Stock_Market_Forecast.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "864873e8",
   "metadata": {},
   "source": [
    "## Setup Environment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "269f6c1f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set Imports\n",
    "import pandas as pd\n",
    "from pandas_datareader import data, wb\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%matplotlib inline\n",
    "pd.set_option('display.max_colwidth', 200)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "111a924f",
   "metadata": {},
   "source": [
    "## Data Exploration (Small Sample Size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a785246a",
   "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>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Open</th>\n",
       "      <th>Close</th>\n",
       "      <th>Volume</th>\n",
       "      <th>Adj Close</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</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>2010-01-04</th>\n",
       "      <td>113.389999</td>\n",
       "      <td>111.510002</td>\n",
       "      <td>112.370003</td>\n",
       "      <td>113.330002</td>\n",
       "      <td>118944600</td>\n",
       "      <td>113.330002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-01-05</th>\n",
       "      <td>113.680000</td>\n",
       "      <td>112.849998</td>\n",
       "      <td>113.260002</td>\n",
       "      <td>113.629997</td>\n",
       "      <td>111579900</td>\n",
       "      <td>113.629997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-01-06</th>\n",
       "      <td>113.989998</td>\n",
       "      <td>113.430000</td>\n",
       "      <td>113.519997</td>\n",
       "      <td>113.709999</td>\n",
       "      <td>116074400</td>\n",
       "      <td>113.709999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-01-07</th>\n",
       "      <td>114.330002</td>\n",
       "      <td>113.180000</td>\n",
       "      <td>113.500000</td>\n",
       "      <td>114.190002</td>\n",
       "      <td>131091100</td>\n",
       "      <td>114.190002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-01-08</th>\n",
       "      <td>114.620003</td>\n",
       "      <td>113.660004</td>\n",
       "      <td>113.889999</td>\n",
       "      <td>114.570000</td>\n",
       "      <td>126402800</td>\n",
       "      <td>114.570000</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-24</th>\n",
       "      <td>193.529999</td>\n",
       "      <td>189.320007</td>\n",
       "      <td>190.630005</td>\n",
       "      <td>193.199997</td>\n",
       "      <td>150812152</td>\n",
       "      <td>193.199997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-25</th>\n",
       "      <td>195.550003</td>\n",
       "      <td>192.830002</td>\n",
       "      <td>193.729996</td>\n",
       "      <td>195.539993</td>\n",
       "      <td>110728317</td>\n",
       "      <td>195.539993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-26</th>\n",
       "      <td>196.679993</td>\n",
       "      <td>194.899994</td>\n",
       "      <td>196.570007</td>\n",
       "      <td>195.089203</td>\n",
       "      <td>129833663</td>\n",
       "      <td>195.089203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-29</th>\n",
       "      <td>196.229996</td>\n",
       "      <td>193.330002</td>\n",
       "      <td>195.110001</td>\n",
       "      <td>193.559998</td>\n",
       "      <td>125918102</td>\n",
       "      <td>193.559998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-03-01</th>\n",
       "      <td>198.210007</td>\n",
       "      <td>194.454193</td>\n",
       "      <td>195.009995</td>\n",
       "      <td>198.110001</td>\n",
       "      <td>141799687</td>\n",
       "      <td>198.110001</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1550 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  High         Low        Open       Close     Volume  \\\n",
       "Date                                                                    \n",
       "2010-01-04  113.389999  111.510002  112.370003  113.330002  118944600   \n",
       "2010-01-05  113.680000  112.849998  113.260002  113.629997  111579900   \n",
       "2010-01-06  113.989998  113.430000  113.519997  113.709999  116074400   \n",
       "2010-01-07  114.330002  113.180000  113.500000  114.190002  131091100   \n",
       "2010-01-08  114.620003  113.660004  113.889999  114.570000  126402800   \n",
       "...                ...         ...         ...         ...        ...   \n",
       "2016-02-24  193.529999  189.320007  190.630005  193.199997  150812152   \n",
       "2016-02-25  195.550003  192.830002  193.729996  195.539993  110728317   \n",
       "2016-02-26  196.679993  194.899994  196.570007  195.089203  129833663   \n",
       "2016-02-29  196.229996  193.330002  195.110001  193.559998  125918102   \n",
       "2016-03-01  198.210007  194.454193  195.009995  198.110001  141799687   \n",
       "\n",
       "             Adj Close  \n",
       "Date                    \n",
       "2010-01-04  113.330002  \n",
       "2010-01-05  113.629997  \n",
       "2010-01-06  113.709999  \n",
       "2010-01-07  114.190002  \n",
       "2010-01-08  114.570000  \n",
       "...                ...  \n",
       "2016-02-24  193.199997  \n",
       "2016-02-25  195.539993  \n",
       "2016-02-26  195.089203  \n",
       "2016-02-29  193.559998  \n",
       "2016-03-01  198.110001  \n",
       "\n",
       "[1550 rows x 6 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Gather data from SPY ETF: this instument represents the stocks of the S&P 500\n",
    "import pandas_datareader as pdr\n",
    "\n",
    "start_date = pd.to_datetime('2010-01-01')\n",
    "stop_date = pd.to_datetime('2016-03-01')\n",
    "\n",
    "spy = pdr.data.get_data_yahoo('SPY', start_date, stop_date)\n",
    "\n",
    "display(spy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "848afdc5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'SPY')"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot data\n",
    "spy_c = spy['Close']\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(15,10))\n",
    "spy_c.plot(color='k')\n",
    "plt.title(\"SPY\", fontsize=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "75b01cd1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "First Open: 112.37000274658203\n",
      "Last Close: 198.11000061035156\n",
      "Delta over period of time: 85.73999786376953\n"
     ]
    }
   ],
   "source": [
    "# Pull data for data analysis purposes\n",
    "first_open = spy['Open'].iloc[0]\n",
    "print(\"First Open: \" + str(first_open))\n",
    "\n",
    "last_close = spy['Close'].iloc[-1]\n",
    "print(\"Last Close: \" + str(last_close))\n",
    "\n",
    "print(\"Delta over period of time: \" + str(last_close - first_open))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "b7f431d7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Date\n",
      "2010-01-04    0.959999\n",
      "2010-01-05    0.369995\n",
      "2010-01-06    0.190002\n",
      "2010-01-07    0.690002\n",
      "2010-01-08    0.680000\n",
      "                ...   \n",
      "2016-02-24    2.569992\n",
      "2016-02-25    1.809998\n",
      "2016-02-26   -1.480804\n",
      "2016-02-29   -1.550003\n",
      "2016-03-01    3.100006\n",
      "Name: Daily Change, Length: 1550, dtype: float64\n",
      "\n",
      "Sum of Daily Change: 40.293968200683594\n",
      "Standard Deviation of Daily Change: 1.1454417328273667\n"
     ]
    }
   ],
   "source": [
    "# Analyze intraday deltas\n",
    "import numpy as np\n",
    "\n",
    "spy['Daily Change'] = pd.Series(spy['Close'] - spy['Open'])\n",
    "print(spy['Daily Change'])\n",
    "\n",
    "daily_change_sum = spy['Daily Change'].sum()\n",
    "daily_change_std_dev = np.std(spy['Daily Change'])\n",
    "\n",
    "print(\"\\n\" + \"Sum of Daily Change: \" + str(daily_change_sum))\n",
    "print(\"Standard Deviation of Daily Change: \" + str(daily_change_std_dev))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "7b8ca3f8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Date\n",
      "2010-01-04         NaN\n",
      "2010-01-05   -0.070000\n",
      "2010-01-06   -0.110001\n",
      "2010-01-07   -0.209999\n",
      "2010-01-08   -0.300003\n",
      "                ...   \n",
      "2016-02-24   -1.690002\n",
      "2016-02-25    0.529999\n",
      "2016-02-26    1.030014\n",
      "2016-02-29    0.020798\n",
      "2016-03-01    1.449997\n",
      "Name: Overnight Change, Length: 1550, dtype: float64\n",
      "\n",
      "Sum of Overnight Change: 45.44602966308594\n",
      "Standard Deviation of Overnight Change: 0.9518729315831574\n"
     ]
    }
   ],
   "source": [
    "# Analyze overnight deltas\n",
    "spy['Overnight Change'] = pd.Series(spy['Open'] - spy['Close'].shift(1))\n",
    "print(spy['Overnight Change'])\n",
    "\n",
    "overnight_sum = spy['Overnight Change'].sum()\n",
    "overnight_std_dev = np.std(spy['Overnight Change'])\n",
    "\n",
    "print(\"\\n\" + \"Sum of Overnight Change: \" + str(overnight_sum))\n",
    "print(\"Standard Deviation of Overnight Change: \" + str(overnight_std_dev))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f5873230",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Daily Upside Change: 0.7818367909880839\n",
      "Average Overnight Upside Change: 0.6346182933265065\n",
      "\n",
      "Average Daily Downside Change: -0.9058683201886606\n",
      "Average Overnight Downside Change: -0.6641292837668926\n"
     ]
    }
   ],
   "source": [
    "# Compare the average change on downside days vs upside days\n",
    "daily_upside_change = spy[spy['Daily Change'] > 0]['Daily Change'].mean()\n",
    "overnight_upside_change = spy[spy['Overnight Change'] > 0]['Overnight Change'].mean()\n",
    "\n",
    "daily_downside_change = spy[spy['Daily Change'] < 0]['Daily Change'].mean()\n",
    "overnight_downside_change = spy[spy['Overnight Change'] < 0]['Overnight Change'].mean()\n",
    "\n",
    "print(\"Average Daily Upside Change: \" + str(daily_upside_change))\n",
    "print(\"Average Overnight Upside Change: \" + str(overnight_upside_change))\n",
    "print(\"\\nAverage Daily Downside Change: \" + str(daily_downside_change))\n",
    "print(\"Average Overnight Downside Change: \" + str(overnight_downside_change))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "83a63292",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Date\n",
      "2010-01-04         NaN\n",
      "2010-01-05    0.264710\n",
      "2010-01-06    0.070406\n",
      "2010-01-07    0.422129\n",
      "2010-01-08    0.332776\n",
      "                ...   \n",
      "2016-02-24    0.457565\n",
      "2016-02-25    1.211178\n",
      "2016-02-26   -0.230536\n",
      "2016-02-29   -0.783849\n",
      "2016-03-01    2.350694\n",
      "Name: Close, Length: 1550, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Calculate returns\n",
    "## Daily returns\n",
    "daily_rtn = ((spy['Close'] - spy['Close'].shift(1)) / spy['Close'].shift(1)) * 100\n",
    "\n",
    "## Intraday returns\n",
    "id_rtn = ((spy['Close'] - spy['Open']) / spy['Open']) * 100\n",
    "\n",
    "## Overnight returns\n",
    "on_rtn = ((spy['Open'] - spy['Close'].shift(1)) / spy['Close'].shift(1)) * 100\n",
    "\n",
    "print(daily_rtn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "cd234d3a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create function to analyze Sharpe Ratio for all strategies\n",
    "def get_stats(s, n=252):\n",
    "    s = s.dropna()\n",
    "    wins = len(s[s>0])\n",
    "    losses = len(s[s<0])\n",
    "    evens = len(s[s==0])\n",
    "    mean_w = round(s[s>0].mean(), 3)\n",
    "    mean_l = round(s[s<0].mean(), 3)\n",
    "    win_r = round(wins/losses, 3)\n",
    "    mean_trd = round(s.mean(), 3)\n",
    "    sd = round(np.std(s), 3)\n",
    "    max_w = round(s.max(), 3)\n",
    "    max_l = round(s.min(), 3)\n",
    "    sharpe_r = round((s.mean()/np.std(s)) * np.sqrt(n), 4)\n",
    "    cnt = len(s)\n",
    "    print(\"Trades: \", cnt, \\\n",
    "          \"\\nWins: \", wins, \\\n",
    "          \"\\nLosses: \", losses, \\\n",
    "          \"\\nBreakeven: \", evens, \\\n",
    "          \"\\nWin/Loss Ratio: \", win_r, \\\n",
    "          \"\\nMean Win: \", mean_w, \\\n",
    "          \"\\nMean Loss: \", mean_l, \\\n",
    "          \"\\nMean: \", mean_trd, \\\n",
    "          \"\\nStd Dev: \", sd, \\\n",
    "          \"\\nMax Win: \", max_w, \\\n",
    "          \"\\nMax Loss: \", max_l, \\\n",
    "          \"\\nSharpe Ratio: \", sharpe_r\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "92739b22",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trades:  1549 \n",
      "Wins:  847 \n",
      "Losses:  699 \n",
      "Breakeven:  3 \n",
      "Win/Loss Ratio:  1.212 \n",
      "Mean Win:  0.688 \n",
      "Mean Loss:  -0.742 \n",
      "Mean:  0.041 \n",
      "Std Dev:  1.007 \n",
      "Max Win:  4.65 \n",
      "Max Loss:  -6.512 \n",
      "Sharpe Ratio:  0.6485\n"
     ]
    }
   ],
   "source": [
    "# Get stats for Daily Returns\n",
    "get_stats(daily_rtn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "5cb3a653",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trades:  1550 \n",
      "Wins:  851 \n",
      "Losses:  690 \n",
      "Breakeven:  9 \n",
      "Win/Loss Ratio:  1.233 \n",
      "Mean Win:  0.515 \n",
      "Mean Loss:  -0.59 \n",
      "Mean:  0.02 \n",
      "Std Dev:  0.757 \n",
      "Max Win:  3.683 \n",
      "Max Loss:  -4.175 \n",
      "Sharpe Ratio:  0.4162\n"
     ]
    }
   ],
   "source": [
    "# Get stats for Intraday Returns\n",
    "get_stats(id_rtn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "95ff1960",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trades:  1549 \n",
      "Wins:  823 \n",
      "Losses:  718 \n",
      "Breakeven:  8 \n",
      "Win/Loss Ratio:  1.146 \n",
      "Mean Win:  0.421 \n",
      "Mean Loss:  -0.436 \n",
      "Mean:  0.022 \n",
      "Std Dev:  0.629 \n",
      "Max Win:  4.09 \n",
      "Max Loss:  -5.131 \n",
      "Sharpe Ratio:  0.5453\n"
     ]
    }
   ],
   "source": [
    "# Get stats for Overnight Returns\n",
    "get_stats(on_rtn)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f438d10f",
   "metadata": {},
   "source": [
    "## Gather Model Data (Extended Sample Size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "fe1f741a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Extending the anlaysis period\n",
    "start_date = pd.to_datetime('2000-01-01')\n",
    "stop_date = pd.to_datetime('2016-03-01')\n",
    "\n",
    "sp = pdr.data.get_data_yahoo('SPY', start_date, stop_date)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "49dea4fb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'SPY')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot extended data\n",
    "fig, ax = plt.subplots(figsize=(15,10))\n",
    "sp['Close'].plot(color='k')\n",
    "plt.title(\"SPY\", fontsize=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "70574955",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate longer period returns\n",
    "long_day_rtn = ((sp['Close'] - sp['Close'].shift(1)) / sp['Close'].shift(1)) * 100\n",
    "long_id_rtn = ((sp['Close'] - sp['Open']) / sp['Open']) * 100\n",
    "long_on_rtn = ((sp['Open'] - sp['Close'].shift(1)) / sp['Close'].shift(1)) * 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "1a9cbce4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Close to Close:  53.60099792480469 \n",
      "Open to Close:  53.73396301269531 \n",
      "Close to Open: -53.73396301269531\n"
     ]
    }
   ],
   "source": [
    "# Calculate point totals for each strategy\n",
    "## Close to Close (Daily)\n",
    "c_to_c = (sp['Close'] - sp['Close'].shift(1)).sum()\n",
    "\n",
    "## Open to Close (Intraday)\n",
    "o_to_c = (sp['Close'] - sp['Open']).sum()\n",
    "\n",
    "## Close to Open (Overnight)\n",
    "c_to_o = (sp['Open'] - sp['Close']).sum()\n",
    "\n",
    "print(\"Close to Close: \", c_to_c, \\\n",
    "      \"\\nOpen to Close: \", o_to_c, \\\n",
    "      \"\\nClose to Open:\", c_to_o\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "678e4f6d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trades:  2054 \n",
      "Wins:  1113 \n",
      "Losses:  938 \n",
      "Breakeven:  3 \n",
      "Win/Loss Ratio:  1.187 \n",
      "Mean Win:  0.851 \n",
      "Mean Loss:  -0.955 \n",
      "Mean:  0.025 \n",
      "Std Dev:  1.388 \n",
      "Max Win:  11.56 \n",
      "Max Loss:  -9.012 \n",
      "Sharpe Ratio:  0.286\n"
     ]
    }
   ],
   "source": [
    "# Get stats for Long Daily Retuns\n",
    "get_stats(long_day_rtn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "6c1f372c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trades:  2055 \n",
      "Wins:  990 \n",
      "Losses:  800 \n",
      "Breakeven:  265 \n",
      "Win/Loss Ratio:  1.238 \n",
      "Mean Win:  0.57 \n",
      "Mean Loss:  -0.647 \n",
      "Mean:  0.023 \n",
      "Std Dev:  0.802 \n",
      "Max Win:  4.42 \n",
      "Max Loss:  -4.625 \n",
      "Sharpe Ratio:  0.4519\n"
     ]
    }
   ],
   "source": [
    "# Get stats for Long Intraday Retuns\n",
    "get_stats(long_id_rtn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "196aa699",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trades:  2054 \n",
      "Wins:  1090 \n",
      "Losses:  956 \n",
      "Breakeven:  8 \n",
      "Win/Loss Ratio:  1.14 \n",
      "Mean Win:  0.585 \n",
      "Mean Loss:  -0.664 \n",
      "Mean:  0.002 \n",
      "Std Dev:  1.101 \n",
      "Max Win:  11.56 \n",
      "Max Loss:  -9.012 \n",
      "Sharpe Ratio:  0.0263\n"
     ]
    }
   ],
   "source": [
    "# Get stats for Long Overnight Retuns\n",
    "get_stats(long_on_rtn)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4fdf7f39",
   "metadata": {},
   "source": [
    "## Build Model with Support Vector Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "bf61c7a6",
   "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>Close</th>\n",
       "      <th>Close Minus 1</th>\n",
       "      <th>Close Minus 2</th>\n",
       "      <th>Close Minus 3</th>\n",
       "      <th>Close Minus 4</th>\n",
       "      <th>Close Minus 5</th>\n",
       "      <th>Close Minus 6</th>\n",
       "      <th>Close Minus 7</th>\n",
       "      <th>Close Minus 8</th>\n",
       "      <th>Close Minus 9</th>\n",
       "      <th>...</th>\n",
       "      <th>Close Minus 11</th>\n",
       "      <th>Close Minus 12</th>\n",
       "      <th>Close Minus 13</th>\n",
       "      <th>Close Minus 14</th>\n",
       "      <th>Close Minus 15</th>\n",
       "      <th>Close Minus 16</th>\n",
       "      <th>Close Minus 17</th>\n",
       "      <th>Close Minus 18</th>\n",
       "      <th>Close Minus 19</th>\n",
       "      <th>Close Minus 20</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</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>2008-01-31</th>\n",
       "      <td>137.806000</td>\n",
       "      <td>135.514008</td>\n",
       "      <td>136.153000</td>\n",
       "      <td>135.315002</td>\n",
       "      <td>132.981003</td>\n",
       "      <td>135.121002</td>\n",
       "      <td>133.774994</td>\n",
       "      <td>130.966003</td>\n",
       "      <td>132.427002</td>\n",
       "      <td>133.229004</td>\n",
       "      <td>...</td>\n",
       "      <td>137.966995</td>\n",
       "      <td>141.488998</td>\n",
       "      <td>139.968994</td>\n",
       "      <td>141.889999</td>\n",
       "      <td>140.768997</td>\n",
       "      <td>138.867996</td>\n",
       "      <td>141.416000</td>\n",
       "      <td>140.960007</td>\n",
       "      <td>144.507996</td>\n",
       "      <td>144.509003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-01</th>\n",
       "      <td>139.488007</td>\n",
       "      <td>137.806000</td>\n",
       "      <td>135.514008</td>\n",
       "      <td>136.153000</td>\n",
       "      <td>135.315002</td>\n",
       "      <td>132.981003</td>\n",
       "      <td>135.121002</td>\n",
       "      <td>133.774994</td>\n",
       "      <td>130.966003</td>\n",
       "      <td>132.427002</td>\n",
       "      <td>...</td>\n",
       "      <td>137.210999</td>\n",
       "      <td>137.966995</td>\n",
       "      <td>141.488998</td>\n",
       "      <td>139.968994</td>\n",
       "      <td>141.889999</td>\n",
       "      <td>140.768997</td>\n",
       "      <td>138.867996</td>\n",
       "      <td>141.416000</td>\n",
       "      <td>140.960007</td>\n",
       "      <td>144.507996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-04</th>\n",
       "      <td>138.033005</td>\n",
       "      <td>139.488007</td>\n",
       "      <td>137.806000</td>\n",
       "      <td>135.514008</td>\n",
       "      <td>136.153000</td>\n",
       "      <td>135.315002</td>\n",
       "      <td>132.981003</td>\n",
       "      <td>135.121002</td>\n",
       "      <td>133.774994</td>\n",
       "      <td>130.966003</td>\n",
       "      <td>...</td>\n",
       "      <td>133.229004</td>\n",
       "      <td>137.210999</td>\n",
       "      <td>137.966995</td>\n",
       "      <td>141.488998</td>\n",
       "      <td>139.968994</td>\n",
       "      <td>141.889999</td>\n",
       "      <td>140.768997</td>\n",
       "      <td>138.867996</td>\n",
       "      <td>141.416000</td>\n",
       "      <td>140.960007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-05</th>\n",
       "      <td>133.636993</td>\n",
       "      <td>138.033005</td>\n",
       "      <td>139.488007</td>\n",
       "      <td>137.806000</td>\n",
       "      <td>135.514008</td>\n",
       "      <td>136.153000</td>\n",
       "      <td>135.315002</td>\n",
       "      <td>132.981003</td>\n",
       "      <td>135.121002</td>\n",
       "      <td>133.774994</td>\n",
       "      <td>...</td>\n",
       "      <td>132.427002</td>\n",
       "      <td>133.229004</td>\n",
       "      <td>137.210999</td>\n",
       "      <td>137.966995</td>\n",
       "      <td>141.488998</td>\n",
       "      <td>139.968994</td>\n",
       "      <td>141.889999</td>\n",
       "      <td>140.768997</td>\n",
       "      <td>138.867996</td>\n",
       "      <td>141.416000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-06</th>\n",
       "      <td>132.673996</td>\n",
       "      <td>133.636993</td>\n",
       "      <td>138.033005</td>\n",
       "      <td>139.488007</td>\n",
       "      <td>137.806000</td>\n",
       "      <td>135.514008</td>\n",
       "      <td>136.153000</td>\n",
       "      <td>135.315002</td>\n",
       "      <td>132.981003</td>\n",
       "      <td>135.121002</td>\n",
       "      <td>...</td>\n",
       "      <td>130.966003</td>\n",
       "      <td>132.427002</td>\n",
       "      <td>133.229004</td>\n",
       "      <td>137.210999</td>\n",
       "      <td>137.966995</td>\n",
       "      <td>141.488998</td>\n",
       "      <td>139.968994</td>\n",
       "      <td>141.889999</td>\n",
       "      <td>140.768997</td>\n",
       "      <td>138.867996</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>2016-02-24</th>\n",
       "      <td>193.199997</td>\n",
       "      <td>192.320007</td>\n",
       "      <td>194.779999</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>192.089996</td>\n",
       "      <td>192.880005</td>\n",
       "      <td>189.779999</td>\n",
       "      <td>186.630005</td>\n",
       "      <td>182.860001</td>\n",
       "      <td>185.270004</td>\n",
       "      <td>...</td>\n",
       "      <td>185.419998</td>\n",
       "      <td>187.949997</td>\n",
       "      <td>191.600006</td>\n",
       "      <td>191.300003</td>\n",
       "      <td>190.160004</td>\n",
       "      <td>193.649994</td>\n",
       "      <td>193.720795</td>\n",
       "      <td>189.110001</td>\n",
       "      <td>188.130005</td>\n",
       "      <td>190.199997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-25</th>\n",
       "      <td>195.539993</td>\n",
       "      <td>193.199997</td>\n",
       "      <td>192.320007</td>\n",
       "      <td>194.779999</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>192.089996</td>\n",
       "      <td>192.880005</td>\n",
       "      <td>189.779999</td>\n",
       "      <td>186.630005</td>\n",
       "      <td>182.860001</td>\n",
       "      <td>...</td>\n",
       "      <td>185.429993</td>\n",
       "      <td>185.419998</td>\n",
       "      <td>187.949997</td>\n",
       "      <td>191.600006</td>\n",
       "      <td>191.300003</td>\n",
       "      <td>190.160004</td>\n",
       "      <td>193.649994</td>\n",
       "      <td>193.720795</td>\n",
       "      <td>189.110001</td>\n",
       "      <td>188.130005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-26</th>\n",
       "      <td>195.089203</td>\n",
       "      <td>195.539993</td>\n",
       "      <td>193.199997</td>\n",
       "      <td>192.320007</td>\n",
       "      <td>194.779999</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>192.089996</td>\n",
       "      <td>192.880005</td>\n",
       "      <td>189.779999</td>\n",
       "      <td>186.630005</td>\n",
       "      <td>...</td>\n",
       "      <td>185.270004</td>\n",
       "      <td>185.429993</td>\n",
       "      <td>185.419998</td>\n",
       "      <td>187.949997</td>\n",
       "      <td>191.600006</td>\n",
       "      <td>191.300003</td>\n",
       "      <td>190.160004</td>\n",
       "      <td>193.649994</td>\n",
       "      <td>193.720795</td>\n",
       "      <td>189.110001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-29</th>\n",
       "      <td>193.559998</td>\n",
       "      <td>195.089203</td>\n",
       "      <td>195.539993</td>\n",
       "      <td>193.199997</td>\n",
       "      <td>192.320007</td>\n",
       "      <td>194.779999</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>192.089996</td>\n",
       "      <td>192.880005</td>\n",
       "      <td>189.779999</td>\n",
       "      <td>...</td>\n",
       "      <td>182.860001</td>\n",
       "      <td>185.270004</td>\n",
       "      <td>185.429993</td>\n",
       "      <td>185.419998</td>\n",
       "      <td>187.949997</td>\n",
       "      <td>191.600006</td>\n",
       "      <td>191.300003</td>\n",
       "      <td>190.160004</td>\n",
       "      <td>193.649994</td>\n",
       "      <td>193.720795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-03-01</th>\n",
       "      <td>198.110001</td>\n",
       "      <td>193.559998</td>\n",
       "      <td>195.089203</td>\n",
       "      <td>195.539993</td>\n",
       "      <td>193.199997</td>\n",
       "      <td>192.320007</td>\n",
       "      <td>194.779999</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>192.089996</td>\n",
       "      <td>192.880005</td>\n",
       "      <td>...</td>\n",
       "      <td>186.630005</td>\n",
       "      <td>182.860001</td>\n",
       "      <td>185.270004</td>\n",
       "      <td>185.429993</td>\n",
       "      <td>185.419998</td>\n",
       "      <td>187.949997</td>\n",
       "      <td>191.600006</td>\n",
       "      <td>191.300003</td>\n",
       "      <td>190.160004</td>\n",
       "      <td>193.649994</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2035 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Close  Close Minus 1  Close Minus 2  Close Minus 3  \\\n",
       "Date                                                                  \n",
       "2008-01-31  137.806000     135.514008     136.153000     135.315002   \n",
       "2008-02-01  139.488007     137.806000     135.514008     136.153000   \n",
       "2008-02-04  138.033005     139.488007     137.806000     135.514008   \n",
       "2008-02-05  133.636993     138.033005     139.488007     137.806000   \n",
       "2008-02-06  132.673996     133.636993     138.033005     139.488007   \n",
       "...                ...            ...            ...            ...   \n",
       "2016-02-24  193.199997     192.320007     194.779999     192.000000   \n",
       "2016-02-25  195.539993     193.199997     192.320007     194.779999   \n",
       "2016-02-26  195.089203     195.539993     193.199997     192.320007   \n",
       "2016-02-29  193.559998     195.089203     195.539993     193.199997   \n",
       "2016-03-01  198.110001     193.559998     195.089203     195.539993   \n",
       "\n",
       "            Close Minus 4  Close Minus 5  Close Minus 6  Close Minus 7  \\\n",
       "Date                                                                     \n",
       "2008-01-31     132.981003     135.121002     133.774994     130.966003   \n",
       "2008-02-01     135.315002     132.981003     135.121002     133.774994   \n",
       "2008-02-04     136.153000     135.315002     132.981003     135.121002   \n",
       "2008-02-05     135.514008     136.153000     135.315002     132.981003   \n",
       "2008-02-06     137.806000     135.514008     136.153000     135.315002   \n",
       "...                   ...            ...            ...            ...   \n",
       "2016-02-24     192.089996     192.880005     189.779999     186.630005   \n",
       "2016-02-25     192.000000     192.089996     192.880005     189.779999   \n",
       "2016-02-26     194.779999     192.000000     192.089996     192.880005   \n",
       "2016-02-29     192.320007     194.779999     192.000000     192.089996   \n",
       "2016-03-01     193.199997     192.320007     194.779999     192.000000   \n",
       "\n",
       "            Close Minus 8  Close Minus 9  ...  Close Minus 11  Close Minus 12  \\\n",
       "Date                                      ...                                   \n",
       "2008-01-31     132.427002     133.229004  ...      137.966995      141.488998   \n",
       "2008-02-01     130.966003     132.427002  ...      137.210999      137.966995   \n",
       "2008-02-04     133.774994     130.966003  ...      133.229004      137.210999   \n",
       "2008-02-05     135.121002     133.774994  ...      132.427002      133.229004   \n",
       "2008-02-06     132.981003     135.121002  ...      130.966003      132.427002   \n",
       "...                   ...            ...  ...             ...             ...   \n",
       "2016-02-24     182.860001     185.270004  ...      185.419998      187.949997   \n",
       "2016-02-25     186.630005     182.860001  ...      185.429993      185.419998   \n",
       "2016-02-26     189.779999     186.630005  ...      185.270004      185.429993   \n",
       "2016-02-29     192.880005     189.779999  ...      182.860001      185.270004   \n",
       "2016-03-01     192.089996     192.880005  ...      186.630005      182.860001   \n",
       "\n",
       "            Close Minus 13  Close Minus 14  Close Minus 15  Close Minus 16  \\\n",
       "Date                                                                         \n",
       "2008-01-31      139.968994      141.889999      140.768997      138.867996   \n",
       "2008-02-01      141.488998      139.968994      141.889999      140.768997   \n",
       "2008-02-04      137.966995      141.488998      139.968994      141.889999   \n",
       "2008-02-05      137.210999      137.966995      141.488998      139.968994   \n",
       "2008-02-06      133.229004      137.210999      137.966995      141.488998   \n",
       "...                    ...             ...             ...             ...   \n",
       "2016-02-24      191.600006      191.300003      190.160004      193.649994   \n",
       "2016-02-25      187.949997      191.600006      191.300003      190.160004   \n",
       "2016-02-26      185.419998      187.949997      191.600006      191.300003   \n",
       "2016-02-29      185.429993      185.419998      187.949997      191.600006   \n",
       "2016-03-01      185.270004      185.429993      185.419998      187.949997   \n",
       "\n",
       "            Close Minus 17  Close Minus 18  Close Minus 19  Close Minus 20  \n",
       "Date                                                                        \n",
       "2008-01-31      141.416000      140.960007      144.507996      144.509003  \n",
       "2008-02-01      138.867996      141.416000      140.960007      144.507996  \n",
       "2008-02-04      140.768997      138.867996      141.416000      140.960007  \n",
       "2008-02-05      141.889999      140.768997      138.867996      141.416000  \n",
       "2008-02-06      139.968994      141.889999      140.768997      138.867996  \n",
       "...                    ...             ...             ...             ...  \n",
       "2016-02-24      193.720795      189.110001      188.130005      190.199997  \n",
       "2016-02-25      193.649994      193.720795      189.110001      188.130005  \n",
       "2016-02-26      190.160004      193.649994      193.720795      189.110001  \n",
       "2016-02-29      191.300003      190.160004      193.649994      193.720795  \n",
       "2016-03-01      191.600006      191.300003      190.160004      193.649994  \n",
       "\n",
       "[2035 rows x 21 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Build df that contains price history for each date (past 20 closes) - predict from one day's close to the next\n",
    "for i in range(1,21,1):\n",
    "    sp.loc[:, 'Close Minus ' + str(i)] = sp['Close'].shift(i)\n",
    "\n",
    "sp20 = sp[[x for x in sp.columns if 'Close Minus' in x or x == 'Close']].iloc[20:]\n",
    "display(sp20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "783a7727",
   "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>Close Minus 20</th>\n",
       "      <th>Close Minus 19</th>\n",
       "      <th>Close Minus 18</th>\n",
       "      <th>Close Minus 17</th>\n",
       "      <th>Close Minus 16</th>\n",
       "      <th>Close Minus 15</th>\n",
       "      <th>Close Minus 14</th>\n",
       "      <th>Close Minus 13</th>\n",
       "      <th>Close Minus 12</th>\n",
       "      <th>Close Minus 11</th>\n",
       "      <th>...</th>\n",
       "      <th>Close Minus 9</th>\n",
       "      <th>Close Minus 8</th>\n",
       "      <th>Close Minus 7</th>\n",
       "      <th>Close Minus 6</th>\n",
       "      <th>Close Minus 5</th>\n",
       "      <th>Close Minus 4</th>\n",
       "      <th>Close Minus 3</th>\n",
       "      <th>Close Minus 2</th>\n",
       "      <th>Close Minus 1</th>\n",
       "      <th>Close</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</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>2008-01-31</th>\n",
       "      <td>144.509003</td>\n",
       "      <td>144.507996</td>\n",
       "      <td>140.960007</td>\n",
       "      <td>141.416000</td>\n",
       "      <td>138.867996</td>\n",
       "      <td>140.768997</td>\n",
       "      <td>141.889999</td>\n",
       "      <td>139.968994</td>\n",
       "      <td>141.488998</td>\n",
       "      <td>137.966995</td>\n",
       "      <td>...</td>\n",
       "      <td>133.229004</td>\n",
       "      <td>132.427002</td>\n",
       "      <td>130.966003</td>\n",
       "      <td>133.774994</td>\n",
       "      <td>135.121002</td>\n",
       "      <td>132.981003</td>\n",
       "      <td>135.315002</td>\n",
       "      <td>136.153000</td>\n",
       "      <td>135.514008</td>\n",
       "      <td>137.806000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-01</th>\n",
       "      <td>144.507996</td>\n",
       "      <td>140.960007</td>\n",
       "      <td>141.416000</td>\n",
       "      <td>138.867996</td>\n",
       "      <td>140.768997</td>\n",
       "      <td>141.889999</td>\n",
       "      <td>139.968994</td>\n",
       "      <td>141.488998</td>\n",
       "      <td>137.966995</td>\n",
       "      <td>137.210999</td>\n",
       "      <td>...</td>\n",
       "      <td>132.427002</td>\n",
       "      <td>130.966003</td>\n",
       "      <td>133.774994</td>\n",
       "      <td>135.121002</td>\n",
       "      <td>132.981003</td>\n",
       "      <td>135.315002</td>\n",
       "      <td>136.153000</td>\n",
       "      <td>135.514008</td>\n",
       "      <td>137.806000</td>\n",
       "      <td>139.488007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-04</th>\n",
       "      <td>140.960007</td>\n",
       "      <td>141.416000</td>\n",
       "      <td>138.867996</td>\n",
       "      <td>140.768997</td>\n",
       "      <td>141.889999</td>\n",
       "      <td>139.968994</td>\n",
       "      <td>141.488998</td>\n",
       "      <td>137.966995</td>\n",
       "      <td>137.210999</td>\n",
       "      <td>133.229004</td>\n",
       "      <td>...</td>\n",
       "      <td>130.966003</td>\n",
       "      <td>133.774994</td>\n",
       "      <td>135.121002</td>\n",
       "      <td>132.981003</td>\n",
       "      <td>135.315002</td>\n",
       "      <td>136.153000</td>\n",
       "      <td>135.514008</td>\n",
       "      <td>137.806000</td>\n",
       "      <td>139.488007</td>\n",
       "      <td>138.033005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-05</th>\n",
       "      <td>141.416000</td>\n",
       "      <td>138.867996</td>\n",
       "      <td>140.768997</td>\n",
       "      <td>141.889999</td>\n",
       "      <td>139.968994</td>\n",
       "      <td>141.488998</td>\n",
       "      <td>137.966995</td>\n",
       "      <td>137.210999</td>\n",
       "      <td>133.229004</td>\n",
       "      <td>132.427002</td>\n",
       "      <td>...</td>\n",
       "      <td>133.774994</td>\n",
       "      <td>135.121002</td>\n",
       "      <td>132.981003</td>\n",
       "      <td>135.315002</td>\n",
       "      <td>136.153000</td>\n",
       "      <td>135.514008</td>\n",
       "      <td>137.806000</td>\n",
       "      <td>139.488007</td>\n",
       "      <td>138.033005</td>\n",
       "      <td>133.636993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-06</th>\n",
       "      <td>138.867996</td>\n",
       "      <td>140.768997</td>\n",
       "      <td>141.889999</td>\n",
       "      <td>139.968994</td>\n",
       "      <td>141.488998</td>\n",
       "      <td>137.966995</td>\n",
       "      <td>137.210999</td>\n",
       "      <td>133.229004</td>\n",
       "      <td>132.427002</td>\n",
       "      <td>130.966003</td>\n",
       "      <td>...</td>\n",
       "      <td>135.121002</td>\n",
       "      <td>132.981003</td>\n",
       "      <td>135.315002</td>\n",
       "      <td>136.153000</td>\n",
       "      <td>135.514008</td>\n",
       "      <td>137.806000</td>\n",
       "      <td>139.488007</td>\n",
       "      <td>138.033005</td>\n",
       "      <td>133.636993</td>\n",
       "      <td>132.673996</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>2016-02-24</th>\n",
       "      <td>190.199997</td>\n",
       "      <td>188.130005</td>\n",
       "      <td>189.110001</td>\n",
       "      <td>193.720795</td>\n",
       "      <td>193.649994</td>\n",
       "      <td>190.160004</td>\n",
       "      <td>191.300003</td>\n",
       "      <td>191.600006</td>\n",
       "      <td>187.949997</td>\n",
       "      <td>185.419998</td>\n",
       "      <td>...</td>\n",
       "      <td>185.270004</td>\n",
       "      <td>182.860001</td>\n",
       "      <td>186.630005</td>\n",
       "      <td>189.779999</td>\n",
       "      <td>192.880005</td>\n",
       "      <td>192.089996</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>194.779999</td>\n",
       "      <td>192.320007</td>\n",
       "      <td>193.199997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-25</th>\n",
       "      <td>188.130005</td>\n",
       "      <td>189.110001</td>\n",
       "      <td>193.720795</td>\n",
       "      <td>193.649994</td>\n",
       "      <td>190.160004</td>\n",
       "      <td>191.300003</td>\n",
       "      <td>191.600006</td>\n",
       "      <td>187.949997</td>\n",
       "      <td>185.419998</td>\n",
       "      <td>185.429993</td>\n",
       "      <td>...</td>\n",
       "      <td>182.860001</td>\n",
       "      <td>186.630005</td>\n",
       "      <td>189.779999</td>\n",
       "      <td>192.880005</td>\n",
       "      <td>192.089996</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>194.779999</td>\n",
       "      <td>192.320007</td>\n",
       "      <td>193.199997</td>\n",
       "      <td>195.539993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-26</th>\n",
       "      <td>189.110001</td>\n",
       "      <td>193.720795</td>\n",
       "      <td>193.649994</td>\n",
       "      <td>190.160004</td>\n",
       "      <td>191.300003</td>\n",
       "      <td>191.600006</td>\n",
       "      <td>187.949997</td>\n",
       "      <td>185.419998</td>\n",
       "      <td>185.429993</td>\n",
       "      <td>185.270004</td>\n",
       "      <td>...</td>\n",
       "      <td>186.630005</td>\n",
       "      <td>189.779999</td>\n",
       "      <td>192.880005</td>\n",
       "      <td>192.089996</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>194.779999</td>\n",
       "      <td>192.320007</td>\n",
       "      <td>193.199997</td>\n",
       "      <td>195.539993</td>\n",
       "      <td>195.089203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-29</th>\n",
       "      <td>193.720795</td>\n",
       "      <td>193.649994</td>\n",
       "      <td>190.160004</td>\n",
       "      <td>191.300003</td>\n",
       "      <td>191.600006</td>\n",
       "      <td>187.949997</td>\n",
       "      <td>185.419998</td>\n",
       "      <td>185.429993</td>\n",
       "      <td>185.270004</td>\n",
       "      <td>182.860001</td>\n",
       "      <td>...</td>\n",
       "      <td>189.779999</td>\n",
       "      <td>192.880005</td>\n",
       "      <td>192.089996</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>194.779999</td>\n",
       "      <td>192.320007</td>\n",
       "      <td>193.199997</td>\n",
       "      <td>195.539993</td>\n",
       "      <td>195.089203</td>\n",
       "      <td>193.559998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-03-01</th>\n",
       "      <td>193.649994</td>\n",
       "      <td>190.160004</td>\n",
       "      <td>191.300003</td>\n",
       "      <td>191.600006</td>\n",
       "      <td>187.949997</td>\n",
       "      <td>185.419998</td>\n",
       "      <td>185.429993</td>\n",
       "      <td>185.270004</td>\n",
       "      <td>182.860001</td>\n",
       "      <td>186.630005</td>\n",
       "      <td>...</td>\n",
       "      <td>192.880005</td>\n",
       "      <td>192.089996</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>194.779999</td>\n",
       "      <td>192.320007</td>\n",
       "      <td>193.199997</td>\n",
       "      <td>195.539993</td>\n",
       "      <td>195.089203</td>\n",
       "      <td>193.559998</td>\n",
       "      <td>198.110001</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2035 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            Close Minus 20  Close Minus 19  Close Minus 18  Close Minus 17  \\\n",
       "Date                                                                         \n",
       "2008-01-31      144.509003      144.507996      140.960007      141.416000   \n",
       "2008-02-01      144.507996      140.960007      141.416000      138.867996   \n",
       "2008-02-04      140.960007      141.416000      138.867996      140.768997   \n",
       "2008-02-05      141.416000      138.867996      140.768997      141.889999   \n",
       "2008-02-06      138.867996      140.768997      141.889999      139.968994   \n",
       "...                    ...             ...             ...             ...   \n",
       "2016-02-24      190.199997      188.130005      189.110001      193.720795   \n",
       "2016-02-25      188.130005      189.110001      193.720795      193.649994   \n",
       "2016-02-26      189.110001      193.720795      193.649994      190.160004   \n",
       "2016-02-29      193.720795      193.649994      190.160004      191.300003   \n",
       "2016-03-01      193.649994      190.160004      191.300003      191.600006   \n",
       "\n",
       "            Close Minus 16  Close Minus 15  Close Minus 14  Close Minus 13  \\\n",
       "Date                                                                         \n",
       "2008-01-31      138.867996      140.768997      141.889999      139.968994   \n",
       "2008-02-01      140.768997      141.889999      139.968994      141.488998   \n",
       "2008-02-04      141.889999      139.968994      141.488998      137.966995   \n",
       "2008-02-05      139.968994      141.488998      137.966995      137.210999   \n",
       "2008-02-06      141.488998      137.966995      137.210999      133.229004   \n",
       "...                    ...             ...             ...             ...   \n",
       "2016-02-24      193.649994      190.160004      191.300003      191.600006   \n",
       "2016-02-25      190.160004      191.300003      191.600006      187.949997   \n",
       "2016-02-26      191.300003      191.600006      187.949997      185.419998   \n",
       "2016-02-29      191.600006      187.949997      185.419998      185.429993   \n",
       "2016-03-01      187.949997      185.419998      185.429993      185.270004   \n",
       "\n",
       "            Close Minus 12  Close Minus 11  ...  Close Minus 9  Close Minus 8  \\\n",
       "Date                                        ...                                 \n",
       "2008-01-31      141.488998      137.966995  ...     133.229004     132.427002   \n",
       "2008-02-01      137.966995      137.210999  ...     132.427002     130.966003   \n",
       "2008-02-04      137.210999      133.229004  ...     130.966003     133.774994   \n",
       "2008-02-05      133.229004      132.427002  ...     133.774994     135.121002   \n",
       "2008-02-06      132.427002      130.966003  ...     135.121002     132.981003   \n",
       "...                    ...             ...  ...            ...            ...   \n",
       "2016-02-24      187.949997      185.419998  ...     185.270004     182.860001   \n",
       "2016-02-25      185.419998      185.429993  ...     182.860001     186.630005   \n",
       "2016-02-26      185.429993      185.270004  ...     186.630005     189.779999   \n",
       "2016-02-29      185.270004      182.860001  ...     189.779999     192.880005   \n",
       "2016-03-01      182.860001      186.630005  ...     192.880005     192.089996   \n",
       "\n",
       "            Close Minus 7  Close Minus 6  Close Minus 5  Close Minus 4  \\\n",
       "Date                                                                     \n",
       "2008-01-31     130.966003     133.774994     135.121002     132.981003   \n",
       "2008-02-01     133.774994     135.121002     132.981003     135.315002   \n",
       "2008-02-04     135.121002     132.981003     135.315002     136.153000   \n",
       "2008-02-05     132.981003     135.315002     136.153000     135.514008   \n",
       "2008-02-06     135.315002     136.153000     135.514008     137.806000   \n",
       "...                   ...            ...            ...            ...   \n",
       "2016-02-24     186.630005     189.779999     192.880005     192.089996   \n",
       "2016-02-25     189.779999     192.880005     192.089996     192.000000   \n",
       "2016-02-26     192.880005     192.089996     192.000000     194.779999   \n",
       "2016-02-29     192.089996     192.000000     194.779999     192.320007   \n",
       "2016-03-01     192.000000     194.779999     192.320007     193.199997   \n",
       "\n",
       "            Close Minus 3  Close Minus 2  Close Minus 1       Close  \n",
       "Date                                                                 \n",
       "2008-01-31     135.315002     136.153000     135.514008  137.806000  \n",
       "2008-02-01     136.153000     135.514008     137.806000  139.488007  \n",
       "2008-02-04     135.514008     137.806000     139.488007  138.033005  \n",
       "2008-02-05     137.806000     139.488007     138.033005  133.636993  \n",
       "2008-02-06     139.488007     138.033005     133.636993  132.673996  \n",
       "...                   ...            ...            ...         ...  \n",
       "2016-02-24     192.000000     194.779999     192.320007  193.199997  \n",
       "2016-02-25     194.779999     192.320007     193.199997  195.539993  \n",
       "2016-02-26     192.320007     193.199997     195.539993  195.089203  \n",
       "2016-02-29     193.199997     195.539993     195.089203  193.559998  \n",
       "2016-03-01     195.539993     195.089203     193.559998  198.110001  \n",
       "\n",
       "[2035 rows x 21 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Reverse columns to have time run left -> right\n",
    "sp20 = sp20.iloc[:, ::-1]\n",
    "display(sp20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "771c63ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import Support Vector Machine (for linear regression purposes) and setup training and test matrices\n",
    "from sklearn.svm import SVR\n",
    "\n",
    "clf = SVR(kernel='linear')\n",
    "x_train = sp20[:-1000]\n",
    "y_train = sp20['Close'].shift(-1)[:-1000]\n",
    "x_test = sp20[-1000:]\n",
    "y_test = sp20['Close'].shift(-1)[-1000:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "3e888f97",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fit model and test out-of-sample data\n",
    "model = clf.fit(x_train, y_train)\n",
    "preds = model.predict(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "4d4e3bfd",
   "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>Next Day Close</th>\n",
       "      <th>Predicted Next Close</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2012-03-09</th>\n",
       "      <td>137.580002</td>\n",
       "      <td>137.665188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-12</th>\n",
       "      <td>140.061996</td>\n",
       "      <td>137.920296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-13</th>\n",
       "      <td>139.910004</td>\n",
       "      <td>139.928730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-14</th>\n",
       "      <td>140.720001</td>\n",
       "      <td>139.900056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-15</th>\n",
       "      <td>140.300003</td>\n",
       "      <td>140.799928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-24</th>\n",
       "      <td>195.539993</td>\n",
       "      <td>192.406784</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-25</th>\n",
       "      <td>195.089203</td>\n",
       "      <td>195.050811</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-26</th>\n",
       "      <td>193.559998</td>\n",
       "      <td>195.204225</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-29</th>\n",
       "      <td>198.110001</td>\n",
       "      <td>193.616204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-03-01</th>\n",
       "      <td>NaN</td>\n",
       "      <td>197.470584</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            Next Day Close  Predicted Next Close\n",
       "Date                                            \n",
       "2012-03-09      137.580002            137.665188\n",
       "2012-03-12      140.061996            137.920296\n",
       "2012-03-13      139.910004            139.928730\n",
       "2012-03-14      140.720001            139.900056\n",
       "2012-03-15      140.300003            140.799928\n",
       "...                    ...                   ...\n",
       "2016-02-24      195.539993            192.406784\n",
       "2016-02-25      195.089203            195.050811\n",
       "2016-02-26      193.559998            195.204225\n",
       "2016-02-29      198.110001            193.616204\n",
       "2016-03-01             NaN            197.470584\n",
       "\n",
       "[1000 rows x 2 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Compare predictions to actuals\n",
    "tf = pd.DataFrame(list(zip(y_test, preds)), columns = ['Next Day Close', 'Predicted Next Close'],\n",
    "                  index=y_test.index)\n",
    "\n",
    "display(tf)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "41a08a8e",
   "metadata": {},
   "source": [
    "## Evaluate Model Performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "e1fa3f92",
   "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>Next Day Close</th>\n",
       "      <th>Predicted Next Close</th>\n",
       "      <th>Current Day Close</th>\n",
       "      <th>Next Day Open</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2012-03-09</th>\n",
       "      <td>137.580002</td>\n",
       "      <td>137.665188</td>\n",
       "      <td>137.570007</td>\n",
       "      <td>137.550003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-12</th>\n",
       "      <td>140.061996</td>\n",
       "      <td>137.920296</td>\n",
       "      <td>137.580002</td>\n",
       "      <td>138.320007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-13</th>\n",
       "      <td>139.910004</td>\n",
       "      <td>139.928730</td>\n",
       "      <td>140.061996</td>\n",
       "      <td>140.100006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-14</th>\n",
       "      <td>140.720001</td>\n",
       "      <td>139.900056</td>\n",
       "      <td>139.910004</td>\n",
       "      <td>140.119995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-15</th>\n",
       "      <td>140.300003</td>\n",
       "      <td>140.799928</td>\n",
       "      <td>140.720001</td>\n",
       "      <td>140.360001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-24</th>\n",
       "      <td>195.539993</td>\n",
       "      <td>192.406784</td>\n",
       "      <td>193.199997</td>\n",
       "      <td>193.729996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-25</th>\n",
       "      <td>195.089203</td>\n",
       "      <td>195.050811</td>\n",
       "      <td>195.539993</td>\n",
       "      <td>196.570007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-26</th>\n",
       "      <td>193.559998</td>\n",
       "      <td>195.204225</td>\n",
       "      <td>195.089203</td>\n",
       "      <td>195.110001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-29</th>\n",
       "      <td>198.110001</td>\n",
       "      <td>193.616204</td>\n",
       "      <td>193.559998</td>\n",
       "      <td>195.009995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-03-01</th>\n",
       "      <td>NaN</td>\n",
       "      <td>197.470584</td>\n",
       "      <td>198.110001</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            Next Day Close  Predicted Next Close  Current Day Close  \\\n",
       "Date                                                                  \n",
       "2012-03-09      137.580002            137.665188         137.570007   \n",
       "2012-03-12      140.061996            137.920296         137.580002   \n",
       "2012-03-13      139.910004            139.928730         140.061996   \n",
       "2012-03-14      140.720001            139.900056         139.910004   \n",
       "2012-03-15      140.300003            140.799928         140.720001   \n",
       "...                    ...                   ...                ...   \n",
       "2016-02-24      195.539993            192.406784         193.199997   \n",
       "2016-02-25      195.089203            195.050811         195.539993   \n",
       "2016-02-26      193.559998            195.204225         195.089203   \n",
       "2016-02-29      198.110001            193.616204         193.559998   \n",
       "2016-03-01             NaN            197.470584         198.110001   \n",
       "\n",
       "            Next Day Open  \n",
       "Date                       \n",
       "2012-03-09     137.550003  \n",
       "2012-03-12     138.320007  \n",
       "2012-03-13     140.100006  \n",
       "2012-03-14     140.119995  \n",
       "2012-03-15     140.360001  \n",
       "...                   ...  \n",
       "2016-02-24     193.729996  \n",
       "2016-02-25     196.570007  \n",
       "2016-02-26     195.110001  \n",
       "2016-02-29     195.009995  \n",
       "2016-03-01            NaN  \n",
       "\n",
       "[1000 rows x 4 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create df to analyze next day close relative to next day open (for buying/selling purposes)\n",
    "cdc = sp[['Close']].iloc[-1000:]\n",
    "ndo = sp[['Open']].iloc[-1000:].shift(-1)\n",
    "\n",
    "tf1 = pd.merge(tf, cdc, left_index = True, right_index = True)\n",
    "tf2 = pd.merge(tf1, ndo, left_index = True, right_index = True)\n",
    "tf2.columns = ['Next Day Close', 'Predicted Next Close', 'Current Day Close', 'Next Day Open']\n",
    "\n",
    "display(tf2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "0f7a2d25",
   "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>Next Day Close</th>\n",
       "      <th>Predicted Next Close</th>\n",
       "      <th>Current Day Close</th>\n",
       "      <th>Next Day Open</th>\n",
       "      <th>Signal</th>\n",
       "      <th>PnL</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</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>2012-03-09</th>\n",
       "      <td>137.580002</td>\n",
       "      <td>137.665188</td>\n",
       "      <td>137.570007</td>\n",
       "      <td>137.550003</td>\n",
       "      <td>1</td>\n",
       "      <td>0.021809</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-12</th>\n",
       "      <td>140.061996</td>\n",
       "      <td>137.920296</td>\n",
       "      <td>137.580002</td>\n",
       "      <td>138.320007</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-13</th>\n",
       "      <td>139.910004</td>\n",
       "      <td>139.928730</td>\n",
       "      <td>140.061996</td>\n",
       "      <td>140.100006</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-14</th>\n",
       "      <td>140.720001</td>\n",
       "      <td>139.900056</td>\n",
       "      <td>139.910004</td>\n",
       "      <td>140.119995</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-03-15</th>\n",
       "      <td>140.300003</td>\n",
       "      <td>140.799928</td>\n",
       "      <td>140.720001</td>\n",
       "      <td>140.360001</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.042745</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-24</th>\n",
       "      <td>195.539993</td>\n",
       "      <td>192.406784</td>\n",
       "      <td>193.199997</td>\n",
       "      <td>193.729996</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-25</th>\n",
       "      <td>195.089203</td>\n",
       "      <td>195.050811</td>\n",
       "      <td>195.539993</td>\n",
       "      <td>196.570007</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-26</th>\n",
       "      <td>193.559998</td>\n",
       "      <td>195.204225</td>\n",
       "      <td>195.089203</td>\n",
       "      <td>195.110001</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.794425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-29</th>\n",
       "      <td>198.110001</td>\n",
       "      <td>193.616204</td>\n",
       "      <td>193.559998</td>\n",
       "      <td>195.009995</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-03-01</th>\n",
       "      <td>NaN</td>\n",
       "      <td>197.470584</td>\n",
       "      <td>198.110001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            Next Day Close  Predicted Next Close  Current Day Close  \\\n",
       "Date                                                                  \n",
       "2012-03-09      137.580002            137.665188         137.570007   \n",
       "2012-03-12      140.061996            137.920296         137.580002   \n",
       "2012-03-13      139.910004            139.928730         140.061996   \n",
       "2012-03-14      140.720001            139.900056         139.910004   \n",
       "2012-03-15      140.300003            140.799928         140.720001   \n",
       "...                    ...                   ...                ...   \n",
       "2016-02-24      195.539993            192.406784         193.199997   \n",
       "2016-02-25      195.089203            195.050811         195.539993   \n",
       "2016-02-26      193.559998            195.204225         195.089203   \n",
       "2016-02-29      198.110001            193.616204         193.559998   \n",
       "2016-03-01             NaN            197.470584         198.110001   \n",
       "\n",
       "            Next Day Open  Signal       PnL  \n",
       "Date                                         \n",
       "2012-03-09     137.550003       1  0.021809  \n",
       "2012-03-12     138.320007       0  0.000000  \n",
       "2012-03-13     140.100006       0  0.000000  \n",
       "2012-03-14     140.119995       0  0.000000  \n",
       "2012-03-15     140.360001       1 -0.042745  \n",
       "...                   ...     ...       ...  \n",
       "2016-02-24     193.729996       0  0.000000  \n",
       "2016-02-25     196.570007       0  0.000000  \n",
       "2016-02-26     195.110001       1 -0.794425  \n",
       "2016-02-29     195.009995       0  0.000000  \n",
       "2016-03-01            NaN       0  0.000000  \n",
       "\n",
       "[1000 rows x 6 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Add signal and profit/loss for the signal\n",
    "def get_signal(r):\n",
    "    if r['Predicted Next Close'] > r['Next Day Open']:\n",
    "        return 1\n",
    "    else:\n",
    "        return 0\n",
    "\n",
    "def get_ret(r):\n",
    "    if r['Signal'] == 1:\n",
    "        return ((r['Next Day Close'] - r['Next Day Open']) / r['Next Day Open']) * 100\n",
    "    else:\n",
    "        return 0\n",
    "\n",
    "tf2 = tf2.assign(Signal = tf2.apply(get_signal, axis=1))\n",
    "tf2 = tf2.assign(PnL = tf2.apply(get_ret, axis=1))\n",
    "\n",
    "display(tf2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "c03190c6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4.253730773925781"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Test if successfully predicted the next day's price\n",
    "(tf2[tf2['Signal']==1]['Next Day Close'] - tf2[tf2['Signal']==1]['Next Day Open']).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "5849a015",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "30.024864196777344"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluate the period we tested in isolation to benchmark how many point the intraday strategy would generate\n",
    "(sp['Close'].iloc[-1000:] - sp['Open'].iloc[-1000:]).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "010c846f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trades:  1000 \n",
      "Wins:  547 \n",
      "Losses:  447 \n",
      "Breakeven:  6 \n",
      "Win/Loss Ratio:  1.224 \n",
      "Mean Win:  0.457 \n",
      "Mean Loss:  -0.513 \n",
      "Mean:  0.02 \n",
      "Std Dev:  0.656 \n",
      "Max Win:  2.756 \n",
      "Max Loss:  -4.175 \n",
      "Sharpe Ratio:  0.4956\n"
     ]
    }
   ],
   "source": [
    "# Get stats for intraday return\n",
    "get_stats((sp['Close'].iloc[-1000:] - sp['Open'].iloc[-1000:]) / sp['Open'].iloc[-1000:] * 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "2ce92596",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trades:  1000 \n",
      "Wins:  210 \n",
      "Losses:  184 \n",
      "Breakeven:  606 \n",
      "Win/Loss Ratio:  1.141 \n",
      "Mean Win:  0.474 \n",
      "Mean Loss:  -0.519 \n",
      "Mean:  0.004 \n",
      "Std Dev:  0.412 \n",
      "Max Win:  2.756 \n",
      "Max Loss:  -2.124 \n",
      "Sharpe Ratio:  0.1593\n"
     ]
    }
   ],
   "source": [
    "# Get stats for model to compare\n",
    "get_stats(tf2['PnL'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e84448f",
   "metadata": {},
   "source": [
    "## Modeling with Dynamic Time Warping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "3aaeb4aa",
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.spatial.distance import euclidean\n",
    "from fastdtw import fastdtw"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "257ad355",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create function that will determine distance between 2 series\n",
    "def dtw_dist(x,y):\n",
    "    distance, path = fastdtw(x, y, dist=euclidean)\n",
    "    return distance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "7737eb77",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split 16 years of time series data into distinct five day periods\n",
    "tseries = []\n",
    "tlen = 5\n",
    "for i in range(tlen, len(sp), tlen):\n",
    "    pctc = sp['Close'].iloc[i-tlen:i].pct_change()[1:].values * 100\n",
    "    res = sp['Close'].iloc[i-tlen: i+1].pct_change()[-1] * 100\n",
    "    tseries.append((pctc, res))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "f3662f06",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([-6.96897812e-04, -2.45521978e+00,  3.23491509e-01, -1.80177925e+00]),\n",
       " 1.3689266269875988)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tseries[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "8de59792",
   "metadata": {},
   "outputs": [],
   "source": [
    "dist_pairs = []\n",
    "for i in range(len(tseries)):\n",
    "    for j in range(len(tseries)):\n",
    "        dist = dtw_dist(tseries[i][0], tseries[j][0])\n",
    "        dist_pairs.append((i, j, dist, tseries[i][1], tseries[j][1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "551bb4fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert to dataframe\n",
    "dist_frame = pd.DataFrame(dist_pairs, columns=['A', 'B', 'Dist', 'A Ret', 'B Ret'])\n",
    "sf = dist_frame[dist_frame['Dist']>0].sort_values(['A', 'B']).reset_index(drop=1)\n",
    "sfe = sf[sf['A']<sf['B']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "4dedb213",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Limit trades to where the distance is less than 1 and the first series has a positive return\n",
    "winf = sfe[(sfe['Dist']<=1) & (sfe['A Ret']>0)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "ca406dff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fb0604221f0>]"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot one of top outcomes\n",
    "plt.plot(np.arange(4), tseries[1][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "2b4ddd65",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fb0602eb520>]"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot a second output\n",
    "plt.plot(np.arange(4), tseries[67][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "2231860f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2696      None\n",
       "2747      None\n",
       "2851      None\n",
       "4581      None\n",
       "4727      None\n",
       "          ... \n",
       "149655    None\n",
       "149667    None\n",
       "150473    None\n",
       "154571    None\n",
       "154573    None\n",
       "Length: 839, dtype: object"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create function to evaluate potential trades\n",
    "excluded = {}\n",
    "return_list = []\n",
    "def get_returns(r):\n",
    "    if excluded.get(r['A']) is None:\n",
    "        return_list.append(r['B Ret'])\n",
    "        if r['B Ret'] < 0:\n",
    "            excluded.update({r['A']:1})\n",
    "\n",
    "winf.apply(get_returns, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "d08d3cec",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trades:  287 \n",
      "Wins:  199 \n",
      "Losses:  88 \n",
      "Breakeven:  0 \n",
      "Win/Loss Ratio:  2.261 \n",
      "Mean Win:  0.478 \n",
      "Mean Loss:  -0.593 \n",
      "Mean:  0.15 \n",
      "Std Dev:  0.667 \n",
      "Max Win:  2.125 \n",
      "Max Loss:  -2.206 \n",
      "Sharpe Ratio:  3.5627\n"
     ]
    }
   ],
   "source": [
    "get_stats(pd.Series(return_list))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6c26d73b",
   "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": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}