Overfitting Prevention

What problem does it solve?

Backtests can overstate performance due to data mining and excessive parameter tuning. This guide helps quantify and detect overfitting in quantitative research so you can distinguish genuine alpha from artifacts.

Core Features & Use Cases

• Deflated Sharpe Ratio (DSR) adjustments for multiple testing and skew/kurtosis folds into the score.
• Combinatorial Purged Cross-Validation (CPCV) for robust out-of-sample evaluation across many paths.
• White's Reality Check and Hansen's SPA Test to validate the significance of the best-performing strategies.
• Minimum Backtest Length (MBL) guidance and degrees-of-freedom penalties to assess statistical reliability.
• End-to-end overfitting detection checklist and pragmatic strategy-acceptance criteria.

Quick Start

Run a comprehensive overfitting assessment on your backtest results by applying DSR, CPCV, MB, and SPA/Reality Check tests across your strategy universe.