Quantitative approach to trading is done via applying mathematical models to various financial instruments. In order to get money for you strategy, mathematical model beneath it should be sound. And to prove that this model worth money one should do proper backtesting. This project aims to provide easy and straitforward backtesting solution.
There are number of python projects for backtesting: backtrader, pyalgotrade, zipline, rqalpha, etc.. When i was trying out them, i was dissatisfied with one or more of the following: event driven, unnecessary complex architecture, no support for trading multiple instruments in convinient way, no proper performance evaluation, etc.. This project solves those issues at cost of not so wide functionality compared to mentioned ones above. Project is designed to be easily build on top of it.
- Data manipulations are made with pandas.
- Backtesting operations are vector( no loops, not event driven).
- Extensive statistical evaluation of strategies.
- Number of visualizations embedded.
- Strategy robustness tests.
- API to work with OHLC data( download, prepare).
- Clean and straitforward project structure.
- PEP8 compliant code.
- Install via setup.py:
git clone git@github.com:bluella/stbt.git
cd stbt
python setup.py install- Install via pip:
pip install stbt- Run tests:
pip install pytest
pytestimport datetime as dt
import pandas as pd
import matplotlib.pyplot as plt
from stbt.simulator import Strategy
from stbt.download_ohlc.cryptocurrency import get_ohlc_cryptocompare
from stbt.operators.technical import skewness
# get trading data from cryptocompare
BTC_TICKER = 'BTC'
ETH_TICKER = 'ETH'
USD_TICKER = 'USD'
END_DATE = dt.datetime(2018, 7, 1, 0, 0, 0)
START_DATE = dt.datetime(2018, 3, 1, 0, 0, 0)
OHLC_BTC = get_ohlc_cryptocompare(BTC_TICKER, USD_TICKER, START_DATE,
end_date=END_DATE, interval_key='day')
OHLC_ETH = get_ohlc_cryptocompare(ETH_TICKER, USD_TICKER, START_DATE,
end_date=END_DATE, interval_key='day')
# create dfs in format that Strategy requires
closes_df = pd.concat([OHLC_BTC['Close'], OHLC_ETH['Close']],
axis=1, keys=['BTC', 'ETH'])
# use imported indicator to create weights
weights_df = skewness(closes_df)
# create strategy
s = Strategy(closes_df, weights_df, cash=100)
# run backtest, robust tests, calculate stats
s.run_all(delay=2, verify_data_integrity=True, instruments_drop=None,
commissions_const=0, capitalization=False)
# check strategy stats
print(s.stats_dict)
# save strategy to futher comparison
s.add_to_pnls_pool()
# plot pool correlation heatmap
heatmap_fig, corr_matrix = s.get_pool_heatmap()
plt.show()
- Improve test coverage.
- More API for data download.
- More technical indicators.
- Portfolio optimization tools.
See CHANGELOG.
This project is licensed under the MIT License - see the LICENSE for details.