docs/example_usage.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Example Usage"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import logging\n",
    "import os\n",
    "\n",
    "logging.basicConfig(level=os.environ.get(\"LOGLEVEL\", \"INFO\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy import stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import l2_orderbook_tops\n",
    "import l2_orderbook_tops.loader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option('display.max_colwidth', None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>.container { width:100% !important; }</style>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from IPython.core.display import display, HTML\n",
    "display(HTML(\"<style>.container { width:100% !important; }</style>\"))\n",
    "pd.set_option('display.max_columns', None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "INPUT_DIR='/storage/m2/market_data/raw/crypto/binance/BTCUSDT/orderbook'\n",
    "SAMPLE_DATE='2020-08-01'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load input data\n",
    "Input data consists of input timestamp, side (bid/ask), price and qty (order size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>timestamp</th>\n",
       "      <th>side</th>\n",
       "      <th>price</th>\n",
       "      <th>qty</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1596240000009</td>\n",
       "      <td>a</td>\n",
       "      <td>11344.84</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1596240000010</td>\n",
       "      <td>b</td>\n",
       "      <td>11343.67</td>\n",
       "      <td>1.895</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1596240000018</td>\n",
       "      <td>b</td>\n",
       "      <td>11336.60</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1596240000024</td>\n",
       "      <td>a</td>\n",
       "      <td>11348.10</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1596240000025</td>\n",
       "      <td>a</td>\n",
       "      <td>11348.71</td>\n",
       "      <td>0.500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       timestamp side     price    qty\n",
       "0  1596240000009    a  11344.84  0.000\n",
       "1  1596240000010    b  11343.67  1.895\n",
       "2  1596240000018    b  11336.60  0.000\n",
       "3  1596240000024    a  11348.10  0.000\n",
       "4  1596240000025    a  11348.71  0.500"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "snapshot, deltas = l2_orderbook_tops.loader.get_binance_orderbook_snap_delta(INPUT_DIR, SAMPLE_DATE)\n",
    "deltas.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Produce TOPS view\n",
    "\n",
    "Library translates this tall and narrow view into a wide view of the top N best bid & ask prices at each timestamp.  \n",
    "\n",
    "$b_0$ is the best (highest) bid price  \n",
    "$bq_0$ is the order depth corresponding to this price\n",
    "\n",
    "$a_0$ is the best (lowest) ask price  \n",
    "$aq_0$ is the order depth corresponding to this price\n",
    "\n",
    "The spread can be defined as: $a_0 - b_0$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The sum of order depth for both bid and ask sides can be calculated by suming the respective quantities, ie: $$ \\sum \\limits _{i=0} ^{n} aq_{i} $$\n",
    "\n",
    "For a wider view of order depth the optional parameter *watch_dollar_dist_depth* can be used to sum order sizes within a static dollar amount from the best bid and ask prices respectively at each tick. The results of this operation are stored in the *a_total* and *b_total* output columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 50.7 s, sys: 21.2 s, total: 1min 11s\n",
      "Wall time: 1min 11s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "tops = l2_orderbook_tops.loader.get_binance_tops(INPUT_DIR, SAMPLE_DATE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 13640551 entries, 15797 to 13656347\n",
      "Data columns (total 35 columns):\n",
      " #   Column   Dtype \n",
      "---  ------   ----- \n",
      " 0   ts       uint64\n",
      " 1   b_0      uint32\n",
      " 2   bq_0     uint32\n",
      " 3   b_1      uint32\n",
      " 4   bq_1     uint32\n",
      " 5   b_2      uint32\n",
      " 6   bq_2     uint32\n",
      " 7   b_3      uint32\n",
      " 8   bq_3     uint32\n",
      " 9   b_4      uint32\n",
      " 10  bq_4     uint32\n",
      " 11  b_5      uint32\n",
      " 12  bq_5     uint32\n",
      " 13  b_6      uint32\n",
      " 14  bq_6     uint32\n",
      " 15  b_7      uint32\n",
      " 16  bq_7     uint32\n",
      " 17  a_0      uint32\n",
      " 18  aq_0     uint32\n",
      " 19  a_1      uint32\n",
      " 20  aq_1     uint32\n",
      " 21  a_2      uint32\n",
      " 22  aq_2     uint32\n",
      " 23  a_3      uint32\n",
      " 24  aq_3     uint32\n",
      " 25  a_4      uint32\n",
      " 26  aq_4     uint32\n",
      " 27  a_5      uint32\n",
      " 28  aq_5     uint32\n",
      " 29  a_6      uint32\n",
      " 30  aq_6     uint32\n",
      " 31  a_7      uint32\n",
      " 32  aq_7     uint32\n",
      " 33  b_total  uint32\n",
      " 34  a_total  uint32\n",
      "dtypes: uint32(34), uint64(1)\n",
      "memory usage: 1.8 GB\n"
     ]
    }
   ],
   "source": [
    "tops.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>ts</th>\n",
       "      <th>b_0</th>\n",
       "      <th>bq_0</th>\n",
       "      <th>b_1</th>\n",
       "      <th>bq_1</th>\n",
       "      <th>b_2</th>\n",
       "      <th>bq_2</th>\n",
       "      <th>b_3</th>\n",
       "      <th>bq_3</th>\n",
       "      <th>b_4</th>\n",
       "      <th>bq_4</th>\n",
       "      <th>b_5</th>\n",
       "      <th>bq_5</th>\n",
       "      <th>b_6</th>\n",
       "      <th>bq_6</th>\n",
       "      <th>b_7</th>\n",
       "      <th>bq_7</th>\n",
       "      <th>a_0</th>\n",
       "      <th>aq_0</th>\n",
       "      <th>a_1</th>\n",
       "      <th>aq_1</th>\n",
       "      <th>a_2</th>\n",
       "      <th>aq_2</th>\n",
       "      <th>a_3</th>\n",
       "      <th>aq_3</th>\n",
       "      <th>a_4</th>\n",
       "      <th>aq_4</th>\n",
       "      <th>a_5</th>\n",
       "      <th>aq_5</th>\n",
       "      <th>a_6</th>\n",
       "      <th>aq_6</th>\n",
       "      <th>a_7</th>\n",
       "      <th>aq_7</th>\n",
       "      <th>b_total</th>\n",
       "      <th>a_total</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>15797</th>\n",
       "      <td>1596240060533</td>\n",
       "      <td>1136758</td>\n",
       "      <td>2083</td>\n",
       "      <td>1136739</td>\n",
       "      <td>1320</td>\n",
       "      <td>1136738</td>\n",
       "      <td>5391</td>\n",
       "      <td>1136705</td>\n",
       "      <td>6580</td>\n",
       "      <td>1136704</td>\n",
       "      <td>1004</td>\n",
       "      <td>1136703</td>\n",
       "      <td>6860</td>\n",
       "      <td>1136698</td>\n",
       "      <td>161</td>\n",
       "      <td>1136697</td>\n",
       "      <td>145</td>\n",
       "      <td>1136759</td>\n",
       "      <td>88</td>\n",
       "      <td>1136808</td>\n",
       "      <td>240</td>\n",
       "      <td>1136897</td>\n",
       "      <td>1141</td>\n",
       "      <td>1136914</td>\n",
       "      <td>2</td>\n",
       "      <td>1136941</td>\n",
       "      <td>1157</td>\n",
       "      <td>1136967</td>\n",
       "      <td>5</td>\n",
       "      <td>1136971</td>\n",
       "      <td>60</td>\n",
       "      <td>1136992</td>\n",
       "      <td>5</td>\n",
       "      <td>8794</td>\n",
       "      <td>328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15798</th>\n",
       "      <td>1596240060533</td>\n",
       "      <td>1136758</td>\n",
       "      <td>2083</td>\n",
       "      <td>1136739</td>\n",
       "      <td>1320</td>\n",
       "      <td>1136738</td>\n",
       "      <td>5391</td>\n",
       "      <td>1136705</td>\n",
       "      <td>6580</td>\n",
       "      <td>1136704</td>\n",
       "      <td>1004</td>\n",
       "      <td>1136703</td>\n",
       "      <td>6860</td>\n",
       "      <td>1136699</td>\n",
       "      <td>161</td>\n",
       "      <td>1136698</td>\n",
       "      <td>161</td>\n",
       "      <td>1136759</td>\n",
       "      <td>88</td>\n",
       "      <td>1136808</td>\n",
       "      <td>240</td>\n",
       "      <td>1136897</td>\n",
       "      <td>1141</td>\n",
       "      <td>1136914</td>\n",
       "      <td>2</td>\n",
       "      <td>1136941</td>\n",
       "      <td>1157</td>\n",
       "      <td>1136967</td>\n",
       "      <td>5</td>\n",
       "      <td>1136971</td>\n",
       "      <td>60</td>\n",
       "      <td>1136992</td>\n",
       "      <td>5</td>\n",
       "      <td>8794</td>\n",
       "      <td>328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15799</th>\n",
       "      <td>1596240060537</td>\n",
       "      <td>1136758</td>\n",
       "      <td>2083</td>\n",
       "      <td>1136739</td>\n",
       "      <td>1320</td>\n",
       "      <td>1136738</td>\n",
       "      <td>5391</td>\n",
       "      <td>1136705</td>\n",
       "      <td>6580</td>\n",
       "      <td>1136704</td>\n",
       "      <td>1004</td>\n",
       "      <td>1136703</td>\n",
       "      <td>6860</td>\n",
       "      <td>1136699</td>\n",
       "      <td>161</td>\n",
       "      <td>1136698</td>\n",
       "      <td>161</td>\n",
       "      <td>1136759</td>\n",
       "      <td>88</td>\n",
       "      <td>1136808</td>\n",
       "      <td>240</td>\n",
       "      <td>1136897</td>\n",
       "      <td>1141</td>\n",
       "      <td>1136914</td>\n",
       "      <td>2</td>\n",
       "      <td>1136941</td>\n",
       "      <td>1157</td>\n",
       "      <td>1136967</td>\n",
       "      <td>5</td>\n",
       "      <td>1136971</td>\n",
       "      <td>60</td>\n",
       "      <td>1136994</td>\n",
       "      <td>1004</td>\n",
       "      <td>8794</td>\n",
       "      <td>328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15800</th>\n",
       "      <td>1596240060537</td>\n",
       "      <td>1136758</td>\n",
       "      <td>2083</td>\n",
       "      <td>1136739</td>\n",
       "      <td>1320</td>\n",
       "      <td>1136738</td>\n",
       "      <td>5391</td>\n",
       "      <td>1136705</td>\n",
       "      <td>6580</td>\n",
       "      <td>1136704</td>\n",
       "      <td>1004</td>\n",
       "      <td>1136703</td>\n",
       "      <td>6860</td>\n",
       "      <td>1136699</td>\n",
       "      <td>161</td>\n",
       "      <td>1136698</td>\n",
       "      <td>161</td>\n",
       "      <td>1136759</td>\n",
       "      <td>88</td>\n",
       "      <td>1136808</td>\n",
       "      <td>240</td>\n",
       "      <td>1136897</td>\n",
       "      <td>1141</td>\n",
       "      <td>1136914</td>\n",
       "      <td>2</td>\n",
       "      <td>1136941</td>\n",
       "      <td>1157</td>\n",
       "      <td>1136958</td>\n",
       "      <td>120</td>\n",
       "      <td>1136967</td>\n",
       "      <td>5</td>\n",
       "      <td>1136971</td>\n",
       "      <td>60</td>\n",
       "      <td>8794</td>\n",
       "      <td>328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15801</th>\n",
       "      <td>1596240060541</td>\n",
       "      <td>1136758</td>\n",
       "      <td>2323</td>\n",
       "      <td>1136739</td>\n",
       "      <td>1320</td>\n",
       "      <td>1136738</td>\n",
       "      <td>5391</td>\n",
       "      <td>1136705</td>\n",
       "      <td>6580</td>\n",
       "      <td>1136704</td>\n",
       "      <td>1004</td>\n",
       "      <td>1136703</td>\n",
       "      <td>6860</td>\n",
       "      <td>1136699</td>\n",
       "      <td>161</td>\n",
       "      <td>1136698</td>\n",
       "      <td>161</td>\n",
       "      <td>1136759</td>\n",
       "      <td>88</td>\n",
       "      <td>1136808</td>\n",
       "      <td>240</td>\n",
       "      <td>1136897</td>\n",
       "      <td>1141</td>\n",
       "      <td>1136914</td>\n",
       "      <td>2</td>\n",
       "      <td>1136941</td>\n",
       "      <td>1157</td>\n",
       "      <td>1136958</td>\n",
       "      <td>120</td>\n",
       "      <td>1136967</td>\n",
       "      <td>5</td>\n",
       "      <td>1136971</td>\n",
       "      <td>60</td>\n",
       "      <td>9034</td>\n",
       "      <td>328</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  ts      b_0  bq_0      b_1  bq_1      b_2  bq_2      b_3  \\\n",
       "15797  1596240060533  1136758  2083  1136739  1320  1136738  5391  1136705   \n",
       "15798  1596240060533  1136758  2083  1136739  1320  1136738  5391  1136705   \n",
       "15799  1596240060537  1136758  2083  1136739  1320  1136738  5391  1136705   \n",
       "15800  1596240060537  1136758  2083  1136739  1320  1136738  5391  1136705   \n",
       "15801  1596240060541  1136758  2323  1136739  1320  1136738  5391  1136705   \n",
       "\n",
       "       bq_3      b_4  bq_4      b_5  bq_5      b_6  bq_6      b_7  bq_7  \\\n",
       "15797  6580  1136704  1004  1136703  6860  1136698   161  1136697   145   \n",
       "15798  6580  1136704  1004  1136703  6860  1136699   161  1136698   161   \n",
       "15799  6580  1136704  1004  1136703  6860  1136699   161  1136698   161   \n",
       "15800  6580  1136704  1004  1136703  6860  1136699   161  1136698   161   \n",
       "15801  6580  1136704  1004  1136703  6860  1136699   161  1136698   161   \n",
       "\n",
       "           a_0  aq_0      a_1  aq_1      a_2  aq_2      a_3  aq_3      a_4  \\\n",
       "15797  1136759    88  1136808   240  1136897  1141  1136914     2  1136941   \n",
       "15798  1136759    88  1136808   240  1136897  1141  1136914     2  1136941   \n",
       "15799  1136759    88  1136808   240  1136897  1141  1136914     2  1136941   \n",
       "15800  1136759    88  1136808   240  1136897  1141  1136914     2  1136941   \n",
       "15801  1136759    88  1136808   240  1136897  1141  1136914     2  1136941   \n",
       "\n",
       "       aq_4      a_5  aq_5      a_6  aq_6      a_7  aq_7  b_total  a_total  \n",
       "15797  1157  1136967     5  1136971    60  1136992     5     8794      328  \n",
       "15798  1157  1136967     5  1136971    60  1136992     5     8794      328  \n",
       "15799  1157  1136967     5  1136971    60  1136994  1004     8794      328  \n",
       "15800  1157  1136958   120  1136967     5  1136971    60     8794      328  \n",
       "15801  1157  1136958   120  1136967     5  1136971    60     9034      328  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tops.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "tops['spread']  = tops['a_0'] - tops['b_0']\n",
    "tops['mid_price'] = tops['b_0'] + tops['spread']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "tops['b_qty'] = tops[[c for c in tops.columns if c.startswith('bq_')]].sum(axis=1)\n",
    "tops['a_qty'] = tops[[c for c in tops.columns if c.startswith('aq_')]].sum(axis=1)\n",
    "tops['tops_imbalance_delta'] = ((tops['b_qty'] - tops['a_qty']) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "sample = tops.tail(100000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pylab import rcParams\n",
    "rcParams['figure.figsize'] = 10, 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x216 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = (sample['mid_price']/100).plot(label='mid', color='black')\n",
    "\n",
    "ax2 = ax.twinx()\n",
    "(sample['tops_imbalance_delta']/1000).ewm(500).mean().plot(ax=ax2, color='green')\n",
    "\n",
    "ax.figure.legend()\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "py36",
   "language": "python",
   "name": "py36"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}