Site icon Premium Alpha

Machine Studying Classification and Portfolio Building

Machine Studying Classification and Portfolio Building


Classification outperforms regression throughout matched machine studying fashions in portfolio development. A stacking ensemble of gradient boosted bushes, random forest, and neural community yields a value-weighted annualized Sharpe ratio of two.08 for classification and 1.39 for regression. This outperformance strengthens with class granularity and persists throughout subsamples and after transaction prices. Spanning checks present that classification retains economically massive alphas after we management for regression, whereas regression alphas shrink considerably as soon as we management for classification. These outcomes point out that classification extracts extra return data than matched regression. Our diagnostics hint classification’s benefit to extra exact separation of return deciles.



Source link

Exit mobile version