stock_predictorRF.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Yahoo finance API for daily stock and index prices\n",
    "import yfinance as yf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                  Open         High          Low        Close  \\\n",
      "Date                                                                            \n",
      "1927-12-30 00:00:00-05:00    17.660000    17.660000    17.660000    17.660000   \n",
      "1928-01-03 00:00:00-05:00    17.760000    17.760000    17.760000    17.760000   \n",
      "1928-01-04 00:00:00-05:00    17.719999    17.719999    17.719999    17.719999   \n",
      "1928-01-05 00:00:00-05:00    17.549999    17.549999    17.549999    17.549999   \n",
      "1928-01-06 00:00:00-05:00    17.660000    17.660000    17.660000    17.660000   \n",
      "...                                ...          ...          ...          ...   \n",
      "2024-02-12 00:00:00-05:00  5026.830078  5048.390137  5016.830078  5021.839844   \n",
      "2024-02-13 00:00:00-05:00  4967.939941  4971.299805  4920.310059  4953.169922   \n",
      "2024-02-14 00:00:00-05:00  4976.439941  5002.520020  4956.450195  5000.620117   \n",
      "2024-02-15 00:00:00-05:00  5003.140137  5032.720215  4999.439941  5029.729980   \n",
      "2024-02-16 00:00:00-05:00  5031.129883  5038.700195  4999.520020  5005.569824   \n",
      "\n",
      "                               Volume  Dividends  Stock Splits  \n",
      "Date                                                            \n",
      "1927-12-30 00:00:00-05:00           0        0.0           0.0  \n",
      "1928-01-03 00:00:00-05:00           0        0.0           0.0  \n",
      "1928-01-04 00:00:00-05:00           0        0.0           0.0  \n",
      "1928-01-05 00:00:00-05:00           0        0.0           0.0  \n",
      "1928-01-06 00:00:00-05:00           0        0.0           0.0  \n",
      "...                               ...        ...           ...  \n",
      "2024-02-12 00:00:00-05:00  3805740000        0.0           0.0  \n",
      "2024-02-13 00:00:00-05:00  4302190000        0.0           0.0  \n",
      "2024-02-14 00:00:00-05:00  3845600000        0.0           0.0  \n",
      "2024-02-15 00:00:00-05:00  4137970000        0.0           0.0  \n",
      "2024-02-16 00:00:00-05:00  3833270000        0.0           0.0  \n",
      "\n",
      "[24148 rows x 7 columns]\n"
     ]
    }
   ],
   "source": [
    "# Enable ticker class to get information about particular stock\n",
    "sp500 = yf.Ticker('^GSPC')\n",
    "# Get historical market data for sp500\n",
    "sp500 = sp500.history(period='max')\n",
    "print(sp500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: title={'center': 'S&P 500'}, xlabel='Date'>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the close prices of sp500, using date/time index as x-axis\n",
    "sp500.plot.line(y='Close', use_index=True, title='S&P 500')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Data cleaning - remove columns that are not needed; dividends and stock splits are more useful for individual stocks\n",
    "del sp500['Dividends']\n",
    "del sp500['Stock Splits']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                  Open         High          Low        Close  \\\n",
      "Date                                                                            \n",
      "1927-12-30 00:00:00-05:00    17.660000    17.660000    17.660000    17.660000   \n",
      "1928-01-03 00:00:00-05:00    17.760000    17.760000    17.760000    17.760000   \n",
      "1928-01-04 00:00:00-05:00    17.719999    17.719999    17.719999    17.719999   \n",
      "1928-01-05 00:00:00-05:00    17.549999    17.549999    17.549999    17.549999   \n",
      "1928-01-06 00:00:00-05:00    17.660000    17.660000    17.660000    17.660000   \n",
      "...                                ...          ...          ...          ...   \n",
      "2024-02-12 00:00:00-05:00  5026.830078  5048.390137  5016.830078  5021.839844   \n",
      "2024-02-13 00:00:00-05:00  4967.939941  4971.299805  4920.310059  4953.169922   \n",
      "2024-02-14 00:00:00-05:00  4976.439941  5002.520020  4956.450195  5000.620117   \n",
      "2024-02-15 00:00:00-05:00  5003.140137  5032.720215  4999.439941  5029.729980   \n",
      "2024-02-16 00:00:00-05:00  5031.129883  5038.700195  4999.520020  5005.569824   \n",
      "\n",
      "                               Volume  Dividends  Stock Splits     Tomorrow  \n",
      "Date                                                                         \n",
      "1927-12-30 00:00:00-05:00           0        0.0           0.0    17.760000  \n",
      "1928-01-03 00:00:00-05:00           0        0.0           0.0    17.719999  \n",
      "1928-01-04 00:00:00-05:00           0        0.0           0.0    17.549999  \n",
      "1928-01-05 00:00:00-05:00           0        0.0           0.0    17.660000  \n",
      "1928-01-06 00:00:00-05:00           0        0.0           0.0    17.500000  \n",
      "...                               ...        ...           ...          ...  \n",
      "2024-02-12 00:00:00-05:00  3805740000        0.0           0.0  4953.169922  \n",
      "2024-02-13 00:00:00-05:00  4302190000        0.0           0.0  5000.620117  \n",
      "2024-02-14 00:00:00-05:00  3845600000        0.0           0.0  5029.729980  \n",
      "2024-02-15 00:00:00-05:00  4137970000        0.0           0.0  5005.569824  \n",
      "2024-02-16 00:00:00-05:00  3833270000        0.0           0.0          NaN  \n",
      "\n",
      "[24148 rows x 8 columns]\n"
     ]
    }
   ],
   "source": [
    "# Target of ML model is to predict if the stock price will go up or down tomorrow; we care more about the directionality of stock prices than the actual price\n",
    "# Set new table column \"Tomorrow\" to be the \"Close\" price of the next day - this sets the target for the ML model\n",
    "sp500[\"Tomorrow\"] = sp500[\"Close\"].shift(-1)\n",
    "print(sp500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                  Open         High          Low        Close  \\\n",
      "Date                                                                            \n",
      "1927-12-30 00:00:00-05:00    17.660000    17.660000    17.660000    17.660000   \n",
      "1928-01-03 00:00:00-05:00    17.760000    17.760000    17.760000    17.760000   \n",
      "1928-01-04 00:00:00-05:00    17.719999    17.719999    17.719999    17.719999   \n",
      "1928-01-05 00:00:00-05:00    17.549999    17.549999    17.549999    17.549999   \n",
      "1928-01-06 00:00:00-05:00    17.660000    17.660000    17.660000    17.660000   \n",
      "...                                ...          ...          ...          ...   \n",
      "2024-02-12 00:00:00-05:00  5026.830078  5048.390137  5016.830078  5021.839844   \n",
      "2024-02-13 00:00:00-05:00  4967.939941  4971.299805  4920.310059  4953.169922   \n",
      "2024-02-14 00:00:00-05:00  4976.439941  5002.520020  4956.450195  5000.620117   \n",
      "2024-02-15 00:00:00-05:00  5003.140137  5032.720215  4999.439941  5029.729980   \n",
      "2024-02-16 00:00:00-05:00  5031.129883  5038.700195  4999.520020  5005.569824   \n",
      "\n",
      "                               Volume  Dividends  Stock Splits     Tomorrow  \\\n",
      "Date                                                                          \n",
      "1927-12-30 00:00:00-05:00           0        0.0           0.0    17.760000   \n",
      "1928-01-03 00:00:00-05:00           0        0.0           0.0    17.719999   \n",
      "1928-01-04 00:00:00-05:00           0        0.0           0.0    17.549999   \n",
      "1928-01-05 00:00:00-05:00           0        0.0           0.0    17.660000   \n",
      "1928-01-06 00:00:00-05:00           0        0.0           0.0    17.500000   \n",
      "...                               ...        ...           ...          ...   \n",
      "2024-02-12 00:00:00-05:00  3805740000        0.0           0.0  4953.169922   \n",
      "2024-02-13 00:00:00-05:00  4302190000        0.0           0.0  5000.620117   \n",
      "2024-02-14 00:00:00-05:00  3845600000        0.0           0.0  5029.729980   \n",
      "2024-02-15 00:00:00-05:00  4137970000        0.0           0.0  5005.569824   \n",
      "2024-02-16 00:00:00-05:00  3833270000        0.0           0.0          NaN   \n",
      "\n",
      "                           Target  \n",
      "Date                               \n",
      "1927-12-30 00:00:00-05:00       1  \n",
      "1928-01-03 00:00:00-05:00       0  \n",
      "1928-01-04 00:00:00-05:00       0  \n",
      "1928-01-05 00:00:00-05:00       1  \n",
      "1928-01-06 00:00:00-05:00       0  \n",
      "...                           ...  \n",
      "2024-02-12 00:00:00-05:00       0  \n",
      "2024-02-13 00:00:00-05:00       1  \n",
      "2024-02-14 00:00:00-05:00       1  \n",
      "2024-02-15 00:00:00-05:00       0  \n",
      "2024-02-16 00:00:00-05:00       0  \n",
      "\n",
      "[24148 rows x 9 columns]\n"
     ]
    }
   ],
   "source": [
    "# Set new table column \"Target\" to be 1 if \"Tomorrow\" > \"Close\", else 0. So, 1 means the stock price went up, 0 means the stock price went down\n",
    "sp500[\"Target\"] = (sp500[\"Tomorrow\"] > sp500[\"Close\"]).astype(int)\n",
    "print(sp500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                  Open         High          Low        Close  \\\n",
      "Date                                                                            \n",
      "1990-01-02 00:00:00-05:00   353.399994   359.690002   351.980011   359.690002   \n",
      "1990-01-03 00:00:00-05:00   359.690002   360.589996   357.890015   358.760010   \n",
      "1990-01-04 00:00:00-05:00   358.760010   358.760010   352.890015   355.670013   \n",
      "1990-01-05 00:00:00-05:00   355.670013   355.670013   351.350006   352.200012   \n",
      "1990-01-08 00:00:00-05:00   352.200012   354.239990   350.540009   353.790009   \n",
      "...                                ...          ...          ...          ...   \n",
      "2024-02-12 00:00:00-05:00  5026.830078  5048.390137  5016.830078  5021.839844   \n",
      "2024-02-13 00:00:00-05:00  4967.939941  4971.299805  4920.310059  4953.169922   \n",
      "2024-02-14 00:00:00-05:00  4976.439941  5002.520020  4956.450195  5000.620117   \n",
      "2024-02-15 00:00:00-05:00  5003.140137  5032.720215  4999.439941  5029.729980   \n",
      "2024-02-16 00:00:00-05:00  5031.129883  5038.700195  4999.520020  5005.569824   \n",
      "\n",
      "                               Volume  Dividends  Stock Splits     Tomorrow  \\\n",
      "Date                                                                          \n",
      "1990-01-02 00:00:00-05:00   162070000        0.0           0.0   358.760010   \n",
      "1990-01-03 00:00:00-05:00   192330000        0.0           0.0   355.670013   \n",
      "1990-01-04 00:00:00-05:00   177000000        0.0           0.0   352.200012   \n",
      "1990-01-05 00:00:00-05:00   158530000        0.0           0.0   353.790009   \n",
      "1990-01-08 00:00:00-05:00   140110000        0.0           0.0   349.619995   \n",
      "...                               ...        ...           ...          ...   \n",
      "2024-02-12 00:00:00-05:00  3805740000        0.0           0.0  4953.169922   \n",
      "2024-02-13 00:00:00-05:00  4302190000        0.0           0.0  5000.620117   \n",
      "2024-02-14 00:00:00-05:00  3845600000        0.0           0.0  5029.729980   \n",
      "2024-02-15 00:00:00-05:00  4137970000        0.0           0.0  5005.569824   \n",
      "2024-02-16 00:00:00-05:00  3833270000        0.0           0.0          NaN   \n",
      "\n",
      "                           Target  \n",
      "Date                               \n",
      "1990-01-02 00:00:00-05:00       0  \n",
      "1990-01-03 00:00:00-05:00       0  \n",
      "1990-01-04 00:00:00-05:00       0  \n",
      "1990-01-05 00:00:00-05:00       1  \n",
      "1990-01-08 00:00:00-05:00       0  \n",
      "...                           ...  \n",
      "2024-02-12 00:00:00-05:00       0  \n",
      "2024-02-13 00:00:00-05:00       1  \n",
      "2024-02-14 00:00:00-05:00       1  \n",
      "2024-02-15 00:00:00-05:00       0  \n",
      "2024-02-16 00:00:00-05:00       0  \n",
      "\n",
      "[8598 rows x 9 columns]\n"
     ]
    }
   ],
   "source": [
    "# Since we're dealing with stock market data, the market could have changed significantly in the past 20 years. We'll only use data from 1990 onwards.\n",
    "sp500 = sp500.loc['1990-01-01':].copy()\n",
    "print(sp500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(min_samples_split=100, random_state=1)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Begin training the model. We'll use a random forest classifier since it usually avoids overfitting (as long as not too deep) and is robust to outliers (non-linear relationships, \n",
    "# which are common in stock market data)\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "# N_estimators is the number of trees in the forest; min_samples_split is the minimum number of samples required to split an internal node (# of samples to consider, protects against overfitting); \n",
    "# random_state is the seed used by the random number generator (to ensure reproducibility of the model's results)\n",
    "model = RandomForestClassifier(n_estimators=100, min_samples_split=100, random_state=1)\n",
    "\n",
    "# Split data into train and test sets. Since this is time series data, cross validation is not good because it shuffles the data, which is not good for time series data \n",
    "# (i.e. using future data that we wouldn't usually have).\n",
    "train = sp500.iloc[:-100]\n",
    "test = sp500.iloc[-100:]\n",
    "\n",
    "predictors = [\"Close\", \"Volume\", \"Open\", \"High\", \"Low\"]\n",
    "model.fit(train[predictors], train[\"Target\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Date\n",
      "2023-09-26 00:00:00-04:00    1\n",
      "2023-09-27 00:00:00-04:00    1\n",
      "2023-09-28 00:00:00-04:00    1\n",
      "2023-09-29 00:00:00-04:00    1\n",
      "2023-10-02 00:00:00-04:00    1\n",
      "                            ..\n",
      "2024-02-12 00:00:00-05:00    0\n",
      "2024-02-13 00:00:00-05:00    0\n",
      "2024-02-14 00:00:00-05:00    0\n",
      "2024-02-15 00:00:00-05:00    0\n",
      "2024-02-16 00:00:00-05:00    0\n",
      "Length: 100, dtype: int32\n",
      "Accuracy:  0.44\n"
     ]
    }
   ],
   "source": [
    "# Get metrics on resulting model\n",
    "from sklearn.metrics import accuracy_score\n",
    "import pandas as pd\n",
    "\n",
    "test_predictions = model.predict(test[predictors])\n",
    "test_predictions = pd.Series(test_predictions, index=test.index) # Convert numpy array to pandas series for easier manipulation\n",
    "print(test_predictions)\n",
    "accuracy = accuracy_score(test[\"Target\"], test_predictions)\n",
    "print(\"Accuracy: \", accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: xlabel='Date'>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the predictions\n",
    "combined = pd.concat([test[\"Target\"], test_predictions], axis=1) # axis=1 means concatenate columns\n",
    "combined.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Build a more robust way of testing the model - currently we're only testing on 100 days of data, which is not enough to get a good idea of the model's performance\n",
    "# Summarize what we've done so far into a function; uses training data to train the model, then uses the resulting model to predict on the test data\n",
    "def predict(train, test, predictors, model):\n",
    "    model.fit(train[predictors], train[\"Target\"])\n",
    "    test_predictions = model.predict(test[predictors])\n",
    "    test_predictions = pd.Series(test_predictions, index=test.index, name=\"Predictions\")\n",
    "    combined_res = pd.concat([test[\"Target\"], test_predictions], axis=1)\n",
    "    return combined_res\n",
    "\n",
    "# Stock market years are about 250 trading days long, so we'll use that as the step size for the backtest. We'll start at 2500 days to give the model enough data to train on.\n",
    "# Essentially, we are training on 10 years of data, then testing on the next year, then take the 11 years of data and predict for the next year, and continue until we run out of data.\n",
    "def backtest(data, model, predictors, start=2500, step=250):\n",
    "    all_predictions = []\n",
    "    for i in range(start, data.shape[0], step):\n",
    "        train = data.iloc[0:i].copy()\n",
    "        test = data.iloc[i:(i+step)].copy()\n",
    "        predictions = predict(train, test, predictors, model)\n",
    "        all_predictions.append(predictions)\n",
    "    return pd.concat(all_predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = backtest(sp500, model, predictors)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.48983273204329286\n"
     ]
    }
   ],
   "source": [
    "# Calculate the accuracy of the backtest\n",
    "accuracy = accuracy_score(predictions[\"Target\"], predictions[\"Predictions\"])\n",
    "print(accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                  Open         High          Low        Close  \\\n",
      "Date                                                                            \n",
      "1993-12-14 00:00:00-05:00   465.730011   466.119995   462.459991   463.059998   \n",
      "1993-12-15 00:00:00-05:00   463.059998   463.690002   461.839996   461.839996   \n",
      "1993-12-16 00:00:00-05:00   461.859985   463.980011   461.859985   463.339996   \n",
      "1993-12-17 00:00:00-05:00   463.339996   466.380005   463.339996   466.380005   \n",
      "1993-12-20 00:00:00-05:00   466.380005   466.899994   465.529999   465.850006   \n",
      "...                                ...          ...          ...          ...   \n",
      "2024-02-09 00:00:00-05:00  5004.169922  5030.060059  5000.339844  5026.609863   \n",
      "2024-02-12 00:00:00-05:00  5026.830078  5048.390137  5016.830078  5021.839844   \n",
      "2024-02-13 00:00:00-05:00  4967.939941  4971.299805  4920.310059  4953.169922   \n",
      "2024-02-14 00:00:00-05:00  4976.439941  5002.520020  4956.450195  5000.620117   \n",
      "2024-02-15 00:00:00-05:00  5003.140137  5032.720215  4999.439941  5029.729980   \n",
      "\n",
      "                               Volume  Dividends  Stock Splits     Tomorrow  \\\n",
      "Date                                                                          \n",
      "1993-12-14 00:00:00-05:00   275050000        0.0           0.0   461.839996   \n",
      "1993-12-15 00:00:00-05:00   331770000        0.0           0.0   463.339996   \n",
      "1993-12-16 00:00:00-05:00   284620000        0.0           0.0   466.380005   \n",
      "1993-12-17 00:00:00-05:00   363750000        0.0           0.0   465.850006   \n",
      "1993-12-20 00:00:00-05:00   255900000        0.0           0.0   465.299988   \n",
      "...                               ...        ...           ...          ...   \n",
      "2024-02-09 00:00:00-05:00  3912990000        0.0           0.0  5021.839844   \n",
      "2024-02-12 00:00:00-05:00  3805740000        0.0           0.0  4953.169922   \n",
      "2024-02-13 00:00:00-05:00  4302190000        0.0           0.0  5000.620117   \n",
      "2024-02-14 00:00:00-05:00  3845600000        0.0           0.0  5029.729980   \n",
      "2024-02-15 00:00:00-05:00  4137970000        0.0           0.0  5005.569824   \n",
      "\n",
      "                           Target  Close_Ratio_2  Trend_2  Close_Ratio_5  \\\n",
      "Date                                                                       \n",
      "1993-12-14 00:00:00-05:00       0       0.997157      1.0       0.996617   \n",
      "1993-12-15 00:00:00-05:00       1       0.998681      0.0       0.995899   \n",
      "1993-12-16 00:00:00-05:00       1       1.001621      1.0       0.999495   \n",
      "1993-12-17 00:00:00-05:00       0       1.003270      2.0       1.004991   \n",
      "1993-12-20 00:00:00-05:00       0       0.999431      1.0       1.003784   \n",
      "...                           ...            ...      ...            ...   \n",
      "2024-02-09 00:00:00-05:00       0       1.002863      2.0       1.008686   \n",
      "2024-02-12 00:00:00-05:00       0       0.999525      1.0       1.004543   \n",
      "2024-02-13 00:00:00-05:00       1       0.993116      0.0       0.990848   \n",
      "2024-02-14 00:00:00-05:00       1       1.004767      1.0       1.000118   \n",
      "2024-02-15 00:00:00-05:00       0       1.002902      2.0       1.004661   \n",
      "\n",
      "                           Trend_5  Close_Ratio_60  Trend_60  Close_Ratio_250  \\\n",
      "Date                                                                            \n",
      "1993-12-14 00:00:00-05:00      1.0        1.000283      32.0         1.028047   \n",
      "1993-12-15 00:00:00-05:00      1.0        0.997329      32.0         1.025151   \n",
      "1993-12-16 00:00:00-05:00      2.0        1.000311      32.0         1.028274   \n",
      "1993-12-17 00:00:00-05:00      3.0        1.006561      32.0         1.034781   \n",
      "1993-12-20 00:00:00-05:00      2.0        1.005120      32.0         1.033359   \n",
      "...                            ...             ...       ...              ...   \n",
      "2024-02-09 00:00:00-05:00      4.0        1.062623      40.0         1.148263   \n",
      "2024-02-12 00:00:00-05:00      4.0        1.059650      39.0         1.146247   \n",
      "2024-02-13 00:00:00-05:00      3.0        1.043508      38.0         1.129730   \n",
      "2024-02-14 00:00:00-05:00      3.0        1.051686      38.0         1.139665   \n",
      "2024-02-15 00:00:00-05:00      3.0        1.055899      38.0         1.145319   \n",
      "\n",
      "                           Trend_250  Close_Ratio_1000  Trend_1000  \n",
      "Date                                                                \n",
      "1993-12-14 00:00:00-05:00      127.0          1.176082       512.0  \n",
      "1993-12-15 00:00:00-05:00      126.0          1.172676       512.0  \n",
      "1993-12-16 00:00:00-05:00      127.0          1.176163       513.0  \n",
      "1993-12-17 00:00:00-05:00      128.0          1.183537       514.0  \n",
      "1993-12-20 00:00:00-05:00      128.0          1.181856       513.0  \n",
      "...                              ...               ...         ...  \n",
      "2024-02-09 00:00:00-05:00      139.0          1.252277       529.0  \n",
      "2024-02-12 00:00:00-05:00      138.0          1.250564       529.0  \n",
      "2024-02-13 00:00:00-05:00      138.0          1.232933       529.0  \n",
      "2024-02-14 00:00:00-05:00      138.0          1.244164       530.0  \n",
      "2024-02-15 00:00:00-05:00      139.0          1.250812       531.0  \n",
      "\n",
      "[7597 rows x 19 columns]\n"
     ]
    }
   ],
   "source": [
    "# Adding additional predictors to the model to see if it improves the accuracy\n",
    "# Ratios are more important for this case - we care more about the relative price of the stock than the absolute price; how much the stock price has changed relative to its past\n",
    "\n",
    "horizons = [2,5,60,250,1000] # 2 days, 5 days, 60 days, 250 days, 1000 days for calculating rolling averages of close price\n",
    "new_predictors = []\n",
    "\n",
    "for horizon in horizons:\n",
    "    rolling_averages = sp500.rolling(horizon).mean()\n",
    "    ratio_column = f\"Close_Ratio_{horizon}\"\n",
    "    sp500[ratio_column] = sp500[\"Close\"] / rolling_averages[\"Close\"]\n",
    "    trend_column = f\"Trend_{horizon}\"\n",
    "    sp500[trend_column] = sp500.shift(1).rolling(horizon).sum()[\"Target\"] # Take the sum of the target variable over the past horizon days\n",
    "    new_predictors += [ratio_column, trend_column]\n",
    "\n",
    "# Close price ratio and trend added as new predictors\n",
    "sp500 = sp500.dropna() # Drop rows with NaN values\n",
    "print(sp500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Update model and change hyperparameters\n",
    "model = RandomForestClassifier(n_estimators=200, min_samples_split=25, random_state=1)\n",
    "\n",
    "# Update predict function\n",
    "def predict(train, test, predictors, model):\n",
    "    model.fit(train[predictors], train[\"Target\"])\n",
    "\n",
    "    # Want more control on what is 1 and 0 (use probabilities)\n",
    "    test_predictions = model.predict_proba(test[predictors])[:,1] # Get second column, which is the probability that stock goes up\n",
    "    test_predictions[test_predictions >= 0.6] = 1\n",
    "    test_predictions[test_predictions < 0.6] = 0\n",
    "    \n",
    "    test_predictions = pd.Series(test_predictions, index=test.index, name=\"Predictions\")\n",
    "    combined_res = pd.concat([test[\"Target\"], test_predictions], axis=1)\n",
    "    return combined_res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = backtest(sp500, model, new_predictors)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.48067490680792624\n"
     ]
    }
   ],
   "source": [
    "# Get new accuracy of model\n",
    "accuracy = accuracy_score(predictions[\"Target\"], predictions[\"Predictions\"])\n",
    "print(accuracy)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.9.15"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}