Skip to content

Latest commit

 

History

History
151 lines (123 loc) · 5.21 KB

README.md

File metadata and controls

151 lines (123 loc) · 5.21 KB
Raw

crypto-strategy

A repository to perform backtests and create trading strategies for cryptocurrencies.

Dependency Review Python package pypi-upload

Install

pip install crypto-strategy[full]

Usage

Backtest Strategy

  1. Moving Average Strategy

  • Description

    BestMaStrategy(symbols, freq, res_dir, flag_filter, flag_stop)
    • symbols: asset name, e.g., BTCUSDT
    • freq: data frequency to use, 1h | 4h
    • res_dir: results directory
    • flag_filter: filter to use, [mmi | ang]
      • mmi: Market Meanness Index filter
      • ang: Linear Regression Angle filter
    • flag_stop: early stop flag, [ts_stop | sl_stop | tp_stop]
      • ts_stop: trailing stop
      • sl_stop: stop loss
      • tp_stop: take profit

  • Example: Find the best params using 1h data with mmi filter and ts_stop

    BestMaStrategy(
    symbols=['BTCUSDT', 'ETHUSDT'], 
    freq='1h', 
    res_dir='results/best-1h-ma-mmi-filter', 
    flag_filters='mmi',
    flag_stop='ts_stop'
)
  1. Breakout Strategy

    • Description: A method to optimize the BO strategy

      • symbols: a list of symbols to be optimzied on, e.g., 'BTCUSDT'
      • freq: currently supported values: '1h' or '4h'
      • res_dir: the output directory
      • flag_filter: currently supported filters: 'vol', 'ang', default: None
      • flag_stop: early stop flag, [ts_stop | sl_stop | tp_stop]
        • ts_stop: trailing stop
        • sl_stop: stop loss
        • tp_stop: take profit

    • Example: Find the best params for bo strategy with vol filter using BTCUSDT 4h data and sl_stop

      BestBoStrategy(
    symbols = 'BTCUSDT',
    freq = '4h', 
    res_dir = 'results/best-4h-bo_rev-vol-filter', 
    flag_filters = 'vol',
    flag_stop = 'sl_stop',
)
  2. MACD Strategy
BestMacdStrategy(symbols, freq, res_dir, flag_filter)
  • symbols: asset name, e.g., BTCUSDT
  • freq: data frequency to use, 1h | 4h
  • res_dir: results directory
  • flag_filter: filter to use, [mmi | ang | stoch | sma]
    • vol: Volume filter
    • ang: Linear Regression Angle filter

Inspect Strategy

  1. MA Strategy

    • Description: A method to inspect the MA strategy with given params 
      InspectMaStrategy(symbols, freq, name, n1, n2, timeperiod, threshold, flag_filter, stop_vars)
      • symbol: the name of the crypto, e.g., 'BTCUSDT'
      • freq: currently supported values are '1h' or '4h'
      • name, n1, n2: the name and the params of the ma strategy, e.g., 'sma', 100, 50
      • timeperiod: param used in either mmi or ang filter
      • threshold: param used in ang filter
      • flag_filter: currently supported fitlers: 'mmi', 'ang', default: None
      • stop_vars: dictionary of stop vars, currently support 'ts_stop', 'sl_stop', 'tp_stop', default None
    • Example: Inspect MA strategy linear_reg with n1=30 & n2=280 and sl_stop=0.1 & tp_stop=0.1 
      InspectMaStrategy(
    symbols, 
    freq='4h', 
    name='linear_reg', n1=30, n2=280, 
    stop_vars={'sl_stop':0.1, 'tp_stop':0.1})

  2. BO Strategy

    • Description: A method to inspect the BO strategy

      InspectBoStrategy(symbol, freq, long_window, short_window, ts_stop, timeperiod, multiplier, threshold, flag_filter, flag_ts_stop)
      • symbols: a list of symbols to be optimzied on, e.g., 'BTCUSDT'
      • freq: currently supported values: '1h' or '4h'
      • long_window, short_window: breakout params
      • flag_filter: currently supported fitlers: 'vol', 'ang', default: None
      • timeperiod, multiplier: volume filter params
      • timeperiod, threshold: angle filter params
      • stop_vars: dictionary of stop vars, currently support 'ts_stop', 'sl_stop', 'tp_stop', default None

    • Example: Inspect 4h BTCUSDT breakout strategy with volume filter and trailing stop

      InspectBoStrategy(
    'BTCUSDT', 
    freq='4h', 
    long_window=100, short_window=50,
    flag_filter='vol', timeperiod=20, multiplier=2,
    stop_vars={'ts_stop':0.1})

CLI

Backtests can also be carried out in command line. To find out more

crypto --help

Example 1: Find the best params for BO strategy with vol filter using 4h data

crypto best-bo-strategy -f 4h -r results/best-4h-bo-vol-filter -g vol -e bo

Example 2: Find the best params for MA strategy with mmi filter and ts_stop using 1h data

crypto best-ma-strategy -f 1h -r results/best-1h-ma-mmi-filter -g mmi -t ts_stop

Tests

pytest