Block sampling the discrete latent variable in dynamic mixture models
Gabriele Fiorentini, Christophe Planas, Alessandro Rossi · 2011
We propose two algorithms for block-sampling the discrete latent variable in dynamic mixture models. The first is a multi-move extension of the single-move Gibbs sampler devised by Gerlach, Carter and Kohn (2000). The second is an adaptive Metropolis-Hastings scheme that performs well even when the number of discrete states is large. Three empirical examples illustrate the gain in eciency achieved. We also show that visual inspection of sample partial autocorrelations of the discrete latent variable helps anticipating whether blocking can be eective.