preprocessing.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "initial_id",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2023-11-14T11:50:15.911790057Z",
     "start_time": "2023-11-14T11:50:15.259137015Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "                                  m_f    A_f     B_f  spd_f  idchg_f\n2023-06-01 00:00:00.025  27201.172622  3.770  10.000    0.1    879.0\n2023-06-01 00:00:00.234  27201.169711  4.195   9.655    0.1   1828.0\n2023-06-01 00:00:00.348  27201.171554  4.169  10.487    0.1   1360.0\n2023-06-01 00:00:00.400  27201.141571  4.069   2.895    0.1   1384.0\n2023-06-01 00:00:00.417  27201.130881  6.717   3.001    0.1     91.0",
      "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>m_f</th>\n      <th>A_f</th>\n      <th>B_f</th>\n      <th>spd_f</th>\n      <th>idchg_f</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>2023-06-01 00:00:00.025</th>\n      <td>27201.172622</td>\n      <td>3.770</td>\n      <td>10.000</td>\n      <td>0.1</td>\n      <td>879.0</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:00.234</th>\n      <td>27201.169711</td>\n      <td>4.195</td>\n      <td>9.655</td>\n      <td>0.1</td>\n      <td>1828.0</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:00.348</th>\n      <td>27201.171554</td>\n      <td>4.169</td>\n      <td>10.487</td>\n      <td>0.1</td>\n      <td>1360.0</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:00.400</th>\n      <td>27201.141571</td>\n      <td>4.069</td>\n      <td>2.895</td>\n      <td>0.1</td>\n      <td>1384.0</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:00.417</th>\n      <td>27201.130881</td>\n      <td>6.717</td>\n      <td>3.001</td>\n      <td>0.1</td>\n      <td>91.0</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import math\n",
    "\n",
    "# Function to add 1 microsecond to duplicates\n",
    "def resolve_duplicates(index):\n",
    "    seen = set()\n",
    "    new_index = []\n",
    "    for dt in index:\n",
    "        while dt in seen:\n",
    "            dt += pd.Timedelta(microseconds=1)\n",
    "        seen.add(dt)\n",
    "        new_index.append(dt)\n",
    "    return new_index\n",
    "\n",
    "def read_spot(path, start, end, res_dup=True):\n",
    "    df = pd.read_csv(path, skiprows=start, nrows=end-start)\n",
    "    df[\"time\"] = pd.to_datetime(df[\"time\"])\n",
    "    df = df.set_index(\"time\")\n",
    "    if res_dup:\n",
    "        df.index = resolve_duplicates(df.index)\n",
    "    return df\n",
    "\n",
    "def read_future(path, start, end, res_dup=True):\n",
    "    df = pd.read_csv(path, skiprows=start, nrows=end-start)\n",
    "    df[\"event_time\"] = pd.to_datetime(df[\"event_time\"], unit=\"ms\")\n",
    "    df = df.set_index(\"event_time\")\n",
    "    df = df.sort_index()\n",
    "    df = df.sort_values(by=[\"update_id\"])\n",
    "    if res_dup:\n",
    "        df.index = resolve_duplicates(df.index)\n",
    "    return df\n",
    "\n",
    "def generate_spot_features(df, features):\n",
    "    df[\"mid_price\"] = (df[\"b\"] * df[\"A\"] + df[\"a\"] * df[\"B\"]) / (df[\"B\"] + df[\"A\"])\n",
    "    for ft in features:\n",
    "        if ft == \"spread\":\n",
    "            df[\"spread\"] = (df[\"a\"] - df[\"b\"]).round(5)\n",
    "        elif ft == \"id_change\":\n",
    "            df[\"id_change\"] = df[\"u\"] - df[\"u\"].shift(1)\n",
    "        elif ft == \"log_return\":\n",
    "            df[\"log_return\"] = df[\"mid_price\"].apply(lambda x: math.log(x)) - df[\"mid_price\"].shift(1).apply(lambda x: math.log(x))\n",
    "            df[\"log_return\"] = df[\"log_return\"].round(4).apply(lambda x: 0.0 if x == -0.0 else x)\n",
    "        elif ft == \"B_change\":\n",
    "            df[\"B_change\"] = df[\"B\"].pct_change()\n",
    "        elif ft == \"A_change\":\n",
    "            df[\"A_change\"] = df[\"A\"].pct_change()\n",
    "        elif ft == \"mid_price\":\n",
    "            continue\n",
    "        else:\n",
    "            raise ValueError(\"Invalid feature name \", ft)\n",
    "        \n",
    "def generate_future_features(df, features):\n",
    "    df[\"mid_price\"] = (df[\"best_bid_price\"] * df[\"best_ask_qty\"] + df[\"best_ask_price\"] * df[\"best_bid_qty\"]) / (df[\"best_bid_qty\"] + df[\"best_ask_qty\"])\n",
    "    for ft in features:\n",
    "        if ft == \"spread\":\n",
    "            df[\"spread\"] = (df[\"best_ask_price\"] - df[\"best_bid_price\"]).round(5)\n",
    "        elif ft == \"id_change\":\n",
    "            df[\"id_change\"] = df[\"update_id\"] - df[\"update_id\"].shift(1)\n",
    "        elif ft == \"log_return\":\n",
    "            df[\"log_return\"] = df[\"mid_price\"].apply(lambda x: math.log(x)) - df[\"mid_price\"].shift(1).apply(lambda x: math.log(x))\n",
    "            df[\"log_return\"] = df[\"log_return\"].round(4).apply(lambda x: 0.0 if x == -0.0 else x)\n",
    "        elif ft == \"B_change\":\n",
    "            df[\"B_change\"] = df[\"best_bid_qty\"].pct_change()\n",
    "        elif ft == \"A_change\":\n",
    "            df[\"A_change\"] = df[\"best_ask_qty\"].pct_change()\n",
    "        elif ft == \"mid_price\":\n",
    "            continue\n",
    "        else:\n",
    "            raise ValueError(\"Invalid feature name: \", ft)\n",
    "        \n",
    "def filter_rows(df, qty_chg):\n",
    "    return df[(abs(df[\"B_change\"]) > qty_chg) | (abs(df[\"A_change\"]) > qty_chg) | (df[\"log_return\"] != 0.0)]\n",
    "\n",
    "def standardize_spot_format(df):\n",
    "    output = df.dropna()\n",
    "    output = output[[\"mid_price\", \"A\", \"B\", \"spread\", \"id_change\", \"log_return\"]]\n",
    "    output.columns = [\"m_s\", \"A_s\", \"B_s\", \"spd_s\", \"idchg_s\", \"logret_s\"]\n",
    "    return output\n",
    "\n",
    "def standardize_futures_format(df):\n",
    "    output = df.dropna()\n",
    "    output = output[[\"mid_price\", \"best_ask_qty\", \"best_bid_qty\", \"spread\", \"id_change\"]]\n",
    "    output.columns = [\"m_f\", \"A_f\", \"B_f\", \"spd_f\", \"idchg_f\"]\n",
    "    return output\n",
    "\n",
    "spot = read_spot(\"dataset/BTCUSDT-bookTicker-2023-06-s\", 0, 50000)\n",
    "generate_spot_features(spot, [\"mid_price\", \"spread\", \"id_change\", \"B_change\", \"A_change\", \"log_return\"])\n",
    "spot = filter_rows(spot, 0.05)\n",
    "spot = standardize_spot_format(spot)\n",
    "\n",
    "futures = read_future(\"dataset/BTCUSDT-bookTicker-2023-06.csv\", 0, 50000)\n",
    "generate_future_features(futures, [\"mid_price\", \"spread\", \"id_change\", \"B_change\", \"A_change\", \"log_return\"])\n",
    "futures = filter_rows(futures, 0.05)\n",
    "futures = standardize_futures_format(futures)\n",
    "\n",
    "spot.head()\n",
    "futures.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "outputs": [
    {
     "data": {
      "text/plain": "0.0037732135834112057"
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import math\n",
    "math.log(27112.90) - math.log(27010.79)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-11-23T11:40:14.545692314Z",
     "start_time": "2023-11-23T11:40:14.536504067Z"
    }
   },
   "id": "4e458a87b2a30641"
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "outputs": [
    {
     "data": {
      "text/plain": "0.003773213583411391"
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "math.log(27112.90 / 27010.79)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-11-23T11:41:37.023718723Z",
     "start_time": "2023-11-23T11:41:36.978986522Z"
    }
   },
   "id": "9f67c68fd7538325"
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "outputs": [
    {
     "data": {
      "text/plain": "0.0037803411155319998"
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(27112.90 - 27010.79) / 27010.79"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-11-23T11:40:22.587752781Z",
     "start_time": "2023-11-23T11:40:22.543884066Z"
    }
   },
   "id": "c0df1ee5ac9f4f13"
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "outputs": [
    {
     "data": {
      "text/plain": "                                     m_s      A_s      B_s  spd_s  idchg_s  \\\n2023-06-01 00:00:00.025000  27210.358381  1.66972  8.64481   0.01      0.0   \n2023-06-01 00:00:00.234000  27210.358381  1.66972  8.64481   0.01      0.0   \n2023-06-01 00:00:00.348000  27210.358381  1.66972  8.64481   0.01      0.0   \n2023-06-01 00:00:00.368963  27210.358811  1.16608  8.63971   0.01     23.0   \n2023-06-01 00:00:00.400000  27210.358811  1.16608  8.63971   0.01      0.0   \n...                                  ...      ...      ...    ...      ...   \n2023-06-01 00:00:06.315702  27211.183390  0.00743  0.00381   0.01      1.0   \n2023-06-01 00:00:06.315788  27211.189816  0.00743  0.39638   0.01      4.0   \n2023-06-01 00:00:06.316952  27211.189853  0.00743  0.49678   0.01      1.0   \n2023-06-01 00:00:06.320000  27211.189853  0.00743  0.49678   0.01      0.0   \n2023-06-01 00:00:06.326190  27211.189879  0.00743  0.60601   0.01      5.0   \n\n                            logret_s           m_f     A_f     B_f  spd_f  \\\n2023-06-01 00:00:00.025000       0.0  27201.172622   3.770  10.000    0.1   \n2023-06-01 00:00:00.234000       0.0  27201.169711   4.195   9.655    0.1   \n2023-06-01 00:00:00.348000       0.0  27201.171554   4.169  10.487    0.1   \n2023-06-01 00:00:00.368963       0.0  27201.171554   4.169  10.487    0.1   \n2023-06-01 00:00:00.400000       0.0  27201.141571   4.069   2.895    0.1   \n...                              ...           ...     ...     ...    ...   \n2023-06-01 00:00:06.315702       0.0  27200.306268  16.748   1.120    0.1   \n2023-06-01 00:00:06.315788       0.0  27200.306268  16.748   1.120    0.1   \n2023-06-01 00:00:06.316952       0.0  27200.306268  16.748   1.120    0.1   \n2023-06-01 00:00:06.320000       0.0  27200.305965  16.743   1.062    0.1   \n2023-06-01 00:00:06.326190       0.0  27200.305965  16.743   1.062    0.1   \n\n                            idchg_f        efp  \n2023-06-01 00:00:00.025000    879.0  -9.185760  \n2023-06-01 00:00:00.234000   1828.0  -9.188670  \n2023-06-01 00:00:00.348000   1360.0  -9.186827  \n2023-06-01 00:00:00.368963   1360.0  -9.187257  \n2023-06-01 00:00:00.400000   1384.0  -9.217240  \n...                             ...        ...  \n2023-06-01 00:00:06.315702     92.0 -10.877121  \n2023-06-01 00:00:06.315788     92.0 -10.883548  \n2023-06-01 00:00:06.316952     92.0 -10.883584  \n2023-06-01 00:00:06.320000    103.0 -10.883888  \n2023-06-01 00:00:06.326190    103.0 -10.883914  \n\n[100 rows x 12 columns]",
      "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>m_s</th>\n      <th>A_s</th>\n      <th>B_s</th>\n      <th>spd_s</th>\n      <th>idchg_s</th>\n      <th>logret_s</th>\n      <th>m_f</th>\n      <th>A_f</th>\n      <th>B_f</th>\n      <th>spd_f</th>\n      <th>idchg_f</th>\n      <th>efp</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>2023-06-01 00:00:00.025000</th>\n      <td>27210.358381</td>\n      <td>1.66972</td>\n      <td>8.64481</td>\n      <td>0.01</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>27201.172622</td>\n      <td>3.770</td>\n      <td>10.000</td>\n      <td>0.1</td>\n      <td>879.0</td>\n      <td>-9.185760</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:00.234000</th>\n      <td>27210.358381</td>\n      <td>1.66972</td>\n      <td>8.64481</td>\n      <td>0.01</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>27201.169711</td>\n      <td>4.195</td>\n      <td>9.655</td>\n      <td>0.1</td>\n      <td>1828.0</td>\n      <td>-9.188670</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:00.348000</th>\n      <td>27210.358381</td>\n      <td>1.66972</td>\n      <td>8.64481</td>\n      <td>0.01</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>27201.171554</td>\n      <td>4.169</td>\n      <td>10.487</td>\n      <td>0.1</td>\n      <td>1360.0</td>\n      <td>-9.186827</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:00.368963</th>\n      <td>27210.358811</td>\n      <td>1.16608</td>\n      <td>8.63971</td>\n      <td>0.01</td>\n      <td>23.0</td>\n      <td>0.0</td>\n      <td>27201.171554</td>\n      <td>4.169</td>\n      <td>10.487</td>\n      <td>0.1</td>\n      <td>1360.0</td>\n      <td>-9.187257</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:00.400000</th>\n      <td>27210.358811</td>\n      <td>1.16608</td>\n      <td>8.63971</td>\n      <td>0.01</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>27201.141571</td>\n      <td>4.069</td>\n      <td>2.895</td>\n      <td>0.1</td>\n      <td>1384.0</td>\n      <td>-9.217240</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    </tr>\n    <tr>\n      <th>2023-06-01 00:00:06.315702</th>\n      <td>27211.183390</td>\n      <td>0.00743</td>\n      <td>0.00381</td>\n      <td>0.01</td>\n      <td>1.0</td>\n      <td>0.0</td>\n      <td>27200.306268</td>\n      <td>16.748</td>\n      <td>1.120</td>\n      <td>0.1</td>\n      <td>92.0</td>\n      <td>-10.877121</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:06.315788</th>\n      <td>27211.189816</td>\n      <td>0.00743</td>\n      <td>0.39638</td>\n      <td>0.01</td>\n      <td>4.0</td>\n      <td>0.0</td>\n      <td>27200.306268</td>\n      <td>16.748</td>\n      <td>1.120</td>\n      <td>0.1</td>\n      <td>92.0</td>\n      <td>-10.883548</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:06.316952</th>\n      <td>27211.189853</td>\n      <td>0.00743</td>\n      <td>0.49678</td>\n      <td>0.01</td>\n      <td>1.0</td>\n      <td>0.0</td>\n      <td>27200.306268</td>\n      <td>16.748</td>\n      <td>1.120</td>\n      <td>0.1</td>\n      <td>92.0</td>\n      <td>-10.883584</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:06.320000</th>\n      <td>27211.189853</td>\n      <td>0.00743</td>\n      <td>0.49678</td>\n      <td>0.01</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>27200.305965</td>\n      <td>16.743</td>\n      <td>1.062</td>\n      <td>0.1</td>\n      <td>103.0</td>\n      <td>-10.883888</td>\n    </tr>\n    <tr>\n      <th>2023-06-01 00:00:06.326190</th>\n      <td>27211.189879</td>\n      <td>0.00743</td>\n      <td>0.60601</td>\n      <td>0.01</td>\n      <td>5.0</td>\n      <td>0.0</td>\n      <td>27200.305965</td>\n      <td>16.743</td>\n      <td>1.062</td>\n      <td>0.1</td>\n      <td>103.0</td>\n      <td>-10.883914</td>\n    </tr>\n  </tbody>\n</table>\n<p>100 rows × 12 columns</p>\n</div>"
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combined_df = pd.concat([spot, futures], axis=1)\n",
    "\n",
    "# set logret_s to 0 if nan\n",
    "combined_df[\"logret_s\"] = combined_df[\"logret_s\"].fillna(0)\n",
    "combined_df[\"idchg_s\"] = combined_df[\"idchg_s\"].fillna(0)\n",
    "\n",
    "# fill nan with values from previous row\n",
    "combined_df = combined_df.fillna(method=\"ffill\")\n",
    "combined_df[\"efp\"] = combined_df[\"m_f\"] - combined_df[\"m_s\"]\n",
    "combined_df.dropna(inplace=True)\n",
    "combined_df.head(100)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-11-14T11:58:34.084968144Z",
     "start_time": "2023-11-14T11:58:34.064554119Z"
    }
   },
   "id": "4c24d007740d4235"
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "outputs": [],
   "source": [
    "training_df = combined_df[[\"A_s\", \"B_s\", \"spd_s\", \"idchg_s\", \"A_f\", \"B_f\", \"spd_f\", \"idchg_f\", \"efp\", \"logret_s\"]]"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-11-14T12:03:02.651743403Z",
     "start_time": "2023-11-14T12:03:02.650055241Z"
    }
   },
   "id": "85d6ea1ba4e8a58c"
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "outputs": [
    {
     "data": {
      "text/plain": "array([[<AxesSubplot:title={'center':'A_s'}>,\n        <AxesSubplot:title={'center':'B_s'}>,\n        <AxesSubplot:title={'center':'spd_s'}>],\n       [<AxesSubplot:title={'center':'idchg_s'}>,\n        <AxesSubplot:title={'center':'A_f'}>,\n        <AxesSubplot:title={'center':'B_f'}>],\n       [<AxesSubplot:title={'center':'spd_f'}>,\n        <AxesSubplot:title={'center':'idchg_f'}>,\n        <AxesSubplot:title={'center':'efp'}>],\n       [<AxesSubplot:title={'center':'logret_s'}>, <AxesSubplot:>,\n        <AxesSubplot:>]], dtype=object)"
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": "<Figure size 1440x1080 with 12 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the distribution of all columns in subplots\n",
    "training_df.hist(bins=10, figsize=(20, 15))"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-11-14T12:03:06.365247575Z",
     "start_time": "2023-11-14T12:03:05.167847391Z"
    }
   },
   "id": "57dd197bce679ac0"
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "outputs": [],
   "source": [
    "# save the dataframe to a csv file\n",
    "training_df.to_csv(\"dataset/ticker/BTCUSDT-bookTicker-2023-06.csv\", index=False)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-11-06T15:38:57.195156706Z",
     "start_time": "2023-11-06T15:38:57.056303081Z"
    }
   },
   "id": "850ff5258c21220"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}