Hands-On Machine Learning for Algorithmic Trading

The architecture – event-driven trading simulation

A zipline algorithm will run for a specified period after an initial setup and executes its trading logic when specific events occur. These events are driven by the trading frequency and can also be scheduled by the algorithm, and result in zipline calling certain methods. The algorithm maintains state through a context dictionary and receives actionable information through a data variable containing point-in-time (PIT) current and historical data. The algorithm returns a DataFrame containing portfolio performance metrics if there were any trades, as well as user-defined metrics that can be used to record, for example, the factor values.

You can execute an algorithm from the command line, ...