Vasicek Beta Shrinkage

Why SurgeFlow pulls noisy stock betas toward a cross-sectional prior, and how model fit adjusts the amount of shrinkage.

Category: Factor Research

Why shrink a beta

Raw OLS betas are noisy, especially for short samples or low-fit stocks. Vasicek shrinkage pulls each stock's beta toward the cross-sectional average beta, weighting by how precise the stock's own estimate is versus the spread of betas across stocks.

How this system computes it

The prior is the per-market average beta. The shrinkage weight is the stock's estimate precision relative to that prior; SurgeFlow also computes a fit-adjusted weight that gives a poorly-fit regression less of its own beta. The build verifies that high-fit stocks keep nearly all their own beta while low-fit stocks are pulled toward the market average. A regime variant weights betas by macro-regime probabilities. - Cross-sectional market prior - Precision-weighted blend - Fit-adjusted variant - Regime-conditional betas

How to read it

The shrunk beta is more stable and trustworthy than raw OLS for noisy names — a low-fit stock's beta is dragged toward the market's roughly 1.0. Regime betas show how a stock's market sensitivity differs across expansion versus stress states.