BVARSV.fit

BVARSV.fit(*, n_draws=2000, n_burn=1000, thin=2, k_vol=0.01, seed=None)

Estimate by Gibbs with the KSC volatility block.

Parameters:

• n_draws (int) – Total sampler iterations.

• n_burn (int) – Burn-in iterations discarded.

• thin (int) – Keep every thin-th post-burn draw.

• k_vol (float) – Prior scale of the log-volatility random walk; raise toward 0.1 when volatility drifts substantially.

• seed (int | Generator | None) – Seed or generator, for reproducibility.

Return type:

BVARSVResult

Returns:

The fitted BVARSVResult.

Raises:

• SpecificationError – If the draw bookkeeping is inconsistent or the prior is unusable.

• NumericalError – If a conditional draw collapses.