quantworks 0.21

Python Algorithmic Trading Framework

QuantWorks

QuantWorks is an event driven algorithmic trading framework. It is a fork of PyAlgoTrade (see Motivation).

QuantWorks provides a Python API for strategy authoring, backtesting, paper trading, and of course live trading via the Broker interface.

To get started using QuantWorks, please take a look at the original PyAlgoTrade tutorial and the full documentation.

Main Features

• Python 3 development
  • Python 2 support is NOT guaranteed in any capacity.
• Event driven.
• Supports Market, Limit, Stop and StopLimit orders.
• Supports any type of time-series data in Pandas or CSV format (like Yahoo! Finance, Google Finance, Quandl and NinjaTrader), as well as database (i.e. sqlite).
• Technical indicators and filters like SMA, WMA, EMA, RSI, Bollinger Bands, Hurst exponent and others.
• Performance metrics like Sharpe ratio and drawdown analysis.
• Event profiler.
• TA-Lib integration.

Motivation

QuantWorks is a fork of PyAlgoTrade by @gbeced. This project aims to be:

• Modern: first-class Python 3 development (Python 2 is EOL as of 2020)
• Extensible: as a framework, robust extension support is a must, and we encourage users of QuantWorks to give back by publishing their extensions (see Extensions)
• Easy to Develop: state-of-the-art tooling (pytest, poetry, travis) and approachable design principles should make it easy for newcomers to contribute.
• Open: as a fork of an Apache 2.0 license project, QuantWorks maintains the spirit of FOSS development. CONTRIBUTING.md forthcoming

Development

QuantWorks is developed and tested using 3.7 and depends on:

• NumPy and SciPy.
• pytz.
• dateutil.
• requests.
• matplotlib for plotting support.
• ws4py for Bitstamp support.
• tornado for Bitstamp support.
• tweepy for Twitter support.

Developer ergonomics are provided by

• poetry
• pytest
• tox
• travis-ci

Extensions

• Bitstamp (bitcoin) live trading is implemented by the quantworks-bitstamp package (https://pypi.org/project/quantworks-bitcoin/)
• Twitter real-time feeds are supported via the quantworks-twitter package (https://pypi.org/project/quantworks-twitter/)