Python for Finance, 3rd Edition

Chapter 23. Vectorized and Event-Based Backtesting

A Note for Early Release Readers

With Early Release ebooks, you get books in their earliest form—the author’s raw and unedited content as they write—so you can take advantage of these technologies long before the official release of these titles.

This will be the 23th chapter of the final book. Please note that the GitHub repo will be made active later on.

If you’d like to be actively involved in reviewing and commenting on this draft, please reach out to the editor at [e-mail].

Vectorized backtests in numpy and pandas are powerful tools for evaluating trading ideas quickly, but professional evaluation also requires an event-based view that mirrors how markets and brokers actually process orders and trades. This chapter develops both layers side by side, using a simple machine learning baseline strategy to connect signals, positions, trades, and performance metrics.

This chapter builds a bridge from EMH-style hypotheses to concrete backtests. It starts from a buy-and-hold baseline for a single liquid instrument, adds a simple logistic-regression strategy for daily direction, and implements both a vectorized and a minimal event-based backtest. The event-based layer deliberately reuses the same notions of prices, returns, signals, orders, and account state as the market and broker engine in Chapter 24, reinforcing the idea that serious backtests must closely resemble the execution environment they stand in for. The ...