Introduction

PyAlgoTrade is an event driven algorithmic trading Python library that I started with one goal in mind: To make it easy to backtest stock trading strategies. This is, given a certain amount of historical data for one stock, I want to check how a certain stock trading strategy behaves.

Paper trading support is in its early stages right now and there are two real time bar feeds supported:

• Bitstamp Example
• Xignite Example

Although initially designed for backtesting purposes, the design should adapt real trading scenarios and future releases will gradually support this.

It should also make it easy to optimize a strategy using multiple computers.

PyAlgoTrade is developed using Python 2.7 and depends on:

• NumPy and SciPy (http://numpy.scipy.org/).
• pytz (http://pytz.sourceforge.net/).
• matplotlib (http://matplotlib.sourceforge.net/) for plotting support.
• ws4py (https://github.com/Lawouach/WebSocket-for-Python) for Bitstamp support.
• tornado (http://www.tornadoweb.org/en/stable/) for Bitstamp support.
• tweepy (https://github.com/tweepy/tweepy) for Twitter support.

so you need to have those installed in order to use this library.

You can install PyAlgoTrade using pip like this:

pip install pyalgotrade