SVPosterior.posterior_replications#

SVPosterior.posterior_replications(n_replications=200, *, seed=None)[source]#

Replicated series, one per posterior draw used.

Each replication takes one kept draw of (mu, phi, sigma2, mean) – and of nu or rho where the posterior carries them – simulates a fresh log-variance path from its stationary start, and the series from it. Draws are taken evenly across the kept sequence when fewer replications than draws are asked for.

Parameters:

  • n_replications (int) – Replicated series.

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

Return type:

NDArray[double]

Returns:

An array of shape (n_replications, n) aligned with observed.

Raises:

SpecificationError – If the count is not positive.