MachineLearningInteractive.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 389,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import sklearn as sk\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import yfinance as yf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 390,
   "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>index</th>\n",
       "      <th>employees</th>\n",
       "      <th>CEO_age</th>\n",
       "      <th>Name</th>\n",
       "      <th>Symbol</th>\n",
       "      <th>sector</th>\n",
       "      <th>IPO_weekday</th>\n",
       "      <th>firstday_volume</th>\n",
       "      <th>inweek_volume</th>\n",
       "      <th>Y_Open_Price</th>\n",
       "      <th>intra_day_up</th>\n",
       "      <th>intra_week_up</th>\n",
       "      <th>intra_month_up</th>\n",
       "      <th>SP500_lastweek_return</th>\n",
       "      <th>positive</th>\n",
       "      <th>negative</th>\n",
       "      <th>neutral</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ipoDate</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>2010-01-26</th>\n",
       "      <td>0</td>\n",
       "      <td>128.000000</td>\n",
       "      <td>57.460211</td>\n",
       "      <td>ANDATEE CHINA MARINE FUEL SERVICES CORP</td>\n",
       "      <td>AMCF</td>\n",
       "      <td>Energy</td>\n",
       "      <td>1</td>\n",
       "      <td>955563.0</td>\n",
       "      <td>5.878947e+06</td>\n",
       "      <td>6.500000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.007087</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-02-11</th>\n",
       "      <td>2</td>\n",
       "      <td>1486.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>GENERAC HOLDINGS INC.</td>\n",
       "      <td>GNRC</td>\n",
       "      <td>Industrials</td>\n",
       "      <td>3</td>\n",
       "      <td>16892087.0</td>\n",
       "      <td>2.179334e+06</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.003156</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-02-11</th>\n",
       "      <td>3</td>\n",
       "      <td>568.000000</td>\n",
       "      <td>62.000000</td>\n",
       "      <td>QUINSTREET, INC</td>\n",
       "      <td>QNST</td>\n",
       "      <td>Technology</td>\n",
       "      <td>3</td>\n",
       "      <td>5372077.0</td>\n",
       "      <td>3.362175e+06</td>\n",
       "      <td>15.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.003156</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-02-10</th>\n",
       "      <td>4</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>61.000000</td>\n",
       "      <td>TERRENO REALTY CORP</td>\n",
       "      <td>TRNO</td>\n",
       "      <td>Real Estate</td>\n",
       "      <td>2</td>\n",
       "      <td>4515261.0</td>\n",
       "      <td>7.616977e+05</td>\n",
       "      <td>18.750000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.002686</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-02-10</th>\n",
       "      <td>5</td>\n",
       "      <td>109.000000</td>\n",
       "      <td>59.000000</td>\n",
       "      <td>PIEDMONT OFFICE REALTY TRUST, INC.</td>\n",
       "      <td>PDM</td>\n",
       "      <td>Real Estate</td>\n",
       "      <td>2</td>\n",
       "      <td>4343012.0</td>\n",
       "      <td>3.551568e+06</td>\n",
       "      <td>14.750000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.002686</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-03-22</th>\n",
       "      <td>3736</td>\n",
       "      <td>59.000000</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>ZAIS Group Holdings, Inc.</td>\n",
       "      <td>ZAIS</td>\n",
       "      <td>Finance</td>\n",
       "      <td>4</td>\n",
       "      <td>1511224.0</td>\n",
       "      <td>3.912183e+05</td>\n",
       "      <td>10.020000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000234</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-06-22</th>\n",
       "      <td>3739</td>\n",
       "      <td>7447.262841</td>\n",
       "      <td>54.787097</td>\n",
       "      <td>ProShares UltraPro Short NASDAQ Biotechnology</td>\n",
       "      <td>ZBIO</td>\n",
       "      <td>Other</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.593089e+05</td>\n",
       "      <td>41.900002</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000964</td>\n",
       "      <td>0.063431</td>\n",
       "      <td>0.033641</td>\n",
       "      <td>0.902928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-07-20</th>\n",
       "      <td>3743</td>\n",
       "      <td>3181.000000</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>Zillow Group, Inc.</td>\n",
       "      <td>ZG</td>\n",
       "      <td>Miscellaneous</td>\n",
       "      <td>2</td>\n",
       "      <td>17549817.0</td>\n",
       "      <td>6.998521e+06</td>\n",
       "      <td>17.776737</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000717</td>\n",
       "      <td>0.111890</td>\n",
       "      <td>0.263831</td>\n",
       "      <td>0.624278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-11-29</th>\n",
       "      <td>3746</td>\n",
       "      <td>7447.262841</td>\n",
       "      <td>54.787097</td>\n",
       "      <td>Credit Suisse AG</td>\n",
       "      <td>ZIV</td>\n",
       "      <td>Finance</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.145535e+04</td>\n",
       "      <td>12.298750</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.001159</td>\n",
       "      <td>0.151475</td>\n",
       "      <td>0.153023</td>\n",
       "      <td>0.695502</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-04-11</th>\n",
       "      <td>3752</td>\n",
       "      <td>5473.000000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>Zoe&amp;#39;s Kitchen, Inc.</td>\n",
       "      <td>ZOES</td>\n",
       "      <td>Consumer Services</td>\n",
       "      <td>4</td>\n",
       "      <td>5765164.0</td>\n",
       "      <td>2.644819e+06</td>\n",
       "      <td>25.612000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.005731</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.330000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1711 rows × 17 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            index    employees    CEO_age  \\\n",
       "ipoDate                                     \n",
       "2010-01-26      0   128.000000  57.460211   \n",
       "2010-02-11      2  1486.000000  50.000000   \n",
       "2010-02-11      3   568.000000  62.000000   \n",
       "2010-02-10      4     6.000000  61.000000   \n",
       "2010-02-10      5   109.000000  59.000000   \n",
       "...           ...          ...        ...   \n",
       "2013-03-22   3736    59.000000  58.000000   \n",
       "2015-06-22   3739  7447.262841  54.787097   \n",
       "2011-07-20   3743  3181.000000  41.000000   \n",
       "2010-11-29   3746  7447.262841  54.787097   \n",
       "2014-04-11   3752  5473.000000  49.000000   \n",
       "\n",
       "                                                     Name Symbol  \\\n",
       "ipoDate                                                            \n",
       "2010-01-26        ANDATEE CHINA MARINE FUEL SERVICES CORP   AMCF   \n",
       "2010-02-11                          GENERAC HOLDINGS INC.   GNRC   \n",
       "2010-02-11                                QUINSTREET, INC   QNST   \n",
       "2010-02-10                            TERRENO REALTY CORP   TRNO   \n",
       "2010-02-10             PIEDMONT OFFICE REALTY TRUST, INC.    PDM   \n",
       "...                                                   ...    ...   \n",
       "2013-03-22                      ZAIS Group Holdings, Inc.   ZAIS   \n",
       "2015-06-22  ProShares UltraPro Short NASDAQ Biotechnology   ZBIO   \n",
       "2011-07-20                             Zillow Group, Inc.     ZG   \n",
       "2010-11-29                               Credit Suisse AG    ZIV   \n",
       "2014-04-11                        Zoe&#39;s Kitchen, Inc.   ZOES   \n",
       "\n",
       "                       sector  IPO_weekday  firstday_volume  inweek_volume  \\\n",
       "ipoDate                                                                      \n",
       "2010-01-26             Energy            1         955563.0   5.878947e+06   \n",
       "2010-02-11        Industrials            3       16892087.0   2.179334e+06   \n",
       "2010-02-11         Technology            3        5372077.0   3.362175e+06   \n",
       "2010-02-10        Real Estate            2        4515261.0   7.616977e+05   \n",
       "2010-02-10        Real Estate            2        4343012.0   3.551568e+06   \n",
       "...                       ...          ...              ...            ...   \n",
       "2013-03-22            Finance            4        1511224.0   3.912183e+05   \n",
       "2015-06-22              Other            0              0.0   1.593089e+05   \n",
       "2011-07-20      Miscellaneous            2       17549817.0   6.998521e+06   \n",
       "2010-11-29            Finance            0              0.0   7.145535e+04   \n",
       "2014-04-11  Consumer Services            4        5765164.0   2.644819e+06   \n",
       "\n",
       "            Y_Open_Price  intra_day_up  intra_week_up  intra_month_up  \\\n",
       "ipoDate                                                                 \n",
       "2010-01-26      6.500000             1              1               0   \n",
       "2010-02-11     13.000000             1              1               1   \n",
       "2010-02-11     15.000000             0              1               0   \n",
       "2010-02-10     18.750000             1              1               1   \n",
       "2010-02-10     14.750000             1              1               1   \n",
       "...                  ...           ...            ...             ...   \n",
       "2013-03-22     10.020000             0              0               0   \n",
       "2015-06-22     41.900002             0              0               1   \n",
       "2011-07-20     17.776737             1              1               1   \n",
       "2010-11-29     12.298750             0              0               0   \n",
       "2014-04-11     25.612000             1              0               0   \n",
       "\n",
       "            SP500_lastweek_return  positive  negative   neutral  \n",
       "ipoDate                                                          \n",
       "2010-01-26              -0.007087  0.330000  0.330000  0.330000  \n",
       "2010-02-11              -0.003156  0.330000  0.330000  0.330000  \n",
       "2010-02-11              -0.003156  0.330000  0.330000  0.330000  \n",
       "2010-02-10              -0.002686  0.330000  0.330000  0.330000  \n",
       "2010-02-10              -0.002686  0.330000  0.330000  0.330000  \n",
       "...                           ...       ...       ...       ...  \n",
       "2013-03-22               0.000234  0.330000  0.330000  0.330000  \n",
       "2015-06-22               0.000964  0.063431  0.033641  0.902928  \n",
       "2011-07-20               0.000717  0.111890  0.263831  0.624278  \n",
       "2010-11-29               0.001159  0.151475  0.153023  0.695502  \n",
       "2014-04-11              -0.005731  0.330000  0.330000  0.330000  \n",
       "\n",
       "[1711 rows x 17 columns]"
      ]
     },
     "execution_count": 390,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Importing dataset and defining path variables\n",
    "path = './data/'\n",
    "file = 'master_data.csv'\n",
    "\n",
    "df = pd.read_csv(path + file, index_col='ipoDate')\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 391,
   "metadata": {},
   "outputs": [],
   "source": [
    "#One hot encoding sector and weekday cols\n",
    "categorical_cols = ['sector', 'IPO_weekday'] \n",
    "df = pd.get_dummies(df, columns = categorical_cols)\n",
    "\n",
    "#Dropping Symbol, Name and Index columns\n",
    "df.drop(columns=['Symbol', 'Name', 'index'], inplace=True, axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 392,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Normalizing numerical variables for ML modeling\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "scaler = MinMaxScaler()\n",
    "\n",
    "normalized_data_train = scaler.fit_transform(df)\n",
    "df_norm = pd.DataFrame(normalized_data_train, columns=df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 393,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Class distribution: \n",
      " intra_day_up      0.692577\n",
      "intra_week_up     0.697838\n",
      "intra_month_up    0.659264\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "#Logistic Regression Modeling\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import train_test_split\n",
    "from imblearn.over_sampling import SMOTE\n",
    "from sklearn import metrics\n",
    "\n",
    "X_d = df_norm.drop(columns= ['intra_day_up', 'intra_week_up', 'intra_month_up'])\n",
    "X_w = df_norm.drop(columns= ['intra_day_up', 'intra_week_up', 'intra_month_up'])\n",
    "X_m = df_norm.drop(columns= ['intra_day_up', 'intra_week_up', 'intra_month_up'])\n",
    "y_d = np.array(df_norm.intra_day_up)\n",
    "y_w = np.array(df_norm.intra_week_up)\n",
    "y_m = np.array(df.intra_month_up)\n",
    "\n",
    "#Looking at class imbalances\n",
    "print('Class distribution: \\n',df_norm[['intra_day_up','intra_week_up', 'intra_month_up']].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 394,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Class distribution after SMOTE: 0.50\n",
      "Class distribution after SMOTE: 0.50\n",
      "Class distribution after SMOTE: 0.50\n"
     ]
    }
   ],
   "source": [
    "#Dealing with class imbalances by oversampling minority class (underperforming IPOs)\n",
    "sm = SMOTE(random_state=42)\n",
    "X_d_bal, y_d_bal = sm.fit_resample(X_d, y_d)\n",
    "X_w_bal, y_w_bal = sm.fit_resample(X_w, y_w)\n",
    "X_m_bal, y_m_bal = sm.fit_resample(X_m, y_m)\n",
    "\n",
    "#Printing mean of y values (should be 0.5, since data should now be balanced)\n",
    "print('Class distribution after SMOTE: {:.2f}'.format(y_d_bal.mean()))\n",
    "print('Class distribution after SMOTE: {:.2f}'.format(y_w_bal.mean()))\n",
    "print('Class distribution after SMOTE: {:.2f}'.format(y_m_bal.mean()))\n",
    "\n",
    "# setting up stratified testing and training sets\n",
    "X_d_train, X_d_test, y_d_train, y_d_test = train_test_split(X_d_bal, y_d_bal, test_size=0.2, random_state=42, stratify = y_d_bal)\n",
    "X_w_train, X_w_test, y_w_train, y_w_test = train_test_split(X_w_bal, y_w_bal, test_size=0.2, random_state=42, stratify = y_w_bal)\n",
    "X_m_train, X_m_test, y_m_train, y_m_test = train_test_split(X_m_bal, y_m_bal, test_size=0.2, random_state=42, stratify = y_m_bal)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 395,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-40 {color: black;background-color: white;}#sk-container-id-40 pre{padding: 0;}#sk-container-id-40 div.sk-toggleable {background-color: white;}#sk-container-id-40 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-40 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-40 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-40 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-40 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-40 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-40 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-40 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-40 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-40 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-40 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-40 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-40 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-40 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-40 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-40 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-40 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-40 div.sk-item {position: relative;z-index: 1;}#sk-container-id-40 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-40 div.sk-item::before, #sk-container-id-40 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-40 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-40 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-40 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-40 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-40 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-40 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-40 div.sk-label-container {text-align: center;}#sk-container-id-40 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-40 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-40\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>LogisticRegression()</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-84\" type=\"checkbox\" checked><label for=\"sk-estimator-id-84\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression()</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "LogisticRegression()"
      ]
     },
     "execution_count": 395,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting Logistic Regression Models\n",
    "logreg_d = LogisticRegression()\n",
    "logreg_w = LogisticRegression()\n",
    "logreg_m = LogisticRegression()\n",
    "\n",
    "logreg_d.fit(X_d_train, y_d_train)\n",
    "logreg_w.fit(X_w_train, y_w_train)\n",
    "logreg_m.fit(X_m_train, y_m_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 396,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of logistic regression classifier on daily up test set: 0.67\n",
      "Accuracy of logistic regression classifier on weekly up test set: 0.65\n",
      "Accuracy of logistic regression classifier on monthly up test set: 0.68\n"
     ]
    }
   ],
   "source": [
    "log_pred_d = logreg_d.predict(X_d_test)\n",
    "log_pred_w = logreg_w.predict(X_w_test)\n",
    "log_pred_m = logreg_m.predict(X_m_test)\n",
    "print('Accuracy of logistic regression classifier on daily up test set: {:.2f}'.format(logreg_d.score(X_d_test, y_d_test)))\n",
    "print('Accuracy of logistic regression classifier on weekly up test set: {:.2f}'.format(logreg_w.score(X_w_test, y_w_test)))\n",
    "print('Accuracy of logistic regression classifier on monthly up test set: {:.2f}'.format(logreg_m.score(X_m_test, y_m_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 397,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.64      0.79      0.71       237\n",
      "         1.0       0.72      0.54      0.62       237\n",
      "\n",
      "    accuracy                           0.67       474\n",
      "   macro avg       0.68      0.67      0.66       474\n",
      "weighted avg       0.68      0.67      0.66       474\n",
      "\n",
      "[[188  49]\n",
      " [108 129]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "\n",
    "print(classification_report(y_d_test, log_pred_d))\n",
    "confusion_matrix = confusion_matrix(y_d_test, log_pred_d)\n",
    "\n",
    "#TN #FP\n",
    "#FN #TP\n",
    "print(confusion_matrix)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 398,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "#Creating grid search for Random Forest Hyperparameter Tuning\n",
    "\n",
    "# Number of trees in random forest\n",
    "n_estimators = [int(x) for x in np.linspace(start = 100, stop = 1500, num = 10)]\n",
    "# Maximum number of levels in tree\n",
    "max_depth = [int(x) for x in np.linspace(10, 110, num = 11)]\n",
    "max_depth.append(None)\n",
    "# Minimum number of samples required to split a node\n",
    "min_samples_split = [2, 5, 10]\n",
    "# Minimum number of samples required at each leaf node\n",
    "min_samples_leaf = [1, 2, 4]\n",
    "# Method of selecting samples for training each tree\n",
    "bootstrap = [True, False]\n",
    "# Create the random grid\n",
    "random_grid = {'n_estimators': n_estimators,\n",
    "               'max_depth': max_depth,\n",
    "               'min_samples_split': min_samples_split,\n",
    "               'min_samples_leaf': min_samples_leaf,\n",
    "               'bootstrap': bootstrap}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 399,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "First randomized search completed.\n",
      "Second randomized search completed.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-41 {color: black;background-color: white;}#sk-container-id-41 pre{padding: 0;}#sk-container-id-41 div.sk-toggleable {background-color: white;}#sk-container-id-41 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-41 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-41 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-41 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-41 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-41 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-41 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-41 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-41 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-41 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-41 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-41 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-41 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-41 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-41 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-41 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-41 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-41 div.sk-item {position: relative;z-index: 1;}#sk-container-id-41 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-41 div.sk-item::before, #sk-container-id-41 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-41 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-41 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-41 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-41 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-41 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-41 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-41 div.sk-label-container {text-align: center;}#sk-container-id-41 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-41 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-41\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomizedSearchCV(cv=10, estimator=RandomForestClassifier(), n_iter=100,\n",
       "                   n_jobs=-1,\n",
       "                   param_distributions={&#x27;bootstrap&#x27;: [True, False],\n",
       "                                        &#x27;max_depth&#x27;: [10, 20, 30, 40, 50, 60,\n",
       "                                                      70, 80, 90, 100, 110,\n",
       "                                                      None],\n",
       "                                        &#x27;min_samples_leaf&#x27;: [1, 2, 4],\n",
       "                                        &#x27;min_samples_split&#x27;: [2, 5, 10],\n",
       "                                        &#x27;n_estimators&#x27;: [100, 255, 411, 566,\n",
       "                                                         722, 877, 1033, 1188,\n",
       "                                                         1344, 1500]},\n",
       "                   random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-85\" type=\"checkbox\" ><label for=\"sk-estimator-id-85\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomizedSearchCV</label><div class=\"sk-toggleable__content\"><pre>RandomizedSearchCV(cv=10, estimator=RandomForestClassifier(), n_iter=100,\n",
       "                   n_jobs=-1,\n",
       "                   param_distributions={&#x27;bootstrap&#x27;: [True, False],\n",
       "                                        &#x27;max_depth&#x27;: [10, 20, 30, 40, 50, 60,\n",
       "                                                      70, 80, 90, 100, 110,\n",
       "                                                      None],\n",
       "                                        &#x27;min_samples_leaf&#x27;: [1, 2, 4],\n",
       "                                        &#x27;min_samples_split&#x27;: [2, 5, 10],\n",
       "                                        &#x27;n_estimators&#x27;: [100, 255, 411, 566,\n",
       "                                                         722, 877, 1033, 1188,\n",
       "                                                         1344, 1500]},\n",
       "                   random_state=42)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-86\" type=\"checkbox\" ><label for=\"sk-estimator-id-86\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier()</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-87\" type=\"checkbox\" ><label for=\"sk-estimator-id-87\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier()</pre></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomizedSearchCV(cv=10, estimator=RandomForestClassifier(), n_iter=100,\n",
       "                   n_jobs=-1,\n",
       "                   param_distributions={'bootstrap': [True, False],\n",
       "                                        'max_depth': [10, 20, 30, 40, 50, 60,\n",
       "                                                      70, 80, 90, 100, 110,\n",
       "                                                      None],\n",
       "                                        'min_samples_leaf': [1, 2, 4],\n",
       "                                        'min_samples_split': [2, 5, 10],\n",
       "                                        'n_estimators': [100, 255, 411, 566,\n",
       "                                                         722, 877, 1033, 1188,\n",
       "                                                         1344, 1500]},\n",
       "                   random_state=42)"
      ]
     },
     "execution_count": 399,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "# Use the random grid to search for best hyperparameters\n",
    "# First create the base models to tune for each predicted label\n",
    "rf_d = RandomForestClassifier()\n",
    "rf_w = RandomForestClassifier()\n",
    "rf_m = RandomForestClassifier()\n",
    "\n",
    "# Random search of parameters, using 10 fold cross validation, \n",
    "# search across 50 different combinations, and use all available cores\n",
    "rf_random_d = RandomizedSearchCV(estimator = rf_d, param_distributions = random_grid, n_iter = 100, cv = 10, random_state=42, n_jobs = -1)\n",
    "rf_random_w = RandomizedSearchCV(estimator = rf_w, param_distributions = random_grid, n_iter = 100, cv = 10, random_state=42, n_jobs = -1)\n",
    "rf_random_m = RandomizedSearchCV(estimator = rf_m, param_distributions = random_grid, n_iter = 100, cv = 10, random_state=42, n_jobs = -1)\n",
    "\n",
    "# Fit all random search models (Takes ca. 12 min to run)\n",
    "rf_random_d.fit(X_d_train, y_d_train)\n",
    "\n",
    "print('First randomized search completed.')\n",
    "\n",
    "rf_random_w.fit(X_w_train, y_w_train)\n",
    "\n",
    "print('Second randomized search completed.')\n",
    "\n",
    "rf_random_m.fit(X_m_train, y_m_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 400,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Balanced accuracy of random forest classifier on daily up test set: 0.79\n",
      "Balanced accuracy of random forest classifier on weekly up test set: 0.76\n",
      "Balanced accuracy of random forest classifier on monthly up test set: 0.76\n"
     ]
    }
   ],
   "source": [
    "#Evaluating accuracy of rf models\n",
    "best_random_rf_d = rf_random_d.best_estimator_\n",
    "best_random_rf_w = rf_random_w.best_estimator_\n",
    "best_random_rf_m = rf_random_m.best_estimator_\n",
    "\n",
    "#Creating random model predictions\n",
    "rf_predictions_d = best_random_rf_d.predict(X_d_test)\n",
    "rf_predictions_w = best_random_rf_w.predict(X_w_test)\n",
    "rf_predictions_m = best_random_rf_m.predict(X_m_test)\n",
    "\n",
    "#Evaluation\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "\n",
    "print('Balanced accuracy of random forest classifier on daily up test set: {:.2f}'.format(balanced_accuracy_score(y_d_test, rf_predictions_d)))\n",
    "print('Balanced accuracy of random forest classifier on weekly up test set: {:.2f}'.format(balanced_accuracy_score(y_w_test, rf_predictions_w)))\n",
    "print('Balanced accuracy of random forest classifier on monthly up test set: {:.2f}'.format(balanced_accuracy_score(y_m_test, rf_predictions_m)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 401,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "First randomized search completed.\n",
      "Second randomized search completed.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-42 {color: black;background-color: white;}#sk-container-id-42 pre{padding: 0;}#sk-container-id-42 div.sk-toggleable {background-color: white;}#sk-container-id-42 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-42 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-42 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-42 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-42 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-42 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-42 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-42 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-42 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-42 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-42 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-42 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-42 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-42 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-42 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-42 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-42 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-42 div.sk-item {position: relative;z-index: 1;}#sk-container-id-42 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-42 div.sk-item::before, #sk-container-id-42 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-42 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-42 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-42 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-42 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-42 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-42 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-42 div.sk-label-container {text-align: center;}#sk-container-id-42 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-42 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-42\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomizedSearchCV(cv=10,\n",
       "                   estimator=XGBClassifier(base_score=None, booster=None,\n",
       "                                           callbacks=None,\n",
       "                                           colsample_bylevel=None,\n",
       "                                           colsample_bynode=None,\n",
       "                                           colsample_bytree=None,\n",
       "                                           early_stopping_rounds=None,\n",
       "                                           enable_categorical=False,\n",
       "                                           eval_metric=None, feature_types=None,\n",
       "                                           gamma=None, gpu_id=None,\n",
       "                                           grow_policy=None,\n",
       "                                           importance_type=None,\n",
       "                                           interaction_constraints=None,\n",
       "                                           learning_rat...\n",
       "                                           n_estimators=100, n_jobs=None,\n",
       "                                           num_parallel_tree=None,\n",
       "                                           predictor=None, random_state=None, ...),\n",
       "                   n_iter=100, n_jobs=-1,\n",
       "                   param_distributions={&#x27;colsample_bytree&#x27;: [0.5],\n",
       "                                        &#x27;gamma&#x27;: [0, 0.1, 0.25, 0.5, 1],\n",
       "                                        &#x27;learning_rate&#x27;: [0.1, 0.01, 0.05, 0.2,\n",
       "                                                          0.25, 0.3],\n",
       "                                        &#x27;max_depth&#x27;: [10, 31, 52, 73, 94, 115,\n",
       "                                                      136, 157, 178, 200],\n",
       "                                        &#x27;reg_lambda&#x27;: [0, 1, 2, 5, 10],\n",
       "                                        &#x27;scale_pos_weight&#x27;: [1, 2, 3, 4, 5],\n",
       "                                        &#x27;subsample&#x27;: [0.8]},\n",
       "                   random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-88\" type=\"checkbox\" ><label for=\"sk-estimator-id-88\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomizedSearchCV</label><div class=\"sk-toggleable__content\"><pre>RandomizedSearchCV(cv=10,\n",
       "                   estimator=XGBClassifier(base_score=None, booster=None,\n",
       "                                           callbacks=None,\n",
       "                                           colsample_bylevel=None,\n",
       "                                           colsample_bynode=None,\n",
       "                                           colsample_bytree=None,\n",
       "                                           early_stopping_rounds=None,\n",
       "                                           enable_categorical=False,\n",
       "                                           eval_metric=None, feature_types=None,\n",
       "                                           gamma=None, gpu_id=None,\n",
       "                                           grow_policy=None,\n",
       "                                           importance_type=None,\n",
       "                                           interaction_constraints=None,\n",
       "                                           learning_rat...\n",
       "                                           n_estimators=100, n_jobs=None,\n",
       "                                           num_parallel_tree=None,\n",
       "                                           predictor=None, random_state=None, ...),\n",
       "                   n_iter=100, n_jobs=-1,\n",
       "                   param_distributions={&#x27;colsample_bytree&#x27;: [0.5],\n",
       "                                        &#x27;gamma&#x27;: [0, 0.1, 0.25, 0.5, 1],\n",
       "                                        &#x27;learning_rate&#x27;: [0.1, 0.01, 0.05, 0.2,\n",
       "                                                          0.25, 0.3],\n",
       "                                        &#x27;max_depth&#x27;: [10, 31, 52, 73, 94, 115,\n",
       "                                                      136, 157, 178, 200],\n",
       "                                        &#x27;reg_lambda&#x27;: [0, 1, 2, 5, 10],\n",
       "                                        &#x27;scale_pos_weight&#x27;: [1, 2, 3, 4, 5],\n",
       "                                        &#x27;subsample&#x27;: [0.8]},\n",
       "                   random_state=42)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-89\" type=\"checkbox\" ><label for=\"sk-estimator-id-89\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=None, gpu_id=None, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
       "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
       "              n_estimators=100, n_jobs=None, num_parallel_tree=None,\n",
       "              predictor=None, random_state=None, ...)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-90\" type=\"checkbox\" ><label for=\"sk-estimator-id-90\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=None, gpu_id=None, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
       "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
       "              n_estimators=100, n_jobs=None, num_parallel_tree=None,\n",
       "              predictor=None, random_state=None, ...)</pre></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomizedSearchCV(cv=10,\n",
       "                   estimator=XGBClassifier(base_score=None, booster=None,\n",
       "                                           callbacks=None,\n",
       "                                           colsample_bylevel=None,\n",
       "                                           colsample_bynode=None,\n",
       "                                           colsample_bytree=None,\n",
       "                                           early_stopping_rounds=None,\n",
       "                                           enable_categorical=False,\n",
       "                                           eval_metric=None, feature_types=None,\n",
       "                                           gamma=None, gpu_id=None,\n",
       "                                           grow_policy=None,\n",
       "                                           importance_type=None,\n",
       "                                           interaction_constraints=None,\n",
       "                                           learning_rat...\n",
       "                                           n_estimators=100, n_jobs=None,\n",
       "                                           num_parallel_tree=None,\n",
       "                                           predictor=None, random_state=None, ...),\n",
       "                   n_iter=100, n_jobs=-1,\n",
       "                   param_distributions={'colsample_bytree': [0.5],\n",
       "                                        'gamma': [0, 0.1, 0.25, 0.5, 1],\n",
       "                                        'learning_rate': [0.1, 0.01, 0.05, 0.2,\n",
       "                                                          0.25, 0.3],\n",
       "                                        'max_depth': [10, 31, 52, 73, 94, 115,\n",
       "                                                      136, 157, 178, 200],\n",
       "                                        'reg_lambda': [0, 1, 2, 5, 10],\n",
       "                                        'scale_pos_weight': [1, 2, 3, 4, 5],\n",
       "                                        'subsample': [0.8]},\n",
       "                   random_state=42)"
      ]
     },
     "execution_count": 401,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from xgboost import XGBClassifier\n",
    "\n",
    "#Creating XGB models for each predicted label\n",
    "xgb_d = XGBClassifier(objective=\"binary:logistic\")\n",
    "xgb_w = XGBClassifier(objective=\"binary:logistic\")\n",
    "xgb_m = XGBClassifier(objective=\"binary:logistic\")\n",
    "\n",
    "#Defining search grid for randomized search\n",
    "param_grid = {\n",
    "    \"max_depth\": [int(x) for x in np.linspace(10, 200, num = 10)],\n",
    "    \"learning_rate\": [0.1, 0.01, 0.05, 0.2, 0.25, 0.3],\n",
    "    \"gamma\": [0, 0.1, 0.25, 0.5, 1],\n",
    "    \"reg_lambda\": [0, 1, 2, 5, 10],\n",
    "    \"scale_pos_weight\": [1, 2, 3, 4, 5],\n",
    "    \"subsample\": [0.8],\n",
    "    \"colsample_bytree\": [0.5],\n",
    "}\n",
    "\n",
    "#Fitting randomized search to XGB models\n",
    "xgb_random_d = RandomizedSearchCV(estimator = xgb_d, param_distributions = param_grid, n_iter = 100, cv = 10, random_state=42, n_jobs = -1)\n",
    "xgb_random_w = RandomizedSearchCV(estimator = xgb_w, param_distributions = param_grid, n_iter = 100, cv = 10, random_state=42, n_jobs = -1)\n",
    "xgb_random_m = RandomizedSearchCV(estimator = xgb_m, param_distributions = param_grid, n_iter = 100, cv = 10, random_state=42, n_jobs = -1)\n",
    "\n",
    "#Fitting search to training data (Takes around 15 min to run)\n",
    "xgb_random_d.fit(X_d_train, y_d_train)\n",
    "\n",
    "print('First randomized search completed.')\n",
    "\n",
    "xgb_random_w.fit(X_w_train, y_w_train)\n",
    "\n",
    "print('Second randomized search completed.')\n",
    "\n",
    "xgb_random_m.fit(X_m_train, y_m_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 402,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Balanced accuracy of xgboost classifier on daily up test set: 0.77\n",
      "Balanced accuracy of xgboost classifier on weekly up test set: 0.74\n",
      "Balanced accuracy of xgboost classifier on monthly up test set: 0.73\n"
     ]
    }
   ],
   "source": [
    "#Evaluating accuracy of xgb models\n",
    "best_random_d = xgb_random_d.best_estimator_\n",
    "best_random_w = xgb_random_w.best_estimator_\n",
    "best_random_m = xgb_random_m.best_estimator_\n",
    "\n",
    "#Creating random model predictions\n",
    "xgb_predictions_d = best_random_d.predict(X_d_test)\n",
    "xgb_predictions_w = best_random_w.predict(X_w_test)\n",
    "xgb_predictions_m = best_random_m.predict(X_m_test)\n",
    "\n",
    "#Evaluation\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "\n",
    "print('Balanced accuracy of xgboost classifier on daily up test set: {:.2f}'.format(balanced_accuracy_score(y_d_test, xgb_predictions_d)))\n",
    "print('Balanced accuracy of xgboost classifier on weekly up test set: {:.2f}'.format(balanced_accuracy_score(y_w_test, xgb_predictions_w)))\n",
    "print('Balanced accuracy of xgboost classifier on monthly up test set: {:.2f}'.format(balanced_accuracy_score(y_m_test, xgb_predictions_m)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 403,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Looking at feature importances across models\n",
    "\n",
    "#Feature importance for rf models\n",
    "feature_names_d = [col for col in X_d.columns]\n",
    "feature_names_w = [col for col in X_w.columns]\n",
    "feature_names_m = [col for col in X_m.columns]\n",
    "rf_d_importance = pd.Series(rf_random_d.best_estimator_.feature_importances_, index = feature_names_d)\n",
    "rf_w_importance = pd.Series(rf_random_w.best_estimator_.feature_importances_, index = feature_names_w)\n",
    "rf_m_importance = pd.Series(rf_random_m.best_estimator_.feature_importances_, index = feature_names_m)\n",
    "\n",
    "#Feature importance for xgb models\n",
    "xgb_d_importance = pd.Series(xgb_random_d.best_estimator_.feature_importances_, index = feature_names_d)\n",
    "xgb_w_importance = pd.Series(xgb_random_w.best_estimator_.feature_importances_, index = feature_names_w)\n",
    "xgb_m_importance = pd.Series(xgb_random_m.best_estimator_.feature_importances_, index = feature_names_m)\n",
    "\n",
    "#Looking at most important features in logistic regression model\n",
    "log_d_importance = pd.Series(abs(logreg_d.coef_[0]), index = X_d.columns)\n",
    "log_w_importance = pd.Series(abs(logreg_w.coef_[0]), index = X_w.columns)\n",
    "log_m_importance = pd.Series(abs(logreg_m.coef_[0]), index = X_m.columns)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 404,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Getting average feature importances\n",
    "xgb_importances = pd.DataFrame(xgb_d_importance).join(pd.DataFrame(xgb_w_importance), rsuffix = '_w').join(pd.DataFrame(xgb_m_importance), rsuffix = '_m')\n",
    "\n",
    "#Dropping trading volumes since they are not included in all models\n",
    "xgb_importances.drop('firstday_volume', inplace = True)\n",
    "xgb_importances.drop('inweek_volume', inplace = True)\n",
    "\n",
    "xgb_importances['Feature Importance'] = xgb_importances.apply(lambda x: x.mean(), axis = 1)\n",
    "xgb_importances = xgb_importances[['Feature Importance']]\n",
    "xgb_importances = xgb_importances.sort_values(by = 'Feature Importance', ascending=False)\n",
    "\n",
    "#Getting average feature importances\n",
    "rf_importances = pd.DataFrame(rf_d_importance).join(pd.DataFrame(rf_w_importance), rsuffix = '_w').join(pd.DataFrame(rf_m_importance), rsuffix = '_m')\n",
    "\n",
    "#Dropping trading volumes since they are not included in all models\n",
    "rf_importances.drop('firstday_volume', inplace = True)\n",
    "rf_importances.drop('inweek_volume', inplace = True)\n",
    "\n",
    "rf_importances['Feature Importance'] = rf_importances.apply(lambda x: x.mean(), axis = 1)\n",
    "rf_importances = rf_importances[['Feature Importance']]\n",
    "rf_importances = rf_importances.sort_values(by = 'Feature Importance', ascending=False)\n",
    "\n",
    "#Getting average feature importances\n",
    "log_importances = pd.DataFrame(log_d_importance).join(pd.DataFrame(log_w_importance), rsuffix = '_w').join(pd.DataFrame(log_m_importance), rsuffix = '_m')\n",
    "\n",
    "#Dropping trading volumes since they are not included in all models\n",
    "log_importances.drop('firstday_volume', inplace = True)\n",
    "log_importances.drop('inweek_volume', inplace = True)\n",
    "\n",
    "log_importances['Feature Importance'] = log_importances.apply(lambda x: x.mean(), axis = 1)\n",
    "log_importances = log_importances[['Feature Importance']]\n",
    "log_importances = log_importances.sort_values(by = 'Feature Importance', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 405,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2500x2500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2, ax3) = plt.subplots(3, 1, figsize = (25,25))\n",
    "rf_importances[:10].plot.bar(ax = ax1)\n",
    "xgb_importances[:10].plot.bar(ax = ax2)\n",
    "log_importances[:10].plot.bar(ax = ax3)\n",
    "ax1.set_title(\"Random Forest Feature importances using MDI\")\n",
    "ax1.set_ylabel(\"Mean decrease in impurity\")\n",
    "ax2.set_title(\"XGBoost Feature importances using MDI\")\n",
    "ax2.set_ylabel(\"XGB Mean decrease in impurity\")\n",
    "ax3.set_title(\"Log Reg Feature importances using Abs. Coef. Value\")\n",
    "ax3.set_ylabel(\"Absolute Coefficient Values\")\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 406,
   "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>pred_label</th>\n",
       "      <th>recall</th>\n",
       "      <th>balanced_accuracy</th>\n",
       "      <th>top_5_features</th>\n",
       "      <th>model</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>intra day underperformance</td>\n",
       "      <td>0.755274</td>\n",
       "      <td>0.791139</td>\n",
       "      <td>[employees, inweek_volume, SP500_lastweek_retu...</td>\n",
       "      <td>RFModel</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>in-week underperformance</td>\n",
       "      <td>0.740586</td>\n",
       "      <td>0.755230</td>\n",
       "      <td>[employees, firstday_volume, inweek_volume, SP...</td>\n",
       "      <td>RFModel</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>in-month underperformance</td>\n",
       "      <td>0.734513</td>\n",
       "      <td>0.761062</td>\n",
       "      <td>[employees, firstday_volume, inweek_volume, SP...</td>\n",
       "      <td>RFModel</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>intra day underperformance</td>\n",
       "      <td>0.789030</td>\n",
       "      <td>0.770042</td>\n",
       "      <td>[sector_Other, sector_Health Care, sector_Real...</td>\n",
       "      <td>XGBoost</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>in-week underperformance</td>\n",
       "      <td>0.799163</td>\n",
       "      <td>0.736402</td>\n",
       "      <td>[sector_Other, sector_Basic Industries, sector...</td>\n",
       "      <td>XGBoost</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>in-month underperformance</td>\n",
       "      <td>0.690265</td>\n",
       "      <td>0.730088</td>\n",
       "      <td>[sector_Other, sector_Real Estate, sector_Heal...</td>\n",
       "      <td>XGBoost</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>intra day underperformance</td>\n",
       "      <td>0.544304</td>\n",
       "      <td>0.668776</td>\n",
       "      <td>[sector_Real Estate, sector_Other, employees, ...</td>\n",
       "      <td>Logreg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>in-week underperformance</td>\n",
       "      <td>0.531381</td>\n",
       "      <td>0.654812</td>\n",
       "      <td>[sector_Other, sector_Transportation, sector_B...</td>\n",
       "      <td>Logreg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>in-month underperformance</td>\n",
       "      <td>0.623894</td>\n",
       "      <td>0.679204</td>\n",
       "      <td>[sector_Other, sector_Basic Materials, sector_...</td>\n",
       "      <td>Logreg</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   pred_label    recall  balanced_accuracy  \\\n",
       "0  intra day underperformance  0.755274           0.791139   \n",
       "1    in-week underperformance  0.740586           0.755230   \n",
       "2   in-month underperformance  0.734513           0.761062   \n",
       "0  intra day underperformance  0.789030           0.770042   \n",
       "1    in-week underperformance  0.799163           0.736402   \n",
       "2   in-month underperformance  0.690265           0.730088   \n",
       "0  intra day underperformance  0.544304           0.668776   \n",
       "1    in-week underperformance  0.531381           0.654812   \n",
       "2   in-month underperformance  0.623894           0.679204   \n",
       "\n",
       "                                      top_5_features    model  \n",
       "0  [employees, inweek_volume, SP500_lastweek_retu...  RFModel  \n",
       "1  [employees, firstday_volume, inweek_volume, SP...  RFModel  \n",
       "2  [employees, firstday_volume, inweek_volume, SP...  RFModel  \n",
       "0  [sector_Other, sector_Health Care, sector_Real...  XGBoost  \n",
       "1  [sector_Other, sector_Basic Industries, sector...  XGBoost  \n",
       "2  [sector_Other, sector_Real Estate, sector_Heal...  XGBoost  \n",
       "0  [sector_Real Estate, sector_Other, employees, ...   Logreg  \n",
       "1  [sector_Other, sector_Transportation, sector_B...   Logreg  \n",
       "2  [sector_Other, sector_Basic Materials, sector_...   Logreg  "
      ]
     },
     "execution_count": 406,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Result aggregation\n",
    "from sklearn.metrics import recall_score\n",
    "\n",
    "#RF results\n",
    "rf_results = pd.DataFrame({'pred_label':[], 'recall': [], 'balanced_accuracy': [], 'top_5_features': []})\n",
    "rf_results['model'] = ['RFModel']*3\n",
    "rf_results.recall = [recall_score(y_d_test, rf_predictions_d), recall_score(y_w_test, rf_predictions_w), recall_score(y_m_test, rf_predictions_m)]\n",
    "rf_results.balanced_accuracy = [balanced_accuracy_score(y_d_test, rf_predictions_d), balanced_accuracy_score(y_w_test, rf_predictions_w), balanced_accuracy_score(y_m_test, rf_predictions_m)]\n",
    "rf_results.top_5_features = [rf_d_importance.nlargest(15).index.tolist(), rf_w_importance.nlargest(15).index.tolist(), rf_m_importance.nlargest(15).index.tolist()]\n",
    "rf_results.pred_label = ['intra day underperformance', 'in-week underperformance', 'in-month underperformance']\n",
    "\n",
    "\n",
    "#XGB results\n",
    "xgb_results = pd.DataFrame({'pred_label':[], 'recall': [], 'balanced_accuracy': [], 'top_5_features': []})\n",
    "xgb_results['model'] = ['XGBoost']*3\n",
    "xgb_results.recall = [recall_score(y_d_test, xgb_predictions_d), recall_score(y_w_test, xgb_predictions_w), recall_score(y_m_test, xgb_predictions_m)]\n",
    "xgb_results.balanced_accuracy = [balanced_accuracy_score(y_d_test, xgb_predictions_d), balanced_accuracy_score(y_w_test, xgb_predictions_w), balanced_accuracy_score(y_m_test, xgb_predictions_m)]\n",
    "xgb_results.top_5_features = [xgb_d_importance.nlargest(15).index.tolist(), xgb_w_importance.nlargest(15).index.tolist(), xgb_m_importance.nlargest(15).index.tolist()]\n",
    "xgb_results.pred_label = ['intra day underperformance', 'in-week underperformance', 'in-month underperformance']\n",
    "\n",
    "\n",
    "#Logreg results\n",
    "log_results = pd.DataFrame({'pred_label':[], 'recall': [], 'balanced_accuracy': [], 'top_5_features': []})\n",
    "log_results['model'] = ['Logreg']*3\n",
    "log_results.recall = [recall_score(y_d_test, log_pred_d), recall_score(y_w_test, log_pred_w), recall_score(y_m_test, log_pred_m)]\n",
    "log_results.balanced_accuracy = [balanced_accuracy_score(y_d_test, log_pred_d), balanced_accuracy_score(y_w_test, log_pred_w), balanced_accuracy_score(y_m_test, log_pred_m)]\n",
    "log_results.top_5_features = [log_d_importance.nlargest(15).index.tolist(), log_w_importance.nlargest(15).index.tolist(), log_m_importance.nlargest(15).index.tolist()]\n",
    "log_results.pred_label = ['intra day underperformance', 'in-week underperformance', 'in-month underperformance']\n",
    "\n",
    "\n",
    "res = pd.concat([rf_results, xgb_results, log_results], axis = 0)\n",
    "res"
   ]
  }
 ],
 "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": "3.9.13"
  },
  "vscode": {
   "interpreter": {
    "hash": "1d1ca3f52486ff76943f2aff3c5b8c36b0f44c2d540260a13bd931443a967605"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}