28788.278333  0.243650  28788.046667  ...   \n",
       "4     28787.839655  0.342845  28787.370690  0.396414  28787.039655  ...   \n",
       "...            ...       ...           ...       ...           ...  ...   \n",
       "2875  30338.193333  0.415267  30337.791667  0.198983  30337.486667  ...   \n",
       "2876  30326.355000  0.436217  30325.738333  0.299033  30325.345000  ...   \n",
       "2877  30318.591111  0.620722  30318.086667  0.219122  30317.767222  ...   \n",
       "2878  30319.371667  0.253267  30319.170000  1.180767  30318.805000  ...   \n",
       "2879  30320.020000  0.357083  30319.876667  0.306200  30319.665000  ...   \n",
       "\n",
       "          as17          ap18      as18          ap19      as19          ap20  \\\n",
       "0     0.487558  28724.361594  0.706101  28724.511594  0.707326  28724.657971   \n",
       "1     4.193586  28760.872414  0.620983  28761.067241  0.773897  28761.250000   \n",
       "2     0.580383  28788.331667  0.421400  28788.506667  0.376617  28788.750000   \n",
       "3     0.346800  28794.290000  1.161250  28794.483000  0.536067  28794.623333   \n",
       "4     0.595362  28793.387931  0.417172  28793.587931  0.473741  28793.763793   \n",
       "...        ...           ...       ...           ...       ...           ...   \n",
       "2875  0.526467  30342.106667  1.050233  30342.235000  1.085950  30342.378333   \n",
       "2876  0.743667  30330.831667  1.051067  30331.013333  0.843717  30331.190000   \n",
       "2877  0.453300  30323.008333  0.712428  30323.123333  0.800967  30323.241667   \n",
       "2878  0.432567  30323.838333  0.591883  30324.008333  0.344667  30324.175000   \n",
       "2879  0.327533  30323.946667  0.375933  30324.131667  0.539817  30324.335000   \n",
       "\n",
       "          as20           WAP    spread  log_returns  \n",
       "0     1.104145  28720.131688  0.000015          NaN  \n",
       "1     0.854017  28756.179152  0.000010     0.001254  \n",
       "2     0.425300  28782.777221  0.000005     0.000925  \n",
       "3     0.395067  28789.546017  0.000005     0.000235  \n",
       "4     0.417776  28788.543348  0.000008    -0.000035  \n",
       "...        ...           ...       ...          ...  \n",
       "2875  0.718317  30338.642871  0.000003     0.000406  \n",
       "2876  1.326483  30326.837494  0.000003    -0.000389  \n",
       "2877  0.675256  30319.209877  0.000003    -0.000252  \n",
       "2878  0.469600  30319.925778  0.000003     0.000024  \n",
       "2879  0.620150  30320.631974  0.000003     0.000023  \n",
       "\n",
       "[2880 rows x 86 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1fcd5cba",
   "metadata": {},
   "outputs": [],
   "source": [
    "def realized_volatility():\n",
    "    list_vol = []\n",
    "    i = 0\n",
    "    for i in data.index:\n",
    "        x = np.std(data.log_returns.iloc[:i])\n",
    "        i += 1\n",
    "        list_vol.append(x)\n",
    "    \n",
    "    return list_vol"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d9a50d3b",
   "metadata": {},
   "outputs": [],
   "source": [
    "data['realized_volatility'] = realized_volatility()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "0ec37049",
   "metadata": {},
   "outputs": [],
   "source": [
    "data['volatility_t+1'] = data['realized_volatility'].shift(-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e510f4f4",
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(['realized_volatility'], axis = 1, inplace = True) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "76732e3d",
   "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>log_price</th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>primary_key</th>\n",
       "      <th>bp1</th>\n",
       "      <th>bs1</th>\n",
       "      <th>bp2</th>\n",
       "      <th>bs2</th>\n",
       "      <th>bp3</th>\n",
       "      <th>bs3</th>\n",
       "      <th>bp4</th>\n",
       "      <th>...</th>\n",
       "      <th>ap18</th>\n",
       "      <th>as18</th>\n",
       "      <th>ap19</th>\n",
       "      <th>as19</th>\n",
       "      <th>ap20</th>\n",
       "      <th>as20</th>\n",
       "      <th>WAP</th>\n",
       "      <th>spread</th>\n",
       "      <th>log_returns</th>\n",
       "      <th>volatility_t+1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10.265354</td>\n",
       "      <td>0</td>\n",
       "      <td>2022-05-19 00:00:15</td>\n",
       "      <td>28720.075362</td>\n",
       "      <td>3.772080</td>\n",
       "      <td>28719.533333</td>\n",
       "      <td>1.405377</td>\n",
       "      <td>28719.011594</td>\n",
       "      <td>0.626551</td>\n",
       "      <td>28718.371014</td>\n",
       "      <td>...</td>\n",
       "      <td>28724.361594</td>\n",
       "      <td>0.706101</td>\n",
       "      <td>28724.511594</td>\n",
       "      <td>0.707326</td>\n",
       "      <td>28724.657971</td>\n",
       "      <td>1.104145</td>\n",
       "      <td>28720.131688</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.266608</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-05-19 00:00:45</td>\n",
       "      <td>28755.960345</td>\n",
       "      <td>1.878828</td>\n",
       "      <td>28755.570690</td>\n",
       "      <td>0.335776</td>\n",
       "      <td>28755.050000</td>\n",
       "      <td>0.573517</td>\n",
       "      <td>28754.684483</td>\n",
       "      <td>...</td>\n",
       "      <td>28760.872414</td>\n",
       "      <td>0.620983</td>\n",
       "      <td>28761.067241</td>\n",
       "      <td>0.773897</td>\n",
       "      <td>28761.250000</td>\n",
       "      <td>0.854017</td>\n",
       "      <td>28756.179152</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>0.001254</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10.267532</td>\n",
       "      <td>2</td>\n",
       "      <td>2022-05-19 00:01:15</td>\n",
       "      <td>28782.703333</td>\n",
       "      <td>1.771600</td>\n",
       "      <td>28782.131667</td>\n",
       "      <td>0.536867</td>\n",
       "      <td>28781.485000</td>\n",
       "      <td>0.402600</td>\n",
       "      <td>28781.140000</td>\n",
       "      <td>...</td>\n",
       "      <td>28788.331667</td>\n",
       "      <td>0.421400</td>\n",
       "      <td>28788.506667</td>\n",
       "      <td>0.376617</td>\n",
       "      <td>28788.750000</td>\n",
       "      <td>0.425300</td>\n",
       "      <td>28782.777221</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000925</td>\n",
       "      <td>0.000165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10.267768</td>\n",
       "      <td>3</td>\n",
       "      <td>2022-05-19 00:01:45</td>\n",
       "      <td>28789.470000</td>\n",
       "      <td>1.767750</td>\n",
       "      <td>28788.766667</td>\n",
       "      <td>0.343217</td>\n",
       "      <td>28788.278333</td>\n",
       "      <td>0.243650</td>\n",
       "      <td>28788.046667</td>\n",
       "      <td>...</td>\n",
       "      <td>28794.290000</td>\n",
       "      <td>1.161250</td>\n",
       "      <td>28794.483000</td>\n",
       "      <td>0.536067</td>\n",
       "      <td>28794.623333</td>\n",
       "      <td>0.395067</td>\n",
       "      <td>28789.546017</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000235</td>\n",
       "      <td>0.000425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10.267733</td>\n",
       "      <td>4</td>\n",
       "      <td>2022-05-19 00:02:15</td>\n",
       "      <td>28788.382759</td>\n",
       "      <td>1.634448</td>\n",
       "      <td>28787.839655</td>\n",
       "      <td>0.342845</td>\n",
       "      <td>28787.370690</td>\n",
       "      <td>0.396414</td>\n",
       "      <td>28787.039655</td>\n",
       "      <td>...</td>\n",
       "      <td>28793.387931</td>\n",
       "      <td>0.417172</td>\n",
       "      <td>28793.587931</td>\n",
       "      <td>0.473741</td>\n",
       "      <td>28793.763793</td>\n",
       "      <td>0.417776</td>\n",
       "      <td>28788.543348</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>-0.000035</td>\n",
       "      <td>0.000517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2875</th>\n",
       "      <td>10.320178</td>\n",
       "      <td>2875</td>\n",
       "      <td>2022-05-19 23:57:45</td>\n",
       "      <td>30338.635000</td>\n",
       "      <td>3.969417</td>\n",
       "      <td>30338.193333</td>\n",
       "      <td>0.415267</td>\n",
       "      <td>30337.791667</td>\n",
       "      <td>0.198983</td>\n",
       "      <td>30337.486667</td>\n",
       "      <td>...</td>\n",
       "      <td>30342.106667</td>\n",
       "      <td>1.050233</td>\n",
       "      <td>30342.235000</td>\n",
       "      <td>1.085950</td>\n",
       "      <td>30342.378333</td>\n",
       "      <td>0.718317</td>\n",
       "      <td>30338.642871</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000406</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2876</th>\n",
       "      <td>10.319788</td>\n",
       "      <td>2876</td>\n",
       "      <td>2022-05-19 23:58:15</td>\n",
       "      <td>30326.763333</td>\n",
       "      <td>1.418667</td>\n",
       "      <td>30326.355000</td>\n",
       "      <td>0.436217</td>\n",
       "      <td>30325.738333</td>\n",
       "      <td>0.299033</td>\n",
       "      <td>30325.345000</td>\n",
       "      <td>...</td>\n",
       "      <td>30330.831667</td>\n",
       "      <td>1.051067</td>\n",
       "      <td>30331.013333</td>\n",
       "      <td>0.843717</td>\n",
       "      <td>30331.190000</td>\n",
       "      <td>1.326483</td>\n",
       "      <td>30326.837494</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-0.000389</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2877</th>\n",
       "      <td>10.319537</td>\n",
       "      <td>2877</td>\n",
       "      <td>2022-05-19 23:58:45</td>\n",
       "      <td>30319.180000</td>\n",
       "      <td>2.653744</td>\n",
       "      <td>30318.591111</td>\n",
       "      <td>0.620722</td>\n",
       "      <td>30318.086667</td>\n",
       "      <td>0.219122</td>\n",
       "      <td>30317.767222</td>\n",
       "      <td>...</td>\n",
       "      <td>30323.008333</td>\n",
       "      <td>0.712428</td>\n",
       "      <td>30323.123333</td>\n",
       "      <td>0.800967</td>\n",
       "      <td>30323.241667</td>\n",
       "      <td>0.675256</td>\n",
       "      <td>30319.209877</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-0.000252</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2878</th>\n",
       "      <td>10.319560</td>\n",
       "      <td>2878</td>\n",
       "      <td>2022-05-19 23:59:15</td>\n",
       "      <td>30319.861667</td>\n",
       "      <td>2.928600</td>\n",
       "      <td>30319.371667</td>\n",
       "      <td>0.253267</td>\n",
       "      <td>30319.170000</td>\n",
       "      <td>1.180767</td>\n",
       "      <td>30318.805000</td>\n",
       "      <td>...</td>\n",
       "      <td>30323.838333</td>\n",
       "      <td>0.591883</td>\n",
       "      <td>30324.008333</td>\n",
       "      <td>0.344667</td>\n",
       "      <td>30324.175000</td>\n",
       "      <td>0.469600</td>\n",
       "      <td>30319.925778</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000024</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2879</th>\n",
       "      <td>10.319584</td>\n",
       "      <td>2879</td>\n",
       "      <td>2022-05-19 23:59:45</td>\n",
       "      <td>30320.560000</td>\n",
       "      <td>2.786183</td>\n",
       "      <td>30320.020000</td>\n",
       "      <td>0.357083</td>\n",
       "      <td>30319.876667</td>\n",
       "      <td>0.306200</td>\n",
       "      <td>30319.665000</td>\n",
       "      <td>...</td>\n",
       "      <td>30323.946667</td>\n",
       "      <td>0.375933</td>\n",
       "      <td>30324.131667</td>\n",
       "      <td>0.539817</td>\n",
       "      <td>30324.335000</td>\n",
       "      <td>0.620150</td>\n",
       "      <td>30320.631974</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2880 rows × 87 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      log_price  Unnamed: 0          primary_key           bp1       bs1  \\\n",
       "0     10.265354           0  2022-05-19 00:00:15  28720.075362  3.772080   \n",
       "1     10.266608           1  2022-05-19 00:00:45  28755.960345  1.878828   \n",
       "2     10.267532           2  2022-05-19 00:01:15  28782.703333  1.771600   \n",
       "3     10.267768           3  2022-05-19 00:01:45  28789.470000  1.767750   \n",
       "4     10.267733           4  2022-05-19 00:02:15  28788.382759  1.634448   \n",
       "...         ...         ...                  ...           ...       ...   \n",
       "2875  10.320178        2875  2022-05-19 23:57:45  30338.635000  3.969417   \n",
       "2876  10.319788        2876  2022-05-19 23:58:15  30326.763333  1.418667   \n",
       "2877  10.319537        2877  2022-05-19 23:58:45  30319.180000  2.653744   \n",
       "2878  10.319560        2878  2022-05-19 23:59:15  30319.861667  2.928600   \n",
       "2879  10.319584        2879  2022-05-19 23:59:45  30320.560000  2.786183   \n",
       "\n",
       "               bp2       bs2           bp3       bs3           bp4  ...  \\\n",
       "0     28719.533333  1.405377  28719.011594  0.626551  28718.371014  ...   \n",
       "1     28755.570690  0.335776  28755.050000  0.573517  28754.684483  ...   \n",
       "2     28782.131667  0.536867  28781.485000  0.402600  28781.140000  ...   \n",
       "3     28788.766667  0.343217  28788.278333  0.243650  28788.046667  ...   \n",
       "4     28787.839655  0.342845  28787.370690  0.396414  28787.039655  ...   \n",
       "...            ...       ...           ...       ...           ...  ...   \n",
       "2875  30338.193333  0.415267  30337.791667  0.198983  30337.486667  ...   \n",
       "2876  30326.355000  0.436217  30325.738333  0.299033  30325.345000  ...   \n",
       "2877  30318.591111  0.620722  30318.086667  0.219122  30317.767222  ...   \n",
       "2878  30319.371667  0.253267  30319.170000  1.180767  30318.805000  ...   \n",
       "2879  30320.020000  0.357083  30319.876667  0.306200  30319.665000  ...   \n",
       "\n",
       "              ap18      as18          ap19      as19          ap20      as20  \\\n",
       "0     28724.361594  0.706101  28724.511594  0.707326  28724.657971  1.104145   \n",
       "1     28760.872414  0.620983  28761.067241  0.773897  28761.250000  0.854017   \n",
       "2     28788.331667  0.421400  28788.506667  0.376617  28788.750000  0.425300   \n",
       "3     28794.290000  1.161250  28794.483000  0.536067  28794.623333  0.395067   \n",
       "4     28793.387931  0.417172  28793.587931  0.473741  28793.763793  0.417776   \n",
       "...            ...       ...           ...       ...           ...       ...   \n",
       "2875  30342.106667  1.050233  30342.235000  1.085950  30342.378333  0.718317   \n",
       "2876  30330.831667  1.051067  30331.013333  0.843717  30331.190000  1.326483   \n",
       "2877  30323.008333  0.712428  30323.123333  0.800967  30323.241667  0.675256   \n",
       "2878  30323.838333  0.591883  30324.008333  0.344667  30324.175000  0.469600   \n",
       "2879  30323.946667  0.375933  30324.131667  0.539817  30324.335000  0.620150   \n",
       "\n",
       "               WAP    spread  log_returns  volatility_t+1  \n",
       "0     28720.131688  0.000015          NaN             NaN  \n",
       "1     28756.179152  0.000010     0.001254        0.000000  \n",
       "2     28782.777221  0.000005     0.000925        0.000165  \n",
       "3     28789.546017  0.000005     0.000235        0.000425  \n",
       "4     28788.543348  0.000008    -0.000035        0.000517  \n",
       "...            ...       ...          ...             ...  \n",
       "2875  30338.642871  0.000003     0.000406        0.000558  \n",
       "2876  30326.837494  0.000003    -0.000389        0.000558  \n",
       "2877  30319.209877  0.000003    -0.000252        0.000558  \n",
       "2878  30319.925778  0.000003     0.000024        0.000558  \n",
       "2879  30320.631974  0.000003     0.000023             NaN  \n",
       "\n",
       "[2880 rows x 87 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "729dbd9b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    2880\n",
       "dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.duplicated().value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "610ef144",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.impute import SimpleImputer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "0138e7c6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputer = SimpleImputer(strategy=\"constant\", fill_value = 0) # Instantiate a SimpleImputer object with your strategy of choice\n",
    "\n",
    "imputer.fit(data[['volatility_t+1']]) # Call the \"fit\" method on the object\n",
    "\n",
    "data['volatility_t+1'] = imputer.transform(data[['volatility_t+1']]) # Call the \"transform\" method on the object\n",
    "\n",
    "imputer.statistics_ # The mean is stored in the transformer's memory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "515c94a7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.001254])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputer2 = SimpleImputer(strategy=\"constant\", fill_value = 0.001254)\n",
    "imputer2.fit(data[['log_returns']]) # Call the \"fit\" method on the object\n",
    "\n",
    "data['log_returns'] = imputer2.transform(data[['log_returns']]) # Call the \"transform\" method on the object\n",
    "\n",
    "imputer2.statistics_ # The mean is stored in the transformer's memory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "6b8dd5fd",
   "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>log_price</th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>primary_key</th>\n",
       "      <th>bp1</th>\n",
       "      <th>bs1</th>\n",
       "      <th>bp2</th>\n",
       "      <th>bs2</th>\n",
       "      <th>bp3</th>\n",
       "      <th>bs3</th>\n",
       "      <th>bp4</th>\n",
       "      <th>...</th>\n",
       "      <th>ap18</th>\n",
       "      <th>as18</th>\n",
       "      <th>ap19</th>\n",
       "      <th>as19</th>\n",
       "      <th>ap20</th>\n",
       "      <th>as20</th>\n",
       "      <th>WAP</th>\n",
       "      <th>spread</th>\n",
       "      <th>log_returns</th>\n",
       "      <th>volatility_t+1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10.265354</td>\n",
       "      <td>0</td>\n",
       "      <td>2022-05-19 00:00:15</td>\n",
       "      <td>28720.075362</td>\n",
       "      <td>3.772080</td>\n",
       "      <td>28719.533333</td>\n",
       "      <td>1.405377</td>\n",
       "      <td>28719.011594</td>\n",
       "      <td>0.626551</td>\n",
       "      <td>28718.371014</td>\n",
       "      <td>...</td>\n",
       "      <td>28724.361594</td>\n",
       "      <td>0.706101</td>\n",
       "      <td>28724.511594</td>\n",
       "      <td>0.707326</td>\n",
       "      <td>28724.657971</td>\n",
       "      <td>1.104145</td>\n",
       "      <td>28720.131688</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.001254</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.266608</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-05-19 00:00:45</td>\n",
       "      <td>28755.960345</td>\n",
       "      <td>1.878828</td>\n",
       "      <td>28755.570690</td>\n",
       "      <td>0.335776</td>\n",
       "      <td>28755.050000</td>\n",
       "      <td>0.573517</td>\n",
       "      <td>28754.684483</td>\n",
       "      <td>...</td>\n",
       "      <td>28760.872414</td>\n",
       "      <td>0.620983</td>\n",
       "      <td>28761.067241</td>\n",
       "      <td>0.773897</td>\n",
       "      <td>28761.250000</td>\n",
       "      <td>0.854017</td>\n",
       "      <td>28756.179152</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>0.001254</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10.267532</td>\n",
       "      <td>2</td>\n",
       "      <td>2022-05-19 00:01:15</td>\n",
       "      <td>28782.703333</td>\n",
       "      <td>1.771600</td>\n",
       "      <td>28782.131667</td>\n",
       "      <td>0.536867</td>\n",
       "      <td>28781.485000</td>\n",
       "      <td>0.402600</td>\n",
       "      <td>28781.140000</td>\n",
       "      <td>...</td>\n",
       "      <td>28788.331667</td>\n",
       "      <td>0.421400</td>\n",
       "      <td>28788.506667</td>\n",
       "      <td>0.376617</td>\n",
       "      <td>28788.750000</td>\n",
       "      <td>0.425300</td>\n",
       "      <td>28782.777221</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000925</td>\n",
       "      <td>0.000165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10.267768</td>\n",
       "      <td>3</td>\n",
       "      <td>2022-05-19 00:01:45</td>\n",
       "      <td>28789.470000</td>\n",
       "      <td>1.767750</td>\n",
       "      <td>28788.766667</td>\n",
       "      <td>0.343217</td>\n",
       "      <td>28788.278333</td>\n",
       "      <td>0.243650</td>\n",
       "      <td>28788.046667</td>\n",
       "      <td>...</td>\n",
       "      <td>28794.290000</td>\n",
       "      <td>1.161250</td>\n",
       "      <td>28794.483000</td>\n",
       "      <td>0.536067</td>\n",
       "      <td>28794.623333</td>\n",
       "      <td>0.395067</td>\n",
       "      <td>28789.546017</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000235</td>\n",
       "      <td>0.000425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10.267733</td>\n",
       "      <td>4</td>\n",
       "      <td>2022-05-19 00:02:15</td>\n",
       "      <td>28788.382759</td>\n",
       "      <td>1.634448</td>\n",
       "      <td>28787.839655</td>\n",
       "      <td>0.342845</td>\n",
       "      <td>28787.370690</td>\n",
       "      <td>0.396414</td>\n",
       "      <td>28787.039655</td>\n",
       "      <td>...</td>\n",
       "      <td>28793.387931</td>\n",
       "      <td>0.417172</td>\n",
       "      <td>28793.587931</td>\n",
       "      <td>0.473741</td>\n",
       "      <td>28793.763793</td>\n",
       "      <td>0.417776</td>\n",
       "      <td>28788.543348</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>-0.000035</td>\n",
       "      <td>0.000517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2875</th>\n",
       "      <td>10.320178</td>\n",
       "      <td>2875</td>\n",
       "      <td>2022-05-19 23:57:45</td>\n",
       "      <td>30338.635000</td>\n",
       "      <td>3.969417</td>\n",
       "      <td>30338.193333</td>\n",
       "      <td>0.415267</td>\n",
       "      <td>30337.791667</td>\n",
       "      <td>0.198983</td>\n",
       "      <td>30337.486667</td>\n",
       "      <td>...</td>\n",
       "      <td>30342.106667</td>\n",
       "      <td>1.050233</td>\n",
       "      <td>30342.235000</td>\n",
       "      <td>1.085950</td>\n",
       "      <td>30342.378333</td>\n",
       "      <td>0.718317</td>\n",
       "      <td>30338.642871</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000406</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2876</th>\n",
       "      <td>10.319788</td>\n",
       "      <td>2876</td>\n",
       "      <td>2022-05-19 23:58:15</td>\n",
       "      <td>30326.763333</td>\n",
       "      <td>1.418667</td>\n",
       "      <td>30326.355000</td>\n",
       "      <td>0.436217</td>\n",
       "      <td>30325.738333</td>\n",
       "      <td>0.299033</td>\n",
       "      <td>30325.345000</td>\n",
       "      <td>...</td>\n",
       "      <td>30330.831667</td>\n",
       "      <td>1.051067</td>\n",
       "      <td>30331.013333</td>\n",
       "      <td>0.843717</td>\n",
       "      <td>30331.190000</td>\n",
       "      <td>1.326483</td>\n",
       "      <td>30326.837494</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-0.000389</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2877</th>\n",
       "      <td>10.319537</td>\n",
       "      <td>2877</td>\n",
       "      <td>2022-05-19 23:58:45</td>\n",
       "      <td>30319.180000</td>\n",
       "      <td>2.653744</td>\n",
       "      <td>30318.591111</td>\n",
       "      <td>0.620722</td>\n",
       "      <td>30318.086667</td>\n",
       "      <td>0.219122</td>\n",
       "      <td>30317.767222</td>\n",
       "      <td>...</td>\n",
       "      <td>30323.008333</td>\n",
       "      <td>0.712428</td>\n",
       "      <td>30323.123333</td>\n",
       "      <td>0.800967</td>\n",
       "      <td>30323.241667</td>\n",
       "      <td>0.675256</td>\n",
       "      <td>30319.209877</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-0.000252</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2878</th>\n",
       "      <td>10.319560</td>\n",
       "      <td>2878</td>\n",
       "      <td>2022-05-19 23:59:15</td>\n",
       "      <td>30319.861667</td>\n",
       "      <td>2.928600</td>\n",
       "      <td>30319.371667</td>\n",
       "      <td>0.253267</td>\n",
       "      <td>30319.170000</td>\n",
       "      <td>1.180767</td>\n",
       "      <td>30318.805000</td>\n",
       "      <td>...</td>\n",
       "      <td>30323.838333</td>\n",
       "      <td>0.591883</td>\n",
       "      <td>30324.008333</td>\n",
       "      <td>0.344667</td>\n",
       "      <td>30324.175000</td>\n",
       "      <td>0.469600</td>\n",
       "      <td>30319.925778</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000024</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2879</th>\n",
       "      <td>10.319584</td>\n",
       "      <td>2879</td>\n",
       "      <td>2022-05-19 23:59:45</td>\n",
       "      <td>30320.560000</td>\n",
       "      <td>2.786183</td>\n",
       "      <td>30320.020000</td>\n",
       "      <td>0.357083</td>\n",
       "      <td>30319.876667</td>\n",
       "      <td>0.306200</td>\n",
       "      <td>30319.665000</td>\n",
       "      <td>...</td>\n",
       "      <td>30323.946667</td>\n",
       "      <td>0.375933</td>\n",
       "      <td>30324.131667</td>\n",
       "      <td>0.539817</td>\n",
       "      <td>30324.335000</td>\n",
       "      <td>0.620150</td>\n",
       "      <td>30320.631974</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2880 rows × 87 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      log_price  Unnamed: 0          primary_key           bp1       bs1  \\\n",
       "0     10.265354           0  2022-05-19 00:00:15  28720.075362  3.772080   \n",
       "1     10.266608           1  2022-05-19 00:00:45  28755.960345  1.878828   \n",
       "2     10.267532           2  2022-05-19 00:01:15  28782.703333  1.771600   \n",
       "3     10.267768           3  2022-05-19 00:01:45  28789.470000  1.767750   \n",
       "4     10.267733           4  2022-05-19 00:02:15  28788.382759  1.634448   \n",
       "...         ...         ...                  ...           ...       ...   \n",
       "2875  10.320178        2875  2022-05-19 23:57:45  30338.635000  3.969417   \n",
       "2876  10.319788        2876  2022-05-19 23:58:15  30326.763333  1.418667   \n",
       "2877  10.319537        2877  2022-05-19 23:58:45  30319.180000  2.653744   \n",
       "2878  10.319560        2878  2022-05-19 23:59:15  30319.861667  2.928600   \n",
       "2879  10.319584        2879  2022-05-19 23:59:45  30320.560000  2.786183   \n",
       "\n",
       "               bp2       bs2           bp3       bs3           bp4  ...  \\\n",
       "0     28719.533333  1.405377  28719.011594  0.626551  28718.371014  ...   \n",
       "1     28755.570690  0.335776  28755.050000  0.573517  28754.684483  ...   \n",
       "2     28782.131667  0.536867  28781.485000  0.402600  28781.140000  ...   \n",
       "3     28788.766667  0.343217  28788.278333  0.243650  28788.046667  ...   \n",
       "4     28787.839655  0.342845  28787.370690  0.396414  28787.039655  ...   \n",
       "...            ...       ...           ...       ...           ...  ...   \n",
       "2875  30338.193333  0.415267  30337.791667  0.198983  30337.486667  ...   \n",
       "2876  30326.355000  0.436217  30325.738333  0.299033  30325.345000  ...   \n",
       "2877  30318.591111  0.620722  30318.086667  0.219122  30317.767222  ...   \n",
       "2878  30319.371667  0.253267  30319.170000  1.180767  30318.805000  ...   \n",
       "2879  30320.020000  0.357083  30319.876667  0.306200  30319.665000  ...   \n",
       "\n",
       "              ap18      as18          ap19      as19          ap20      as20  \\\n",
       "0     28724.361594  0.706101  28724.511594  0.707326  28724.657971  1.104145   \n",
       "1     28760.872414  0.620983  28761.067241  0.773897  28761.250000  0.854017   \n",
       "2     28788.331667  0.421400  28788.506667  0.376617  28788.750000  0.425300   \n",
       "3     28794.290000  1.161250  28794.483000  0.536067  28794.623333  0.395067   \n",
       "4     28793.387931  0.417172  28793.587931  0.473741  28793.763793  0.417776   \n",
       "...            ...       ...           ...       ...           ...       ...   \n",
       "2875  30342.106667  1.050233  30342.235000  1.085950  30342.378333  0.718317   \n",
       "2876  30330.831667  1.051067  30331.013333  0.843717  30331.190000  1.326483   \n",
       "2877  30323.008333  0.712428  30323.123333  0.800967  30323.241667  0.675256   \n",
       "2878  30323.838333  0.591883  30324.008333  0.344667  30324.175000  0.469600   \n",
       "2879  30323.946667  0.375933  30324.131667  0.539817  30324.335000  0.620150   \n",
       "\n",
       "               WAP    spread  log_returns  volatility_t+1  \n",
       "0     28720.131688  0.000015     0.001254        0.000000  \n",
       "1     28756.179152  0.000010     0.001254        0.000000  \n",
       "2     28782.777221  0.000005     0.000925        0.000165  \n",
       "3     28789.546017  0.000005     0.000235        0.000425  \n",
       "4     28788.543348  0.000008    -0.000035        0.000517  \n",
       "...            ...       ...          ...             ...  \n",
       "2875  30338.642871  0.000003     0.000406        0.000558  \n",
       "2876  30326.837494  0.000003    -0.000389        0.000558  \n",
       "2877  30319.209877  0.000003    -0.000252        0.000558  \n",
       "2878  30319.925778  0.000003     0.000024        0.000558  \n",
       "2879  30320.631974  0.000003     0.000023        0.000000  \n",
       "\n",
       "[2880 rows x 87 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "6f246fde",
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(['Unnamed: 0'], axis = 1, inplace = True) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "bf505412",
   "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>log_price</th>\n",
       "      <th>primary_key</th>\n",
       "      <th>bp1</th>\n",
       "      <th>bs1</th>\n",
       "      <th>bp2</th>\n",
       "      <th>bs2</th>\n",
       "      <th>bp3</th>\n",
       "      <th>bs3</th>\n",
       "      <th>bp4</th>\n",
       "      <th>bs4</th>\n",
       "      <th>...</th>\n",
       "      <th>ap18</th>\n",
       "      <th>as18</th>\n",
       "      <th>ap19</th>\n",
       "      <th>as19</th>\n",
       "      <th>ap20</th>\n",
       "      <th>as20</th>\n",
       "      <th>WAP</th>\n",
       "      <th>spread</th>\n",
       "      <th>log_returns</th>\n",
       "      <th>volatility_t+1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10.265354</td>\n",
       "      <td>2022-05-19 00:00:15</td>\n",
       "      <td>28720.075362</td>\n",
       "      <td>3.772080</td>\n",
       "      <td>28719.533333</td>\n",
       "      <td>1.405377</td>\n",
       "      <td>28719.011594</td>\n",
       "      <td>0.626551</td>\n",
       "      <td>28718.371014</td>\n",
       "      <td>0.505087</td>\n",
       "      <td>...</td>\n",
       "      <td>28724.361594</td>\n",
       "      <td>0.706101</td>\n",
       "      <td>28724.511594</td>\n",
       "      <td>0.707326</td>\n",
       "      <td>28724.657971</td>\n",
       "      <td>1.104145</td>\n",
       "      <td>28720.131688</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.001254</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.266608</td>\n",
       "      <td>2022-05-19 00:00:45</td>\n",
       "      <td>28755.960345</td>\n",
       "      <td>1.878828</td>\n",
       "      <td>28755.570690</td>\n",
       "      <td>0.335776</td>\n",
       "      <td>28755.050000</td>\n",
       "      <td>0.573517</td>\n",
       "      <td>28754.684483</td>\n",
       "      <td>0.737259</td>\n",
       "      <td>...</td>\n",
       "      <td>28760.872414</td>\n",
       "      <td>0.620983</td>\n",
       "      <td>28761.067241</td>\n",
       "      <td>0.773897</td>\n",
       "      <td>28761.250000</td>\n",
       "      <td>0.854017</td>\n",
       "      <td>28756.179152</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>0.001254</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10.267532</td>\n",
       "      <td>2022-05-19 00:01:15</td>\n",
       "      <td>28782.703333</td>\n",
       "      <td>1.771600</td>\n",
       "      <td>28782.131667</td>\n",
       "      <td>0.536867</td>\n",
       "      <td>28781.485000</td>\n",
       "      <td>0.402600</td>\n",
       "      <td>28781.140000</td>\n",
       "      <td>0.382467</td>\n",
       "      <td>...</td>\n",
       "      <td>28788.331667</td>\n",
       "      <td>0.421400</td>\n",
       "      <td>28788.506667</td>\n",
       "      <td>0.376617</td>\n",
       "      <td>28788.750000</td>\n",
       "      <td>0.425300</td>\n",
       "      <td>28782.777221</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000925</td>\n",
       "      <td>0.000165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10.267768</td>\n",
       "      <td>2022-05-19 00:01:45</td>\n",
       "      <td>28789.470000</td>\n",
       "      <td>1.767750</td>\n",
       "      <td>28788.766667</td>\n",
       "      <td>0.343217</td>\n",
       "      <td>28788.278333</td>\n",
       "      <td>0.243650</td>\n",
       "      <td>28788.046667</td>\n",
       "      <td>0.325300</td>\n",
       "      <td>...</td>\n",
       "      <td>28794.290000</td>\n",
       "      <td>1.161250</td>\n",
       "      <td>28794.483000</td>\n",
       "      <td>0.536067</td>\n",
       "      <td>28794.623333</td>\n",
       "      <td>0.395067</td>\n",
       "      <td>28789.546017</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000235</td>\n",
       "      <td>0.000425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10.267733</td>\n",
       "      <td>2022-05-19 00:02:15</td>\n",
       "      <td>28788.382759</td>\n",
       "      <td>1.634448</td>\n",
       "      <td>28787.839655</td>\n",
       "      <td>0.342845</td>\n",
       "      <td>28787.370690</td>\n",
       "      <td>0.396414</td>\n",
       "      <td>28787.039655</td>\n",
       "      <td>0.232483</td>\n",
       "      <td>...</td>\n",
       "      <td>28793.387931</td>\n",
       "      <td>0.417172</td>\n",
       "      <td>28793.587931</td>\n",
       "      <td>0.473741</td>\n",
       "      <td>28793.763793</td>\n",
       "      <td>0.417776</td>\n",
       "      <td>28788.543348</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>-0.000035</td>\n",
       "      <td>0.000517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2875</th>\n",
       "      <td>10.320178</td>\n",
       "      <td>2022-05-19 23:57:45</td>\n",
       "      <td>30338.635000</td>\n",
       "      <td>3.969417</td>\n",
       "      <td>30338.193333</td>\n",
       "      <td>0.415267</td>\n",
       "      <td>30337.791667</td>\n",
       "      <td>0.198983</td>\n",
       "      <td>30337.486667</td>\n",
       "      <td>0.348533</td>\n",
       "      <td>...</td>\n",
       "      <td>30342.106667</td>\n",
       "      <td>1.050233</td>\n",
       "      <td>30342.235000</td>\n",
       "      <td>1.085950</td>\n",
       "      <td>30342.378333</td>\n",
       "      <td>0.718317</td>\n",
       "      <td>30338.642871</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000406</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2876</th>\n",
       "      <td>10.319788</td>\n",
       "      <td>2022-05-19 23:58:15</td>\n",
       "      <td>30326.763333</td>\n",
       "      <td>1.418667</td>\n",
       "      <td>30326.355000</td>\n",
       "      <td>0.436217</td>\n",
       "      <td>30325.738333</td>\n",
       "      <td>0.299033</td>\n",
       "      <td>30325.345000</td>\n",
       "      <td>0.070417</td>\n",
       "      <td>...</td>\n",
       "      <td>30330.831667</td>\n",
       "      <td>1.051067</td>\n",
       "      <td>30331.013333</td>\n",
       "      <td>0.843717</td>\n",
       "      <td>30331.190000</td>\n",
       "      <td>1.326483</td>\n",
       "      <td>30326.837494</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-0.000389</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2877</th>\n",
       "      <td>10.319537</td>\n",
       "      <td>2022-05-19 23:58:45</td>\n",
       "      <td>30319.180000</td>\n",
       "      <td>2.653744</td>\n",
       "      <td>30318.591111</td>\n",
       "      <td>0.620722</td>\n",
       "      <td>30318.086667</td>\n",
       "      <td>0.219122</td>\n",
       "      <td>30317.767222</td>\n",
       "      <td>0.326067</td>\n",
       "      <td>...</td>\n",
       "      <td>30323.008333</td>\n",
       "      <td>0.712428</td>\n",
       "      <td>30323.123333</td>\n",
       "      <td>0.800967</td>\n",
       "      <td>30323.241667</td>\n",
       "      <td>0.675256</td>\n",
       "      <td>30319.209877</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-0.000252</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2878</th>\n",
       "      <td>10.319560</td>\n",
       "      <td>2022-05-19 23:59:15</td>\n",
       "      <td>30319.861667</td>\n",
       "      <td>2.928600</td>\n",
       "      <td>30319.371667</td>\n",
       "      <td>0.253267</td>\n",
       "      <td>30319.170000</td>\n",
       "      <td>1.180767</td>\n",
       "      <td>30318.805000</td>\n",
       "      <td>1.711983</td>\n",
       "      <td>...</td>\n",
       "      <td>30323.838333</td>\n",
       "      <td>0.591883</td>\n",
       "      <td>30324.008333</td>\n",
       "      <td>0.344667</td>\n",
       "      <td>30324.175000</td>\n",
       "      <td>0.469600</td>\n",
       "      <td>30319.925778</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000024</td>\n",
       "      <td>0.000558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2879</th>\n",
       "      <td>10.319584</td>\n",
       "      <td>2022-05-19 23:59:45</td>\n",
       "      <td>30320.560000</td>\n",
       "      <td>2.786183</td>\n",
       "      <td>30320.020000</td>\n",
       "      <td>0.357083</td>\n",
       "      <td>30319.876667</td>\n",
       "      <td>0.306200</td>\n",
       "      <td>30319.665000</td>\n",
       "      <td>0.771800</td>\n",
       "      <td>...</td>\n",
       "      <td>30323.946667</td>\n",
       "      <td>0.375933</td>\n",
       "      <td>30324.131667</td>\n",
       "      <td>0.539817</td>\n",
       "      <td>30324.335000</td>\n",
       "      <td>0.620150</td>\n",
       "      <td>30320.631974</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2880 rows × 86 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      log_price          primary_key           bp1       bs1           bp2  \\\n",
       "0     10.265354  2022-05-19 00:00:15  28720.075362  3.772080  28719.533333   \n",
       "1     10.266608  2022-05-19 00:00:45  28755.960345  1.878828  28755.570690   \n",
       "2     10.267532  2022-05-19 00:01:15  28782.703333  1.771600  28782.131667   \n",
       "3     10.267768  2022-05-19 00:01:45  28789.470000  1.767750  28788.766667   \n",
       "4     10.267733  2022-05-19 00:02:15  28788.382759  1.634448  28787.839655   \n",
       "...         ...                  ...           ...       ...           ...   \n",
       "2875  10.320178  2022-05-19 23:57:45  30338.635000  3.969417  30338.193333   \n",
       "2876  10.319788  2022-05-19 23:58:15  30326.763333  1.418667  30326.355000   \n",
       "2877  10.319537  2022-05-19 23:58:45  30319.180000  2.653744  30318.591111   \n",
       "2878  10.319560  2022-05-19 23:59:15  30319.861667  2.928600  30319.371667   \n",
       "2879  10.319584  2022-05-19 23:59:45  30320.560000  2.786183  30320.020000   \n",
       "\n",
       "           bs2           bp3       bs3           bp4       bs4  ...  \\\n",
       "0     1.405377  28719.011594  0.626551  28718.371014  0.505087  ...   \n",
       "1     0.335776  28755.050000  0.573517  28754.684483  0.737259  ...   \n",
       "2     0.536867  28781.485000  0.402600  28781.140000  0.382467  ...   \n",
       "3     0.343217  28788.278333  0.243650  28788.046667  0.325300  ...   \n",
       "4     0.342845  28787.370690  0.396414  28787.039655  0.232483  ...   \n",
       "...        ...           ...       ...           ...       ...  ...   \n",
       "2875  0.415267  30337.791667  0.198983  30337.486667  0.348533  ...   \n",
       "2876  0.436217  30325.738333  0.299033  30325.345000  0.070417  ...   \n",
       "2877  0.620722  30318.086667  0.219122  30317.767222  0.326067  ...   \n",
       "2878  0.253267  30319.170000  1.180767  30318.805000  1.711983  ...   \n",
       "2879  0.357083  30319.876667  0.306200  30319.665000  0.771800  ...   \n",
       "\n",
       "              ap18      as18          ap19      as19          ap20      as20  \\\n",
       "0     28724.361594  0.706101  28724.511594  0.707326  28724.657971  1.104145   \n",
       "1     28760.872414  0.620983  28761.067241  0.773897  28761.250000  0.854017   \n",
       "2     28788.331667  0.421400  28788.506667  0.376617  28788.750000  0.425300   \n",
       "3     28794.290000  1.161250  28794.483000  0.536067  28794.623333  0.395067   \n",
       "4     28793.387931  0.417172  28793.587931  0.473741  28793.763793  0.417776   \n",
       "...            ...       ...           ...       ...           ...       ...   \n",
       "2875  30342.106667  1.050233  30342.235000  1.085950  30342.378333  0.718317   \n",
       "2876  30330.831667  1.051067  30331.013333  0.843717  30331.190000  1.326483   \n",
       "2877  30323.008333  0.712428  30323.123333  0.800967  30323.241667  0.675256   \n",
       "2878  30323.838333  0.591883  30324.008333  0.344667  30324.175000  0.469600   \n",
       "2879  30323.946667  0.375933  30324.131667  0.539817  30324.335000  0.620150   \n",
       "\n",
       "               WAP    spread  log_returns  volatility_t+1  \n",
       "0     28720.131688  0.000015     0.001254        0.000000  \n",
       "1     28756.179152  0.000010     0.001254        0.000000  \n",
       "2     28782.777221  0.000005     0.000925        0.000165  \n",
       "3     28789.546017  0.000005     0.000235        0.000425  \n",
       "4     28788.543348  0.000008    -0.000035        0.000517  \n",
       "...            ...       ...          ...             ...  \n",
       "2875  30338.642871  0.000003     0.000406        0.000558  \n",
       "2876  30326.837494  0.000003    -0.000389        0.000558  \n",
       "2877  30319.209877  0.000003    -0.000252        0.000558  \n",
       "2878  30319.925778  0.000003     0.000024        0.000558  \n",
       "2879  30320.631974  0.000003     0.000023        0.000000  \n",
       "\n",
       "[2880 rows x 86 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "511f1a2b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['log_price', 'primary_key', 'bp1', 'bs1', 'bp2', 'bs2', 'bp3', 'bs3',\n",
       "       'bp4', 'bs4', 'bp5', 'bs5', 'bp6', 'bs6', 'bp7', 'bs7', 'bp8', 'bs8',\n",
       "       'bp9', 'bs9', 'bp10', 'bs10', 'bp11', 'bs11', 'bp12', 'bs12', 'bp13',\n",
       "       'bs13', 'bp14', 'bs14', 'bp15', 'bs15', 'bp16', 'bs16', 'bp17', 'bs17',\n",
       "       'bp18', 'bs18', 'bp19', 'bs19', 'bp20', 'bs20', 'ap1', 'as1', 'ap2',\n",
       "       'as2', 'ap3', 'as3', 'ap4', 'as4', 'ap5', 'as5', 'ap6', 'as6', 'ap7',\n",
       "       'as7', 'ap8', 'as8', 'ap9', 'as9', 'ap10', 'as10', 'ap11', 'as11',\n",
       "       'ap12', 'as12', 'ap13', 'as13', 'ap14', 'as14', 'ap15', 'as15', 'ap16',\n",
       "       'as16', 'ap17', 'as17', 'ap18', 'as18', 'ap19', 'as19', 'ap20', 'as20',\n",
       "       'WAP', 'spread', 'log_returns', 'volatility_t+1'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "0bf38910",
   "metadata": {},
   "outputs": [],
   "source": [
    "#sum of all bid quantities\n",
    "data['bid depth'] = data[['bs1', 'bs2', 'bs3','bs4', 'bs5', 'bs6','bs7', 'bs8', 'bs9','bs10',\n",
    "                         'bs11', 'bs12', 'bs13','bs14', 'bs15', 'bs16','bs17', 'bs18', 'bs19','bs20']].sum(axis=1)\n",
    "      "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "ffe84ec2",
   "metadata": {},
   "outputs": [],
   "source": [
    "#sum of all bid quantities\n",
    "data['ask depth'] = data[['as1', 'as2', 'as3','as4', 'as5', 'as6','as7', 'as8', 'as9','as10',\n",
    "                         'as11', 'as12', 'as13','as14', 'as15', 'as16','as17', 'as18', 'as19','as20']].sum(axis=1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "4deb5e33",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Order Flow Imbalance (OFI) \n",
    "#relative quantities of bids vs asks\n",
    "#full depth (approx 20 levels) OFI\n",
    "data['FDOFI'] = (data['bid depth']-data['ask depth'])/(data['bid depth']+data['ask depth'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "ab5a1ec8",
   "metadata": {},
   "outputs": [],
   "source": [
    "data=data.set_index('primary_key')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "50ba1cfe",
   "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>log_price</th>\n",
       "      <th>bp1</th>\n",
       "      <th>bs1</th>\n",
       "      <th>bp2</th>\n",
       "      <th>bs2</th>\n",
       "      <th>bp3</th>\n",
       "      <th>bs3</th>\n",
       "      <th>bp4</th>\n",
       "      <th>bs4</th>\n",
       "      <th>bp5</th>\n",
       "      <th>...</th>\n",
       "      <th>as19</th>\n",
       "      <th>ap20</th>\n",
       "      <th>as20</th>\n",
       "      <th>WAP</th>\n",
       "      <th>spread</th>\n",
       "      <th>log_returns</th>\n",
       "      <th>volatility_t+1</th>\n",
       "      <th>bid depth</th>\n",
       "      <th>ask depth</th>\n",
       "      <th>FDOFI</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>primary_key</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2022-05-19 00:00:15</th>\n",
       "      <td>10.265354</td>\n",
       "      <td>28720.075362</td>\n",
       "      <td>3.772080</td>\n",
       "      <td>28719.533333</td>\n",
       "      <td>1.405377</td>\n",
       "      <td>28719.011594</td>\n",
       "      <td>0.626551</td>\n",
       "      <td>28718.371014</td>\n",
       "      <td>0.505087</td>\n",
       "      <td>28717.980435</td>\n",
       "      <td>...</td>\n",
       "      <td>0.707326</td>\n",
       "      <td>28724.657971</td>\n",
       "      <td>1.104145</td>\n",
       "      <td>28720.131688</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.001254</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.837920</td>\n",
       "      <td>15.108906</td>\n",
       "      <td>0.356340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 00:00:45</th>\n",
       "      <td>10.266608</td>\n",
       "      <td>28755.960345</td>\n",
       "      <td>1.878828</td>\n",
       "      <td>28755.570690</td>\n",
       "      <td>0.335776</td>\n",
       "      <td>28755.050000</td>\n",
       "      <td>0.573517</td>\n",
       "      <td>28754.684483</td>\n",
       "      <td>0.737259</td>\n",
       "      <td>28754.279310</td>\n",
       "      <td>...</td>\n",
       "      <td>0.773897</td>\n",
       "      <td>28761.250000</td>\n",
       "      <td>0.854017</td>\n",
       "      <td>28756.179152</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>0.001254</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12.962207</td>\n",
       "      <td>23.700362</td>\n",
       "      <td>-0.292892</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 00:01:15</th>\n",
       "      <td>10.267532</td>\n",
       "      <td>28782.703333</td>\n",
       "      <td>1.771600</td>\n",
       "      <td>28782.131667</td>\n",
       "      <td>0.536867</td>\n",
       "      <td>28781.485000</td>\n",
       "      <td>0.402600</td>\n",
       "      <td>28781.140000</td>\n",
       "      <td>0.382467</td>\n",
       "      <td>28780.815000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.376617</td>\n",
       "      <td>28788.750000</td>\n",
       "      <td>0.425300</td>\n",
       "      <td>28782.777221</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000925</td>\n",
       "      <td>0.000165</td>\n",
       "      <td>11.303767</td>\n",
       "      <td>11.144067</td>\n",
       "      <td>0.007114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 00:01:45</th>\n",
       "      <td>10.267768</td>\n",
       "      <td>28789.470000</td>\n",
       "      <td>1.767750</td>\n",
       "      <td>28788.766667</td>\n",
       "      <td>0.343217</td>\n",
       "      <td>28788.278333</td>\n",
       "      <td>0.243650</td>\n",
       "      <td>28788.046667</td>\n",
       "      <td>0.325300</td>\n",
       "      <td>28787.726667</td>\n",
       "      <td>...</td>\n",
       "      <td>0.536067</td>\n",
       "      <td>28794.623333</td>\n",
       "      <td>0.395067</td>\n",
       "      <td>28789.546017</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000235</td>\n",
       "      <td>0.000425</td>\n",
       "      <td>10.402750</td>\n",
       "      <td>14.287550</td>\n",
       "      <td>-0.157341</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 00:02:15</th>\n",
       "      <td>10.267733</td>\n",
       "      <td>28788.382759</td>\n",
       "      <td>1.634448</td>\n",
       "      <td>28787.839655</td>\n",
       "      <td>0.342845</td>\n",
       "      <td>28787.370690</td>\n",
       "      <td>0.396414</td>\n",
       "      <td>28787.039655</td>\n",
       "      <td>0.232483</td>\n",
       "      <td>28786.703448</td>\n",
       "      <td>...</td>\n",
       "      <td>0.473741</td>\n",
       "      <td>28793.763793</td>\n",
       "      <td>0.417776</td>\n",
       "      <td>28788.543348</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>-0.000035</td>\n",
       "      <td>0.000517</td>\n",
       "      <td>9.357000</td>\n",
       "      <td>11.494069</td>\n",
       "      <td>-0.102492</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 88 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     log_price           bp1       bs1           bp2  \\\n",
       "primary_key                                                            \n",
       "2022-05-19 00:00:15  10.265354  28720.075362  3.772080  28719.533333   \n",
       "2022-05-19 00:00:45  10.266608  28755.960345  1.878828  28755.570690   \n",
       "2022-05-19 00:01:15  10.267532  28782.703333  1.771600  28782.131667   \n",
       "2022-05-19 00:01:45  10.267768  28789.470000  1.767750  28788.766667   \n",
       "2022-05-19 00:02:15  10.267733  28788.382759  1.634448  28787.839655   \n",
       "\n",
       "                          bs2           bp3       bs3           bp4       bs4  \\\n",
       "primary_key                                                                     \n",
       "2022-05-19 00:00:15  1.405377  28719.011594  0.626551  28718.371014  0.505087   \n",
       "2022-05-19 00:00:45  0.335776  28755.050000  0.573517  28754.684483  0.737259   \n",
       "2022-05-19 00:01:15  0.536867  28781.485000  0.402600  28781.140000  0.382467   \n",
       "2022-05-19 00:01:45  0.343217  28788.278333  0.243650  28788.046667  0.325300   \n",
       "2022-05-19 00:02:15  0.342845  28787.370690  0.396414  28787.039655  0.232483   \n",
       "\n",
       "                              bp5  ...      as19          ap20      as20  \\\n",
       "primary_key                        ...                                     \n",
       "2022-05-19 00:00:15  28717.980435  ...  0.707326  28724.657971  1.104145   \n",
       "2022-05-19 00:00:45  28754.279310  ...  0.773897  28761.250000  0.854017   \n",
       "2022-05-19 00:01:15  28780.815000  ...  0.376617  28788.750000  0.425300   \n",
       "2022-05-19 00:01:45  28787.726667  ...  0.536067  28794.623333  0.395067   \n",
       "2022-05-19 00:02:15  28786.703448  ...  0.473741  28793.763793  0.417776   \n",
       "\n",
       "                              WAP    spread  log_returns  volatility_t+1  \\\n",
       "primary_key                                                                \n",
       "2022-05-19 00:00:15  28720.131688  0.000015     0.001254        0.000000   \n",
       "2022-05-19 00:00:45  28756.179152  0.000010     0.001254        0.000000   \n",
       "2022-05-19 00:01:15  28782.777221  0.000005     0.000925        0.000165   \n",
       "2022-05-19 00:01:45  28789.546017  0.000005     0.000235        0.000425   \n",
       "2022-05-19 00:02:15  28788.543348  0.000008    -0.000035        0.000517   \n",
       "\n",
       "                     bid depth  ask depth     FDOFI  \n",
       "primary_key                                          \n",
       "2022-05-19 00:00:15  31.837920  15.108906  0.356340  \n",
       "2022-05-19 00:00:45  12.962207  23.700362 -0.292892  \n",
       "2022-05-19 00:01:15  11.303767  11.144067  0.007114  \n",
       "2022-05-19 00:01:45  10.402750  14.287550 -0.157341  \n",
       "2022-05-19 00:02:15   9.357000  11.494069 -0.102492  \n",
       "\n",
       "[5 rows x 88 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "89986011",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "80ec1f28",
   "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": [
    "data[['log_price']].boxplot();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "40d2cdeb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data[['WAP']].boxplot();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "6858aa27",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data[['spread']].boxplot();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "56dce47e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data[['log_returns']].boxplot();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "0899f07b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAj0AAAGdCAYAAAD5ZcJyAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAxtUlEQVR4nO3df3RU9Z3/8VcmyUwSQhIRyA9EiALCIgsKJo2K1CUQBbtEYfnRtGLKgrbSyqaWCisgW7dpESyLhxbZ7cFtVUAsQhsDkoafSogYoYr8KNqoPWiCiMkAgWQy8/n+wTd3GQmYQGRgPs/HORyYz33fO++bc27mxb2fOzfCGGMEAAAQ5lyhbgAAAOBSIPQAAAArEHoAAIAVCD0AAMAKhB4AAGAFQg8AALACoQcAAFiB0AMAAKwQFeoGLieBQECffPKJ2rdvr4iIiFC3AwAAWsAYo2PHjiktLU0u17nP5xB6zvDJJ5+oa9euoW4DAABcgL///e+65pprzrmc0HOG9u3bSzr9Q0tISAhxNwDaks/n04YNGzR8+HBFR0eHuh0Abcjr9apr167O5/i5EHrO0HRJKyEhgdADhBmfz6e4uDglJCQQeoAw9VVTU5jIDAAArEDoAQAAViD0AAAAKxB6AACAFQg9AADACoQeAABgBUIPAACwAqEHAABYgdADIOz5/X5t2bJFW7du1ZYtW+T3+0PdEoAQIPQACGurV6/W9ddfr2HDhunpp5/WsGHDdP3112v16tWhbg3AJUboARC2Vq9erdGjR+vw4cNB44cPH9bo0aMJPoBlCD0AwpLf79dDDz0kSRo6dKi2bdum5cuXa9u2bRo6dKgk6fvf/z6XugCLEHoAhKXNmzfrs88+0+233661a9cqMzNTsbGxyszM1Nq1a3X77bfr8OHD2rx5c6hbBXCJEHoAhKWmMDN37lw1NjZq0aJFWrp0qRYtWqTGxkbNmTMnqA5A+IsKdQMA8HV69tlnlZOTo8bGRklScXGxHnvsMd17770h7gzApRZhjDGhbuJy4fV6lZiYqNraWiUkJIS6HQAXobS0VNnZ2ZKkzp07Ky8vTydOnFC7du30wgsvOJOb//znPztzfABcmVr6+U3oOQOhBwgfJ0+eVFxcnCTJ7XaroaHBWXbm67q6OsXGxoakRwBto6Wf38zpARCWnn32WeffPp8vaFnTpa4v1wEIb4QeAGHp4MGDkqRu3bopMjIyaJnL5VK3bt2C6gCEPyYyAwhLERERkqSPPvpI99xzj4YNG6aDBw+qZ8+eKikpUVFRUVAdgPB3QWd6Fi9erO7duysmJkaZmZl68803z1u/atUq9e7dWzExMerXr5+Ki4uDlhtjNHv2bKWmpio2NlbZ2dln/e/r6NGjysvLU0JCgpKSkjRp0iQdP348qOa1117TN77xDbVv316dOnXS6NGj9eGHH17ILgK4wt1yyy2STs/fWb58uXw+n/72t7/J5/Np+fLlcrvdQXUAwl+rQ8/KlStVUFCgOXPm6O2331b//v2Vk5Nz1te8N9m+fbsmTJigSZMmadeuXcrNzVVubq727Nnj1MybN0+LFi3SkiVLVF5ernbt2iknJ0enTp1yavLy8vTee+85/0PbunWrpkyZ4iyvrKzUqFGj9E//9E/avXu3XnvtNR05ckT33Xdfa3cRQBj44osvJEkNDQ1q3769Hn30URUXF+vRRx9V+/btnYnMTXUALGBaKSMjwzz88MPOa7/fb9LS0kxhYWGz9WPHjjUjR44MGsvMzDQPPvigMcaYQCBgUlJSzFNPPeUsr6mpMR6PxyxfvtwYY8zevXuNJLNz506nZt26dSYiIsIcOnTIGGPMqlWrTFRUlPH7/U7NH//4RxMREWEaGhpatG+1tbVGkqmtrW1RPYDL1/PPP28kfeWf559/PtStArhILf38btWcnoaGBlVUVGjGjBnOmMvlUnZ2tsrKyppdp6ysTAUFBUFjOTk5WrNmjaTTZ2iqqqqc79OQpMTERGVmZqqsrEzjx49XWVmZkpKSNGjQIKcmOztbLpdL5eXluvfeezVw4EC5XC4tW7ZMDzzwgI4fP67f//73ys7OVnR0dLO91dfXq76+3nnt9Xolnb7T48t3ewC4snTo0MH5t8vlUiAQaPZ1hw4dON6BK1xLj+FWhZ4jR47I7/crOTk5aDw5OVn79+9vdp2qqqpm66uqqpzlTWPnq+ncuXNw41FR6tChg1OTnp6uDRs2aOzYsXrwwQfl9/uVlZV11vyhMxUWFmru3LlnjW/YsMH5fg8AV6ZXXnnF+XdUVFTQ9/Sc+XrlypVBywBceerq6lpUFzZ3b1VVVWny5MmaOHGiJkyYoGPHjmn27NkaM2aMSkpKmr1DY8aMGUFnobxer7p27arhw4fz5YTAFe63v/2t8++hQ4cqPT1d77//vnr06KHKykqtW7dO0umbJEaMGBGqNgG0gaYrNV+lVaGnY8eOioyMVHV1ddB4dXW1UlJSml0nJSXlvPVNf1dXVys1NTWoZsCAAU7NlydKNzY26ujRo876ixcvVmJioubNm+fUPP/88+ratavKy8v1jW9846zePB6PPB7PWePR0dHnvCQG4Mpw8uRJSdKgQYO0YcMG+f1+SafP5EZGRmrgwIGqqKjQyZMnOd6BK1xLj+FW3b3ldrs1cOBAlZaWOmOBQEClpaXKyspqdp2srKygekkqKSlx6tPT05WSkhJU4/V6VV5e7tRkZWWppqZGFRUVTs3GjRsVCASUmZkp6fSpLZcreHeavpDszGv5AOzQNAfwrbfecm5Pb+J2u53fJ2fOFQQQ5lo7Q3rFihXG4/GY5557zuzdu9dMmTLFJCUlmaqqKmOMMd/97nfNY4895tS/8cYbJioqysyfP9/s27fPzJkzx0RHR5t3333XqfnFL35hkpKSzNq1a80777xjRo0aZdLT083JkyedmrvuusvcdNNNpry83Lz++uumZ8+eZsKECc7y0tJSExERYebOnWv++te/moqKCpOTk2O6detm6urqWrRv3L0FhI8NGzY4d2i53W4zbtw4k5+fb8aNG2fcbrezbMOGDaFuFcBFaunnd6tDjzHGPPPMM+baa681brfbZGRkmB07djjLhgwZYiZOnBhU/9JLL5levXoZt9tt+vbta1599dWg5YFAwMyaNcskJycbj8djhg4dag4cOBBU8/nnn5sJEyaY+Ph4k5CQYPLz882xY8eCapYvX25uuukm065dO9OpUyfzz//8z2bfvn0t3i9CDxA+XnvttRbdsv7aa6+FulUAF6mln988Zf0MPGUdCB+zZs3Sk08++ZV1jz/+uH72s59dgo4AfF14yjoAq7V0Lh9z/gB7EHoAhKWkpKQ2rQNw5SP0AAhLLX2mFs/eAuxB6AEQlv72t7+1aR2AKx+hB0BY2rdvX5vWAbjyEXoAhKUPPvigTesAXPkIPQDCUksfQNjSOgBXPkIPgLDU0q8g46vKAHsQegAAgBUIPQAAwAqEHgAAYAVCDwAAsAKhBwAAWIHQAwAArEDoAQAAViD0AAAAKxB6AACAFQg9AADACoQeAABgBUIPAACwAqEHAABYgdADAACsQOgBAABWIPQAAAArEHoAAIAVCD0AAMAKhB4AAGAFQg8AALACoQcAAFiB0AMAAKxA6AEAAFYg9AAAACsQegAAgBUIPQAAwAqEHgAAYAVCDwAAsAKhBwAAWIHQAwAArEDoAQAAViD0AAAAKxB6AACAFQg9AADACoQeAABgBUIPAACwAqEHAABYgdADAACsQOgBAABWIPQAAAArEHoAAIAVCD0AAMAKhB4AAGCFqFA3AADnUldXp/3793/t7/P2229f0Hq9e/dWXFxcG3cD4OtC6AFw2dq/f78GDhz4tb/Phb5HRUWFbr755jbuBsDXhdAD4LLVu3dvVVRUXNC6I0aMUHV19VfWJScnq7i4+ILeo3fv3he0HoDQIPQAuGzFxcVd8JmUd999V507d25RXadOnS7oPQBcWZjIDCAsderUSYmJieetSUxMJPAAFiH0AAhbNTU15ww+iYmJqqmpubQNAQgpQg+AsFZTU6PDhw8r7ZquUnSM0q7pqsOHDxN4AAsRegCEvU6dOunV13epW8HLevX1XVzSAixF6AEAAFYg9AAAACsQegAAgBUIPQAAwAqEHgAAYAVCDwAAsAKhBwAAWIHQAwAArEDoAQAAViD0AAAAKxB6AACAFS4o9CxevFjdu3dXTEyMMjMz9eabb563ftWqVerdu7diYmLUr18/FRcXBy03xmj27NlKTU1VbGyssrOzdfDgwaCao0ePKi8vTwkJCUpKStKkSZN0/Pjxs7Yzf/589erVSx6PR126dNF//ud/XsguAgCAMNPq0LNy5UoVFBRozpw5evvtt9W/f3/l5OTo8OHDzdZv375dEyZM0KRJk7Rr1y7l5uYqNzdXe/bscWrmzZunRYsWacmSJSovL1e7du2Uk5OjU6dOOTV5eXl67733VFJSoqKiIm3dulVTpkwJeq9HHnlE//M//6P58+dr//79+uMf/6iMjIzW7iIAAAhHppUyMjLMww8/7Lz2+/0mLS3NFBYWNls/duxYM3LkyKCxzMxM8+CDDxpjjAkEAiYlJcU89dRTzvKamhrj8XjM8uXLjTHG7N2710gyO3fudGrWrVtnIiIizKFDh5yaqKgos3///tbukqO2ttZIMrW1tRe8DQCXp10fHjHdflpkdn14JNStAGhjLf38jmpNQGpoaFBFRYVmzJjhjLlcLmVnZ6usrKzZdcrKylRQUBA0lpOTozVr1kiSKisrVVVVpezsbGd5YmKiMjMzVVZWpvHjx6usrExJSUkaNGiQU5OdnS2Xy6Xy8nLde++9+tOf/qTrrrtORUVFuuuuu2SMUXZ2tubNm6cOHTo021t9fb3q6+ud116vV5Lk8/nk8/la86MBcJlrbGx0/ub4BsJLS4/pVoWeI0eOyO/3Kzk5OWg8OTlZ+/fvb3adqqqqZuurqqqc5U1j56vp3LlzcONRUerQoYNT87e//U0fffSRVq1apd/97nfy+/36t3/7N40ZM0YbN25strfCwkLNnTv3rPENGzYoLi6u2XUAXJn+flySorRjxw4d2vNV1QCuJHV1dS2qa1XouZwFAgHV19frd7/7nXr16iVJ+u1vf6uBAwfqwIEDuuGGG85aZ8aMGUFnobxer7p27arhw4crISHhkvUOhKMPPz+hE/X+ULfhqK+qld7dp849+qlbSmKo23G080Sq+9XtQt0GcEVrulLzVVoVejp27KjIyEhVV1cHjVdXVyslJaXZdVJSUs5b3/R3dXW1UlNTg2oGDBjg1Hx5onRjY6OOHj3qrJ+amqqoqCgn8EhSnz59JEkff/xxs6HH4/HI4/GcNR4dHa3o6Ohm9wfAV6s8ckLDFr4R6jaaNf2VfaFu4SybHv2m0jsSfIAL1dLP7FaFHrfbrYEDB6q0tFS5ubmSTp9hKS0t1dSpU5tdJysrS6WlpZo2bZozVlJSoqysLElSenq6UlJSVFpa6oQcr9er8vJyff/733e2UVNTo4qKCg0cOFCStHHjRgUCAWVmZkqSbrvtNjU2NuqDDz7Q9ddfL0n661//Kknq1q1ba3YTwEU6UX96/szCcQPUo3N8iLs57cTJehVtLtM938xSu9iz/7MTCu8fPq5pK3c7Py8AX69WX94qKCjQxIkTNWjQIGVkZGjhwoU6ceKE8vPzJUn333+/unTposLCQkmnbyMfMmSIFixYoJEjR2rFihV66623tHTpUklSRESEpk2bpieffFI9e/ZUenq6Zs2apbS0NCdY9enTR3fddZcmT56sJUuWyOfzaerUqRo/frzS0tIknZ7YfPPNN+t73/ueFi5cqEAgoIcffljDhg0LOvsD4NLp0TleN3a5PC4l+Xw+VXWSbu52FWdyAUu1OvSMGzdOn332mWbPnq2qqioNGDBA69evdyYif/zxx3K5/u/rf2699Va9+OKLevzxxzVz5kz17NlTa9as0Y033ujUTJ8+XSdOnNCUKVNUU1Oj22+/XevXr1dMTIxT88ILL2jq1KkaOnSoXC6XRo8erUWLFjnLXS6X/vSnP+mHP/yh7rjjDrVr10533323FixYcEE/GAAAEF4ijDEm1E1cLrxerxITE1VbW8tEZuAi7DlUq3ueeV1FP7z9sjrTU1xcrBEjRlw2Z3oux58TcCVq6ec3z94CAABWIPQAAAArhM339AC4fNT7T8kVc0iV3gNyxVwed281Njbqk8ZPtO/oPkVFXR6/+iq9x+WKOaR6/ylJXN4Cvm6Xx5EPIKx8cuIjtUt/RjPfDHUnZ/v1+l+HuoUg7dKlT04M0EAlf3UxgItC6AHQ5tLaddOJyh/qv8YN0PWXyff0NDY26o3X39Btt9922Zzp+eDwcT2ycrfS7uS7xIBL4fI48gGEFU9kjAKnuig94Qb9w9WXx2Ubn8+nyqhK9enQ57K5eytwqlaBU5/JExnz1cUALhoTmQEAgBUIPQAAwAqEHgAAYAXm9ABocyd9fkmnv3H4cnHiZL3e+kxK+eiLy+qBowAuHUIPgDb3wf//MH9s9bsh7uTLovT793eGuomztPPwqxi4FDjSALS54X1TJEnXd45XbHRkiLs57cCntfrxy+9qwZh+uiH18rijTDodeNI7tgt1G4AVCD0A2lyHdm6Nz7g21G0EaWxslCRd36kdD/cELMVEZgAAYAVCDwAAsAKhBwAAWIHQAwAArEDoAQAAViD0AAAAKxB6AACAFQg9AADACoQeAABgBUIPAACwAqEHAABYgdADAACsQOgBAABWIPQAAAArEHoAAIAVCD0AAMAKhB4AAGAFQg8AALACoQcAAFghKtQNAMC51NXVaf/+/W2yrQOf1qi+6n3t2xOrwOdJbbLN3r17Ky4urk22BeDrR+gBcNnav3+/Bg4c2Kbb/Pb/tt22KioqdPPNN7fdBgF8rQg9AC5bvXv3VkVFRZts6/jJer26qUwj78xSfKynTbbZu3fvNtkOgEuD0APgshUXF9dmZ1J8Pp++OHJYWRmDFB0d3SbbBHBlYSIzAACwAqEHAABYgdADAACsQOgBAABWIPQAAAArEHoAAIAVCD0AAMAKhB4AAGAFQg8AALACoQcAAFiB0AMAAKxA6AEAAFYg9AAAACsQegAAgBUIPQAAwAqEHgAAYAVCDwAAsAKhBwAAWIHQAwAArEDoAQAAViD0AAAAKxB6AACAFQg9AADACoQeAGHP7/dry5Yt2rp1q7Zs2SK/3x/qlgCEAKEHQFhbvXq1evTooWHDhunpp5/WsGHD1KNHD61evTrUrQG4xAg9AMLW6tWrNWbMGPXr10/btm3T8uXLtW3bNvXr109jxowh+ACWiTDGmFA3cbnwer1KTExUbW2tEhISQt0OgIvg9/vVo0cP9evXT2vWrJHf71dxcbFGjBihyMhI5ebmas+ePTp48KAiIyND3S6Ai9DSz2/O9AAIS9u2bdOHH36omTNnyuUK/lXncrk0Y8YMVVZWatu2bSHqEMClRugBEJY+/fRTSdKNN97Y7PKm8aY6AOGP0AMgLKWmpkqS9uzZ0+zypvGmOgDhj9ADICwNHjxY3bt3189//nMFAoGgZYFAQIWFhUpPT9fgwYND1CGAS43QAyAsRUZGasGCBSoqKlJubq527NihkydPaseOHcrNzVVRUZHmz5/PJGbAIlGhbgAAvi733XefXn75Zf34xz/WHXfc4Yynp6fr5Zdf1n333RfC7gBcatyyfgZuWQfCk9/v16ZNm7Ru3TrdfffduvPOOznDA4SRr/WW9cWLF6t79+6KiYlRZmam3nzzzfPWr1q1Sr1791ZMTIz69eun4uLioOXGGM2ePVupqamKjY1Vdna2Dh48GFRz9OhR5eXlKSEhQUlJSZo0aZKOHz/e7Pu9//77at++vZKSki5k9wCEmcjISA0ZMkR33HGHhgwZQuABLNXq0LNy5UoVFBRozpw5evvtt9W/f3/l5OTo8OHDzdZv375dEyZM0KRJk7Rr1y7l5uY6XwrWZN68eVq0aJGWLFmi8vJytWvXTjk5OTp16pRTk5eXp/fee08lJSUqKirS1q1bNWXKlLPez+fzacKECUxOBAAAwUwrZWRkmIcffth57ff7TVpamiksLGy2fuzYsWbkyJFBY5mZmebBBx80xhgTCARMSkqKeeqpp5zlNTU1xuPxmOXLlxtjjNm7d6+RZHbu3OnUrFu3zkRERJhDhw4FbXv69OnmO9/5jlm2bJlJTExs1b7V1tYaSaa2trZV6wG4/DU0NJg1a9aYhoaGULcCoI219PO7VWd6GhoaVFFRoezsbGfM5XIpOztbZWVlza5TVlYWVC9JOTk5Tn1lZaWqqqqCahITE5WZmenUlJWVKSkpSYMGDXJqsrOz5XK5VF5e7oxt3LhRq1at0uLFi1uzWwAAwAKtunvryJEj8vv9Sk5ODhpPTk7W/v37m12nqqqq2fqqqipnedPY+Wo6d+4c3HhUlDp06ODUfP7553rggQf0/PPPt3gScn19verr653XXq9X0ulLZD6fr0XbAHBlaDqmObaB8NPS4zpsblmfPHmyvv3tbwfdlvpVCgsLNXfu3LPGN2zYoLi4uLZsD8BloqSkJNQtAGhjdXV1LaprVejp2LGjIiMjVV1dHTReXV2tlJSUZtdJSUk5b33T39XV1UFfB19dXa0BAwY4NV+eKN3Y2KijR48662/cuFF//OMfNX/+fEmn7wgLBAKKiorS0qVL9b3vfe+s3mbMmKGCggLntdfrVdeuXTV8+HBuWQfCjM/nU0lJiYYNG6bo6OhQtwOgDTVdqfkqrQo9brdbAwcOVGlpqXJzcyWd/jr30tJSTZ06tdl1srKyVFpaqmnTpjljJSUlysrKknT6S8JSUlJUWlrqhByv16vy8nJ9//vfd7ZRU1OjiooKDRw4UNLpkBMIBJSZmSnp9Lwfv9/vvMfatWv1y1/+Utu3b1eXLl2a7c3j8cjj8Zw1Hh0dzS9FIExxfAPhp6XHdKsvbxUUFGjixIkaNGiQMjIytHDhQp04cUL5+fmSpPvvv19dunRRYWGhJOmRRx7RkCFDtGDBAo0cOVIrVqzQW2+9paVLl0qSIiIiNG3aND355JPq2bOn0tPTNWvWLKWlpTnBqk+fPrrrrrs0efJkLVmyRD6fT1OnTtX48eOVlpbm1JzprbfeksvlOucTlgEAgF1aHXrGjRunzz77TLNnz1ZVVZUGDBig9evXOxORP/74Y7lc/3dT2K233qoXX3xRjz/+uGbOnKmePXtqzZo1QWFk+vTpOnHihKZMmaKamhrdfvvtWr9+vWJiYpyaF154QVOnTtXQoUPlcrk0evRoLVq06GL2HQAAWITHUJyBx1AA4cvn86m4uFgjRozg8hYQZr7Wx1AAAABcaQg9AADACoQeAABgBUIPAACwAqEHAABYgdADAACsQOgBAABWIPQAAAArEHoAAIAVCD0AAMAKhB4AAGAFQg8AALACoQcAAFiB0AMAAKxA6AEAAFYg9AAAACsQegAAgBUIPQAAwAqEHgAAYAVCDwAAsAKhBwAAWIHQAwAArEDoAQAAViD0AAAAKxB6AACAFQg9AADACoQeAABgBUIPAACwAqEHAABYgdADAACsQOgBAABWIPQAAAArEHoAAIAVCD0AAMAKhB4AAGAFQg8AALACoQcAAFiB0AMAAKxA6AEAAFYg9AAAACsQegAAgBUIPQAAwAqEHgAAYAVCDwAAsAKhBwAAWIHQAwAArEDoAQAAViD0AAAAKxB6AACAFQg9AADACoQeAABgBUIPAACwAqEHAABYgdADAACsQOgBAABWIPQAAAArEHoAAIAVCD0AAMAKhB4AAGAFQg8AALACoQcAAFiB0AMAAKxA6AEAAFYg9AAAACsQegAAgBUIPQAAwAqEHgAAYAVCDwAAsAKhBwAAWIHQAwAArEDoAQAAVrig0LN48WJ1795dMTExyszM1Jtvvnne+lWrVql3796KiYlRv379VFxcHLTcGKPZs2crNTVVsbGxys7O1sGDB4Nqjh49qry8PCUkJCgpKUmTJk3S8ePHneWbN2/WqFGjlJqaqnbt2mnAgAF64YUXLmT3AABAGGp16Fm5cqUKCgo0Z84cvf322+rfv79ycnJ0+PDhZuu3b9+uCRMmaNKkSdq1a5dyc3OVm5urPXv2ODXz5s3TokWLtGTJEpWXl6tdu3bKycnRqVOnnJq8vDy99957KikpUVFRkbZu3aopU6YEvc8//uM/6g9/+IPeeecd5efn6/7771dRUVFrdxEAAIQj00oZGRnm4Ycfdl77/X6TlpZmCgsLm60fO3asGTlyZNBYZmamefDBB40xxgQCAZOSkmKeeuopZ3lNTY3xeDxm+fLlxhhj9u7daySZnTt3OjXr1q0zERER5tChQ+fsdcSIESY/P7/F+1ZbW2skmdra2havA+DK0NDQYNasWWMaGhpC3QqANtbSz++o1gSkhoYGVVRUaMaMGc6Yy+VSdna2ysrKml2nrKxMBQUFQWM5OTlas2aNJKmyslJVVVXKzs52licmJiozM1NlZWUaP368ysrKlJSUpEGDBjk12dnZcrlcKi8v17333tvse9fW1qpPnz7n3J/6+nrV19c7r71eryTJ5/PJ5/Odcz0AV56mY5pjGwg/LT2uWxV6jhw5Ir/fr+Tk5KDx5ORk7d+/v9l1qqqqmq2vqqpyljeNna+mc+fOwY1HRalDhw5OzZe99NJL2rlzp5599tlz7k9hYaHmzp171viGDRsUFxd3zvUAXLlKSkpC3QKANlZXV9eiulaFnivFpk2blJ+fr//+7/9W3759z1k3Y8aMoLNQXq9XXbt21fDhw5WQkHApWgVwifh8PpWUlGjYsGGKjo4OdTsA2lDTlZqv0qrQ07FjR0VGRqq6ujpovLq6WikpKc2uk5KSct76pr+rq6uVmpoaVDNgwACn5ssTpRsbG3X06NGz3nfLli361re+pV/96le6//77z7s/Ho9HHo/nrPHo6Gh+KQJhiuMbCD8tPaZbdfeW2+3WwIEDVVpa6owFAgGVlpYqKyur2XWysrKC6qXTp5eb6tPT05WSkhJU4/V6VV5e7tRkZWWppqZGFRUVTs3GjRsVCASUmZnpjG3evFkjR47UL3/5y6A7uwAAAFp9eaugoEATJ07UoEGDlJGRoYULF+rEiRPKz8+XJN1///3q0qWLCgsLJUmPPPKIhgwZogULFmjkyJFasWKF3nrrLS1dulSSFBERoWnTpunJJ59Uz549lZ6erlmzZiktLU25ubmSpD59+uiuu+7S5MmTtWTJEvl8Pk2dOlXjx49XWlqapNOXtO655x498sgjGj16tDPXx+12q0OHDhf9gwIAAFe4C7k17JlnnjHXXnutcbvdJiMjw+zYscNZNmTIEDNx4sSg+pdeesn06tXLuN1u07dvX/Pqq68GLQ8EAmbWrFkmOTnZeDweM3ToUHPgwIGgms8//9xMmDDBxMfHm4SEBJOfn2+OHTvmLJ84caKRdNafIUOGtHi/uGUdCF/csg6Er5Z+fkcYY0wIM9dlxev1KjExUbW1tUxkBsKMz+dTcXGxRowYwZweIMy09PObZ28BAAArEHoAAIAVCD0Awp7f79eWLVu0detWbdmyRX6/P9QtAQgBQg+AsLZ69Wr16NFDw4YN09NPP61hw4apR48eWr16dahbA3CJEXoAhK3Vq1drzJgx6tevn7Zt26bly5dr27Zt6tevn8aMGUPwASzD3Vtn4O4tIHz4/X716NFD/fr105o1a+T3+527tyIjI5Wbm6s9e/bo4MGDioyMDHW7AC4Cd28BsNq2bdv04YcfaubMmXK5gn/VuVwuzZgxQ5WVldq2bVuIOgRwqRF6AISlTz/9VJJ04403Nru8abypDkD4I/QACEtNDzDes2dPs8ubxs980DGA8EboARCWBg8erO7du+vnP/+5AoFA0LJAIKDCwkKlp6dr8ODBIeoQwKVG6AEQliIjI7VgwQIVFRUpNzdXO3bs0MmTJ7Vjxw7l5uaqqKhI8+fPZxIzYJFWP2UdAK4U9913n15++WX9+Mc/1h133OGMp6en6+WXX9Z9990Xwu4AXGrcsn4GblkHwpPf79emTZu0bt063X333brzzjs5wwOEkZZ+fnOmB0DYi4yM1JAhQ3TixAkNGTKEwANYijk9AADACoQeAABgBUIPAACwAqEHAABYgdADAACsQOgBAABWIPQAAAArEHoAAIAVCD0AAMAKhB4AAGAFQg8AALACoQcAAFiB0AMAAKxA6AEAAFYg9AAAACsQegAAgBUIPQAAwAqEHgAAYAVCDwAAsAKhBwAAWIHQAwAArEDoAQAAViD0AAAAKxB6AACAFQg9AADACoQeAABgBUIPAACwAqEHAABYgdADAACsQOgBAABWIPQAAAArEHoAAIAVCD0AAMAKhB4AAGAFQg8AALACoQcAAFiB0AMAAKxA6AEAAFYg9AAAACsQegAAgBUIPQAAwAqEHgAAYAVCDwAAsAKhBwAAWIHQAwAArEDoAQAAViD0AAAAKxB6AACAFQg9AADACoQeAABgBUIPAACwAqEHAABYgdADAACsQOgBAABWIPQAAAArEHoAhL2TJ0/qRz/6kZ544gn96Ec/0smTJ0PdEoAQuKDQs3jxYnXv3l0xMTHKzMzUm2++ed76VatWqXfv3oqJiVG/fv1UXFwctNwYo9mzZys1NVWxsbHKzs7WwYMHg2qOHj2qvLw8JSQkKCkpSZMmTdLx48eDat555x0NHjxYMTEx6tq1q+bNm3chuwcgjOTm5iouLk5LlizR7t27tWTJEsXFxSk3NzfUrQG4xFodelauXKmCggLNmTNHb7/9tvr376+cnBwdPny42frt27drwoQJmjRpknbt2qXc3Fzl5uZqz549Ts28efO0aNEiLVmyROXl5WrXrp1ycnJ06tQppyYvL0/vvfeeSkpKVFRUpK1bt2rKlCnOcq/Xq+HDh6tbt26qqKjQU089pSeeeEJLly5t7S4CCBO5ublau3at3G63pk+frt/85jeaPn263G631q5dS/ABbGNaKSMjwzz88MPOa7/fb9LS0kxhYWGz9WPHjjUjR44MGsvMzDQPPvigMcaYQCBgUlJSzFNPPeUsr6mpMR6PxyxfvtwYY8zevXuNJLNz506nZt26dSYiIsIcOnTIGGPMr3/9a3PVVVeZ+vp6p+anP/2pueGGG1q8b7W1tUaSqa2tbfE6AC5PdXV1RpJxu92mvr7eNDQ0mDVr1piGhgZTX19v3G63kWTq6upC3SqAi9TSz++o1gSkhoYGVVRUaMaMGc6Yy+VSdna2ysrKml2nrKxMBQUFQWM5OTlas2aNJKmyslJVVVXKzs52licmJiozM1NlZWUaP368ysrKlJSUpEGDBjk12dnZcrlcKi8v17333quysjLdcccdcrvdQe/zy1/+Ul988YWuuuqqs3qrr69XfX2989rr9UqSfD6ffD5fK34yAC43Tb93pk2bpoiICOeY9vl8io6O1o9+9CPNnz9fBQUFWrRoUShbBXCRWvqZ3arQc+TIEfn9fiUnJweNJycna//+/c2uU1VV1Wx9VVWVs7xp7Hw1nTt3Dm48KkodOnQIqklPTz9rG03Lmgs9hYWFmjt37lnjGzZsUFxcXLP7A+DKsGPHDklSenp60DzCkpISZ7yp7svzDAFcWerq6lpU16rQE25mzJgRdBbK6/Wqa9euGj58uBISEkLYGYCLtX79eu3evVuVlZWaNGmSfD6fSkpKNGzYMEVHR2vmzJmSpG984xsaMWJEiLsFcDGartR8lVaFno4dOyoyMlLV1dVB49XV1UpJSWl2nZSUlPPWN/1dXV2t1NTUoJoBAwY4NV+eKN3Y2KijR48Gbae59znzPb7M4/HI4/GcNR4dHa3o6Ohm1wFwZXj66ae1ZMkSLVy4UD/72c+cYzo6OlrGGOeS1tNPP83xDlzhWnoMt+ruLbfbrYEDB6q0tNQZCwQCKi0tVVZWVrPrZGVlBdVLp08vN9Wnp6crJSUlqMbr9aq8vNypycrKUk1NjSoqKpyajRs3KhAIKDMz06nZunVr0HW9kpIS3XDDDc1e2gIQ3mJjYzVq1Cg1NDSoffv2mjlzpg4dOqSZM2eqffv2amho0KhRoxQbGxvqVgFcKq2dIb1ixQrj8XjMc889Z/bu3WumTJlikpKSTFVVlTHGmO9+97vmsccec+rfeOMNExUVZebPn2/27dtn5syZY6Kjo827777r1PziF78wSUlJZu3ateadd94xo0aNMunp6ebkyZNOzV133WVuuukmU15ebl5//XXTs2dPM2HCBGd5TU2NSU5ONt/97nfNnj17zIoVK0xcXJx59tlnW7xv3L0FhJ9Ro0YZSWf9GTVqVKhbA9BGWvr53erQY4wxzzzzjLn22muN2+02GRkZZseOHc6yIUOGmIkTJwbVv/TSS6ZXr17G7Xabvn37mldffTVoeSAQMLNmzTLJycnG4/GYoUOHmgMHDgTVfP7552bChAkmPj7eJCQkmPz8fHPs2LGgmr/85S/m9ttvNx6Px3Tp0sX84he/aNV+EXqA8FRXV2ceeughM2DAAPPQQw9xmzoQZlr6+R1hjDEhO810mfF6vUpMTFRtbS0TmYEw4/P5VFxcrBEjRjCHBwgzLf385tlbAADACoQeAABgBUIPAACwAqEHAABYgdADAACsQOgBAABWIPQAAAArEHoAAIAVCD0AAMAKrXrKerhr+nLqlj6iHsCVw+fzqa6uTl6vl29kBsJM0+f2Vz1kgtBzhmPHjkmSunbtGuJOAABAax07dkyJiYnnXM6zt84QCAT0ySefqH379oqIiAh1OwDakNfrVdeuXfX3v/+dZ+sBYcYYo2PHjiktLU0u17ln7hB6AFiBBwoDYCIzAACwAqEHAABYgdADwAoej0dz5syRx+MJdSsAQoQ5PQAAwAqc6QEAAFYg9AAAACsQegAAgBUIPQBCIiIiQmvWrGnT7Xz44YeKiIjQ7t27JUmbN29WRESEampqLvp9AFz5CD0ArghPPPGEBgwYcNb4p59+qrvvvrvZdW699VZ9+umnztfSP/fcc0pKSmrTvh544AHl5uZe1DY+/fRTffvb31avXr3kcrk0bdq0NukNQDBCD4ArWkpKyjlvQ3e73UpJSQn5Y2WazkCdS319vTp16qTHH39c/fv3v4SdAXYh9ABotaVLlyotLU2BQCBofNSoUfre974nSfrNb36j66+/Xm63WzfccIN+//vfn3ebP/3pT9WrVy/FxcXpuuuu06xZs+Tz+SSdPkMzd+5c/eUvf1FERIQiIiL03HPPSTr/ZbIzL29t3rxZ+fn5qq2tdbbxxBNP6D/+4z904403nrXugAEDNGvWrPP2/MQTT+h///d/tXbtWmebmzdvPu86zenevbv+67/+S/fff/95H5YI4OLwlHUArfYv//Iv+uEPf6hNmzZp6NChkqSjR49q/fr1Ki4u1iuvvKJHHnlECxcuVHZ2toqKipSfn69rrrlGd955Z7PbbN++vZ577jmlpaXp3Xff1eTJk9W+fXtNnz5d48aN0549e7R+/Xr9+c9/lqRWh4Nbb71VCxcu1OzZs3XgwAFJUnx8vGpqajR37lzt3LlTt9xyiyRp165deuedd7R69erzbvPRRx/Vvn375PV6tWzZMklShw4dWtUXgEuH0AOg1a666irdfffdevHFF53Q8/LLL6tjx4668847NXjwYD3wwAP6wQ9+IEkqKCjQjh07NH/+/HOGnscff9z5d/fu3fXoo49qxYoVmj59umJjYxUfH6+oqCilpKRcUM9ut1uJiYmKiIgI2kZ8fLxycnK0bNkyJ/QsW7ZMQ4YM0XXXXXfebcbHxys2Nlb19fUX3BeAS4fLWwAuSF5env7whz+ovr5ekvTCCy9o/Pjxcrlc2rdvn2677bag+ttuu0379u075/ZWrlyp2267TSkpKYqPj9fjjz+ujz/++GvdhyaTJ0/W8uXLderUKTU0NOjFF190LtNdqL59+yo+Pl7x8fHq27evJDmv4+Pjzzn5GsDXhzM9AC7It771LRlj9Oqrr+qWW27Rtm3b9Ktf/eqCtlVWVqa8vDzNnTtXOTk5SkxM1IoVK7RgwYI27rp53/rWt+TxePTKK6/I7XbL5/NpzJgxF7XN4uJiZ07SoUOH9M1vftO5lV6SYmNjL2r7AFqP0APggsTExOi+++7TCy+8oPfff1833HCDbr75ZklSnz599MYbb2jixIlO/RtvvKF/+Id/aHZb27dvV7du3fTv//7vzthHH30UVON2u+X3+y+q53NtIyoqShMnTtSyZcvkdrs1fvz4FoeSc22zW7duQduXpB49elxg5wDaAqEHwAXLy8vTPffco/fee0/f+c53nPGf/OQnGjt2rG666SZlZ2frT3/6k1avXu1MQv6ynj176uOPP9aKFSt0yy236NVXX9Urr7wSVNO9e3dVVlZq9+7duuaaa9S+fftWPzG9e/fuOn78uEpLS9W/f3/FxcUpLi5OkvSv//qv6tOnj6TTAa0123zttdd04MABXX311UpMTFR0dHSr+pLknAU6fvy4PvvsM+3evVtut/ucQRHABTAAcIH8fr9JTU01kswHH3wQtOzXv/61ue6660x0dLTp1auX+d3vfhe0XJJ55ZVXnNc/+clPzNVXX23i4+PNuHHjzK9+9SuTmJjoLD916pQZPXq0SUpKMpLMsmXLztpOZWWlkWR27dpljDFm06ZNRpL54osvnO089NBD5uqrrzaSzJw5c4J6Gjx4sOnbt2+rfgaHDx82w4YNM/Hx8UaS2bRp01k1TX2dj6Sz/nTr1q1VvQA4vwhjjAlF2AKAy4kxRj179tQPfvADFRQUhLodAF8DLm8BsN5nn32mFStWqKqqSvn5+aFuB8DXhNADwHqdO3dWx44dtXTpUl111VVBy+Lj48+53rp16zR48OCvuz0AbYTLWwBwHu+///45l3Xp0oVbz4ErCKEHAABYgW9kBgAAViD0AAAAKxB6AACAFQg9AADACoQeAABgBUIPAACwAqEHAABYgdADAACs8P8AZAoBAXqALHUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data[['volatility_t+1']].boxplot();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5281224a",
   "metadata": {},
   "source": [
    "# Scaling the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "c7b9c95c",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "05553239",
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = MinMaxScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "b905e05a",
   "metadata": {},
   "outputs": [],
   "source": [
    "data[data.columns] = scaler.fit_transform(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "84467ef2",
   "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>log_price</th>\n",
       "      <th>bp1</th>\n",
       "      <th>bs1</th>\n",
       "      <th>bp2</th>\n",
       "      <th>bs2</th>\n",
       "      <th>bp3</th>\n",
       "      <th>bs3</th>\n",
       "      <th>bp4</th>\n",
       "      <th>bs4</th>\n",
       "      <th>bp5</th>\n",
       "      <th>...</th>\n",
       "      <th>as19</th>\n",
       "      <th>ap20</th>\n",
       "      <th>as20</th>\n",
       "      <th>WAP</th>\n",
       "      <th>spread</th>\n",
       "      <th>log_returns</th>\n",
       "      <th>volatility_t+1</th>\n",
       "      <th>bid depth</th>\n",
       "      <th>ask depth</th>\n",
       "      <th>FDOFI</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>primary_key</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2022-05-19 00:00:15</th>\n",
       "      <td>0.022792</td>\n",
       "      <td>0.022122</td>\n",
       "      <td>0.017682</td>\n",
       "      <td>0.022097</td>\n",
       "      <td>0.010072</td>\n",
       "      <td>0.022076</td>\n",
       "      <td>0.016586</td>\n",
       "      <td>0.021914</td>\n",
       "      <td>0.014328</td>\n",
       "      <td>0.021840</td>\n",
       "      <td>...</td>\n",
       "      <td>0.077064</td>\n",
       "      <td>0.022187</td>\n",
       "      <td>0.169244</td>\n",
       "      <td>0.022108</td>\n",
       "      <td>0.064543</td>\n",
       "      <td>0.381071</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.073940</td>\n",
       "      <td>0.149910</td>\n",
       "      <td>0.639872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 00:00:45</th>\n",
       "      <td>0.043030</td>\n",
       "      <td>0.041692</td>\n",
       "      <td>0.007254</td>\n",
       "      <td>0.041750</td>\n",
       "      <td>0.002395</td>\n",
       "      <td>0.041730</td>\n",
       "      <td>0.015180</td>\n",
       "      <td>0.041717</td>\n",
       "      <td>0.020940</td>\n",
       "      <td>0.041636</td>\n",
       "      <td>...</td>\n",
       "      <td>0.085127</td>\n",
       "      <td>0.042124</td>\n",
       "      <td>0.127689</td>\n",
       "      <td>0.041766</td>\n",
       "      <td>0.038022</td>\n",
       "      <td>0.381097</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.024647</td>\n",
       "      <td>0.256198</td>\n",
       "      <td>0.267512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 00:01:15</th>\n",
       "      <td>0.057948</td>\n",
       "      <td>0.056277</td>\n",
       "      <td>0.006663</td>\n",
       "      <td>0.056235</td>\n",
       "      <td>0.003838</td>\n",
       "      <td>0.056147</td>\n",
       "      <td>0.010647</td>\n",
       "      <td>0.056145</td>\n",
       "      <td>0.010836</td>\n",
       "      <td>0.056107</td>\n",
       "      <td>...</td>\n",
       "      <td>0.037006</td>\n",
       "      <td>0.057108</td>\n",
       "      <td>0.056464</td>\n",
       "      <td>0.056272</td>\n",
       "      <td>0.011147</td>\n",
       "      <td>0.355279</td>\n",
       "      <td>0.199778</td>\n",
       "      <td>0.020316</td>\n",
       "      <td>0.100860</td>\n",
       "      <td>0.439577</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 00:01:45</th>\n",
       "      <td>0.061742</td>\n",
       "      <td>0.059967</td>\n",
       "      <td>0.006642</td>\n",
       "      <td>0.059853</td>\n",
       "      <td>0.002448</td>\n",
       "      <td>0.059852</td>\n",
       "      <td>0.006432</td>\n",
       "      <td>0.059911</td>\n",
       "      <td>0.009208</td>\n",
       "      <td>0.059876</td>\n",
       "      <td>...</td>\n",
       "      <td>0.056320</td>\n",
       "      <td>0.060308</td>\n",
       "      <td>0.051441</td>\n",
       "      <td>0.059963</td>\n",
       "      <td>0.011140</td>\n",
       "      <td>0.301312</td>\n",
       "      <td>0.514418</td>\n",
       "      <td>0.017963</td>\n",
       "      <td>0.139749</td>\n",
       "      <td>0.345256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 00:02:15</th>\n",
       "      <td>0.061180</td>\n",
       "      <td>0.059374</td>\n",
       "      <td>0.005907</td>\n",
       "      <td>0.059348</td>\n",
       "      <td>0.002445</td>\n",
       "      <td>0.059357</td>\n",
       "      <td>0.010483</td>\n",
       "      <td>0.059362</td>\n",
       "      <td>0.006564</td>\n",
       "      <td>0.059318</td>\n",
       "      <td>...</td>\n",
       "      <td>0.048770</td>\n",
       "      <td>0.059840</td>\n",
       "      <td>0.055214</td>\n",
       "      <td>0.059416</td>\n",
       "      <td>0.023907</td>\n",
       "      <td>0.280179</td>\n",
       "      <td>0.626419</td>\n",
       "      <td>0.015232</td>\n",
       "      <td>0.105190</td>\n",
       "      <td>0.376714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 23:57:45</th>\n",
       "      <td>0.907378</td>\n",
       "      <td>0.904823</td>\n",
       "      <td>0.018769</td>\n",
       "      <td>0.904834</td>\n",
       "      <td>0.002965</td>\n",
       "      <td>0.904897</td>\n",
       "      <td>0.005247</td>\n",
       "      <td>0.904900</td>\n",
       "      <td>0.009869</td>\n",
       "      <td>0.904878</td>\n",
       "      <td>...</td>\n",
       "      <td>0.122925</td>\n",
       "      <td>0.903624</td>\n",
       "      <td>0.105144</td>\n",
       "      <td>0.904752</td>\n",
       "      <td>0.000064</td>\n",
       "      <td>0.314677</td>\n",
       "      <td>0.676196</td>\n",
       "      <td>0.017869</td>\n",
       "      <td>0.104124</td>\n",
       "      <td>0.408075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 23:58:15</th>\n",
       "      <td>0.901098</td>\n",
       "      <td>0.898348</td>\n",
       "      <td>0.004719</td>\n",
       "      <td>0.898378</td>\n",
       "      <td>0.003116</td>\n",
       "      <td>0.898324</td>\n",
       "      <td>0.007901</td>\n",
       "      <td>0.898278</td>\n",
       "      <td>0.001949</td>\n",
       "      <td>0.898280</td>\n",
       "      <td>...</td>\n",
       "      <td>0.093584</td>\n",
       "      <td>0.897528</td>\n",
       "      <td>0.206182</td>\n",
       "      <td>0.898314</td>\n",
       "      <td>0.000071</td>\n",
       "      <td>0.252438</td>\n",
       "      <td>0.676141</td>\n",
       "      <td>0.017467</td>\n",
       "      <td>0.151608</td>\n",
       "      <td>0.322110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 23:58:45</th>\n",
       "      <td>0.897040</td>\n",
       "      <td>0.894213</td>\n",
       "      <td>0.011522</td>\n",
       "      <td>0.894144</td>\n",
       "      <td>0.004440</td>\n",
       "      <td>0.894151</td>\n",
       "      <td>0.005782</td>\n",
       "      <td>0.894146</td>\n",
       "      <td>0.009229</td>\n",
       "      <td>0.894110</td>\n",
       "      <td>...</td>\n",
       "      <td>0.088406</td>\n",
       "      <td>0.893198</td>\n",
       "      <td>0.097990</td>\n",
       "      <td>0.894154</td>\n",
       "      <td>0.000076</td>\n",
       "      <td>0.263214</td>\n",
       "      <td>0.676051</td>\n",
       "      <td>0.017791</td>\n",
       "      <td>0.111075</td>\n",
       "      <td>0.393519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 23:59:15</th>\n",
       "      <td>0.897421</td>\n",
       "      <td>0.894584</td>\n",
       "      <td>0.013036</td>\n",
       "      <td>0.894569</td>\n",
       "      <td>0.001802</td>\n",
       "      <td>0.894742</td>\n",
       "      <td>0.031284</td>\n",
       "      <td>0.894712</td>\n",
       "      <td>0.048701</td>\n",
       "      <td>0.894641</td>\n",
       "      <td>...</td>\n",
       "      <td>0.033136</td>\n",
       "      <td>0.893706</td>\n",
       "      <td>0.063823</td>\n",
       "      <td>0.894545</td>\n",
       "      <td>0.000075</td>\n",
       "      <td>0.284753</td>\n",
       "      <td>0.675934</td>\n",
       "      <td>0.027510</td>\n",
       "      <td>0.123202</td>\n",
       "      <td>0.459037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-05-19 23:59:45</th>\n",
       "      <td>0.897796</td>\n",
       "      <td>0.894965</td>\n",
       "      <td>0.012252</td>\n",
       "      <td>0.894923</td>\n",
       "      <td>0.002548</td>\n",
       "      <td>0.895127</td>\n",
       "      <td>0.008091</td>\n",
       "      <td>0.895181</td>\n",
       "      <td>0.021924</td>\n",
       "      <td>0.895162</td>\n",
       "      <td>...</td>\n",
       "      <td>0.056774</td>\n",
       "      <td>0.893793</td>\n",
       "      <td>0.088835</td>\n",
       "      <td>0.894930</td>\n",
       "      <td>0.000075</td>\n",
       "      <td>0.284728</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.021191</td>\n",
       "      <td>0.215764</td>\n",
       "      <td>0.278242</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2880 rows × 88 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     log_price       bp1       bs1       bp2       bs2  \\\n",
       "primary_key                                                              \n",
       "2022-05-19 00:00:15   0.022792  0.022122  0.017682  0.022097  0.010072   \n",
       "2022-05-19 00:00:45   0.043030  0.041692  0.007254  0.041750  0.002395   \n",
       "2022-05-19 00:01:15   0.057948  0.056277  0.006663  0.056235  0.003838   \n",
       "2022-05-19 00:01:45   0.061742  0.059967  0.006642  0.059853  0.002448   \n",
       "2022-05-19 00:02:15   0.061180  0.059374  0.005907  0.059348  0.002445   \n",
       "...                        ...       ...       ...       ...       ...   \n",
       "2022-05-19 23:57:45   0.907378  0.904823  0.018769  0.904834  0.002965   \n",
       "2022-05-19 23:58:15   0.901098  0.898348  0.004719  0.898378  0.003116   \n",
       "2022-05-19 23:58:45   0.897040  0.894213  0.011522  0.894144  0.004440   \n",
       "2022-05-19 23:59:15   0.897421  0.894584  0.013036  0.894569  0.001802   \n",
       "2022-05-19 23:59:45   0.897796  0.894965  0.012252  0.894923  0.002548   \n",
       "\n",
       "                          bp3       bs3       bp4       bs4       bp5  ...  \\\n",
       "primary_key                                                            ...   \n",
       "2022-05-19 00:00:15  0.022076  0.016586  0.021914  0.014328  0.021840  ...   \n",
       "2022-05-19 00:00:45  0.041730  0.015180  0.041717  0.020940  0.041636  ...   \n",
       "2022-05-19 00:01:15  0.056147  0.010647  0.056145  0.010836  0.056107  ...   \n",
       "2022-05-19 00:01:45  0.059852  0.006432  0.059911  0.009208  0.059876  ...   \n",
       "2022-05-19 00:02:15  0.059357  0.010483  0.059362  0.006564  0.059318  ...   \n",
       "...                       ...       ...       ...       ...       ...  ...   \n",
       "2022-05-19 23:57:45  0.904897  0.005247  0.904900  0.009869  0.904878  ...   \n",
       "2022-05-19 23:58:15  0.898324  0.007901  0.898278  0.001949  0.898280  ...   \n",
       "2022-05-19 23:58:45  0.894151  0.005782  0.894146  0.009229  0.894110  ...   \n",
       "2022-05-19 23:59:15  0.894742  0.031284  0.894712  0.048701  0.894641  ...   \n",
       "2022-05-19 23:59:45  0.895127  0.008091  0.895181  0.021924  0.895162  ...   \n",
       "\n",
       "                         as19      ap20      as20       WAP    spread  \\\n",
       "primary_key                                                             \n",
       "2022-05-19 00:00:15  0.077064  0.022187  0.169244  0.022108  0.064543   \n",
       "2022-05-19 00:00:45  0.085127  0.042124  0.127689  0.041766  0.038022   \n",
       "2022-05-19 00:01:15  0.037006  0.057108  0.056464  0.056272  0.011147   \n",
       "2022-05-19 00:01:45  0.056320  0.060308  0.051441  0.059963  0.011140   \n",
       "2022-05-19 00:02:15  0.048770  0.059840  0.055214  0.059416  0.023907   \n",
       "...                       ...       ...       ...       ...       ...   \n",
       "2022-05-19 23:57:45  0.122925  0.903624  0.105144  0.904752  0.000064   \n",
       "2022-05-19 23:58:15  0.093584  0.897528  0.206182  0.898314  0.000071   \n",
       "2022-05-19 23:58:45  0.088406  0.893198  0.097990  0.894154  0.000076   \n",
       "2022-05-19 23:59:15  0.033136  0.893706  0.063823  0.894545  0.000075   \n",
       "2022-05-19 23:59:45  0.056774  0.893793  0.088835  0.894930  0.000075   \n",
       "\n",
       "                     log_returns  volatility_t+1  bid depth  ask depth  \\\n",
       "primary_key                                                              \n",
       "2022-05-19 00:00:15     0.381071        0.000000   0.073940   0.149910   \n",
       "2022-05-19 00:00:45     0.381097        0.000000   0.024647   0.256198   \n",
       "2022-05-19 00:01:15     0.355279        0.199778   0.020316   0.100860   \n",
       "2022-05-19 00:01:45     0.301312        0.514418   0.017963   0.139749   \n",
       "2022-05-19 00:02:15     0.280179        0.626419   0.015232   0.105190   \n",
       "...                          ...             ...        ...        ...   \n",
       "2022-05-19 23:57:45     0.314677        0.676196   0.017869   0.104124   \n",
       "2022-05-19 23:58:15     0.252438        0.676141   0.017467   0.151608   \n",
       "2022-05-19 23:58:45     0.263214        0.676051   0.017791   0.111075   \n",
       "2022-05-19 23:59:15     0.284753        0.675934   0.027510   0.123202   \n",
       "2022-05-19 23:59:45     0.284728        0.000000   0.021191   0.215764   \n",
       "\n",
       "                        FDOFI  \n",
       "primary_key                    \n",
       "2022-05-19 00:00:15  0.639872  \n",
       "2022-05-19 00:00:45  0.267512  \n",
       "2022-05-19 00:01:15  0.439577  \n",
       "2022-05-19 00:01:45  0.345256  \n",
       "2022-05-19 00:02:15  0.376714  \n",
       "...                       ...  \n",
       "2022-05-19 23:57:45  0.408075  \n",
       "2022-05-19 23:58:15  0.322110  \n",
       "2022-05-19 23:58:45  0.393519  \n",
       "2022-05-19 23:59:15  0.459037  \n",
       "2022-05-19 23:59:45  0.278242  \n",
       "\n",
       "[2880 rows x 88 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "04e3e541",
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "1e3a28a7",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.histplot(data['volatility_t+1'], kde = True, stat='density');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "51ec4340",
   "metadata": {},
   "outputs": [],
   "source": [
    "# above 0.75 -> high\n",
    "#between 0.65 & 0.75 -> medium\n",
    "# less than 0.65 -> low"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3efce7c0",
   "metadata": {},
   "source": [
    "Let's keep the levels this way, and compare as we keep improving the models. Keep in mind that it is only a *30 minute snapchot* thus it might not have captured the real level of volatility"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f28075f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# data['volatility_levels'] = pd.cut(x = data['realized_volatility'],\n",
    "#                        bins=[data['realized_volatility'].min()-1,\n",
    "#                              np.percentile(data['realized_volatility'], 25),\n",
    "#                              np.percentile(data['realized_volatility'], 50),\n",
    "#                              np.percentile(data['realized_volatility'],75),\n",
    "#                              data['realized_volatility'].max()+1], \n",
    "#                        labels=['low', 'medium','high'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bcdcbecc",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "data.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c3020151",
   "metadata": {},
   "source": [
    "#  Models"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8031b137",
   "metadata": {},
   "source": [
    "## Baseline model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1bcf6e03",
   "metadata": {},
   "outputs": [],
   "source": [
    "#baseline_score = 0.25"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09948082",
   "metadata": {},
   "source": [
    "List of model to try:\n",
    "\n",
    "- Time Series\n",
    "- RNN\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "a77581af",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_size = 0.65\n",
    "index = round(train_size*data.shape[0])\n",
    "df_train = data.iloc[:index]\n",
    "df_test = data.iloc[index:]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "22ff9d41",
   "metadata": {},
   "source": [
    "## Linear Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "b5431206",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "11078384",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression()"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Instanciate the model\n",
    "linear = LinearRegression()\n",
    "\n",
    "# Define X and y\n",
    "X = df_train.drop(columns = 'volatility_t+1')\n",
    "y = df_train['volatility_t+1']\n",
    "X_test = df_test.drop(columns = 'volatility_t+1')\n",
    "y_test = df_test['volatility_t+1']\n",
    "# Train the model on the data\n",
    "linear.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "d0ce4bba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.4940299 , 0.51848786, 0.5547123 , ..., 0.4917857 , 0.48312113,\n",
       "       0.54528739])"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "linear.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "cf0f194d",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-28.88967725965776"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "linear.score(X_test,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "292ca2e9",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Creating the Pipeline\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.model_selection import cross_val_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "9e6d5f2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "pipe = Pipeline([\n",
    "    ('scaler', MinMaxScaler()),\n",
    "    ('linear', LinearRegression())\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "6bd32bb5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-429.25231950039307"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_val_score(pipe, X, y, cv=10, scoring='r2').mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "8ed6b4ac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(steps=[('scaler', MinMaxScaler()), ('linear', LinearRegression())])"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "d2b089c0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.4940299 , 0.51848786, 0.5547123 , ..., 0.4917857 , 0.48312113,\n",
       "       0.54528739])"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prediction = pipe.predict(X_test)\n",
    "prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "96762f32",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-28.889677254164535"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe_score = pipe.score(X_test, y_test)\n",
    "pipe_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "99aa9251",
   "metadata": {},
   "outputs": [],
   "source": [
    "import joblib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "93597453",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['model.joblib']"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(pipe, 'model.joblib')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cffe7e77",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.13"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}