Out-of-sample testing is a critical component of designing and evaluating trading systems.
I stopped considering Out-of-Sample as a strategic validation tool when I started using it primarily as an invalidation criterion. I find it extremely useful to question and refute some of my more questionable work.
Just like backtest, you still need to use OOS with caution. Incorporating it into a WFA improves the results.
I stopped considering Out-of-Sample as a strategic validation tool when I started using it primarily as an invalidation criterion. I find it extremely useful to question and refute some of my more questionable work.
Just like backtest, you still need to use OOS with caution. Incorporating it into a WFA improves the results.