The Evolution of Walk-Forward Backtesting
梳理了逐步向前回测的构建过程，从简单向量化方法到事件驱动架构，并介绍函数式响应式编程带来的额外优势，适合量化交易者参考。
Many papers discuss what should be included in a walk-forward backtest, but few focus on how to go about building one. This article aims to demystify the processes and consolidate the various concepts. It begins with an overview of the professional systematic trading workflow, highlighting the functionalities that various backtesting patterns should include. Three development patterns are examined, guiding the reader from a simple vectorized approach via for-loop iterative backtesting to cutting-edge event-driven architectures. Finally, an alternative approach based on functional reactive programming is introduced, and the article illustrates how this paradigm offers additional benefits for walk-forward backtesting.