Algo & backtesting
Walk-forward analysis
Also known as: walk forward, out-of-sample testing
Walk-forward analysis optimises a strategy on one window of data and tests it on the following unseen window, repeatedly rolling forward.
By always validating on out-of-sample data, it gives a more honest estimate of live performance than a single backtest and helps detect overfitting. A strategy that holds up across many walk-forward windows is more likely to be robust.
Related terms
Build strategies around walk-forward analysis without code — drag nodes onto the Algovex canvas, backtest, and export.
Build a strategy free →