Methods // Bayesian Calibration

Bayesian Calibration for Synthetic Regimes

Set priors: Define constraints based on policy, liquidity, or risk limits.
Update parameters: Fit priors to observed moments and target metrics.
Lock manifests: Store posterior parameters for replay and review.

Calibrate synthetic regimes against priors, constraints, and evidence so your scenarios remain realistic while still exploring tail-risk territory.

Why Calibration Matters

Ensure synthetic regimes stay anchored to observable market behavior.

Quantify confidence bounds rather than relying on point estimates.

Store priors, updates, and posterior results for governance review.

Keep synthetic scenarios grounded while expanding coverage of rare regimes.