Importance sampling type estimators based on approximate marginal Markov chain Monte Carlo

By Matti Vihola, Jouni Helske and Jordan Franks in Bayesian Inference

September 3, 2020



We consider importance sampling (IS) type weighted estimators based on Markov chain Monte Carlo (MCMC) targeting an approximate marginal of the target distribution. In the context of Bayesian latent variable models, the MCMC typically operates on the hyper parameters, and the subsequent weighting may be based on IS or sequential Monte Carlo (SMC), but allows for multilevel techniques as well. The IS approach provides a natural alternative to delayed acceptance (DA) pseudo-marginal/particle MCMC, and has many advantages over DA, including a straightforward parallelisation and additional flexibility in MCMC implementation. We detail minimal conditions which ensure strong consistency of the suggested estimators, and provide central limit theorems with expressions for asymptotic variances. We demonstrate how our method can make use of SMC in the state space models context, using Laplace approximations and time-discretised diffusions. Our experimental results are promising and show that the IS type approach can provide substantial gains relative to an analogous DA scheme, and is often competitive even without parallelisation.

Posted on:
September 3, 2020
1 minute read, 163 words
Bayesian Inference
Markov Chain Monte Carlo Sequential Monte Carlo
See Also:
Estimating Causal Effects from Panel Data with Dynamic Multivariate Panel Models
Spatio-temporal modeling of co-dynamics of smallpox, measles and pertussis in pre-healthcare Finland
dynamite: An R Package for Dynamic Multivariate Panel Models