Importance sampling correction versus standard averages of reversible MCMCs in terms of the asymptotic variance
Franks, J., & Vihola, M. (2020). Importance sampling correction versus standard averages of reversible MCMCs in terms of the asymptotic variance. Stochastic Processes and Their Applications, 130(10), 6157-6183. https://doi.org/10.1016/j.spa.2020.05.006
Published inStochastic Processes and Their Applications
© 2020 Elsevier BV
We establish an ordering criterion for the asymptotic variances of two consistent Markov chain Monte Carlo (MCMC) estimators: an importance sampling (IS) estimator, based on an approximate reversible chain and subsequent IS weighting, and a standard MCMC estimator, based on an exact reversible chain. Essentially, we relax the criterion of the Peskun type covariance ordering by considering two different invariant probabilities, and obtain, in place of a strict ordering of asymptotic variances, a bound of the asymptotic variance of IS by that of the direct MCMC. Simple examples show that IS can have arbitrarily better or worse asymptotic variance than Metropolis–Hastings and delayed-acceptance (DA) MCMC. Our ordering implies that IS is guaranteed to be competitive up to a factor depending on the supremum of the (marginal) IS weight. We elaborate upon the criterion in case of unbiased estimators as part of an auxiliary variable framework. We show how the criterion implies asymptotic variance guarantees for IS in terms of pseudo-marginal (PM) and DA corrections, essentially if the ratio of exact and approximate likelihoods is bounded. We also show that convergence of the IS chain can be less affected by unbounded high-variance unbiased estimators than PM and DA chains. ...
ISSN Search the Publication Forum0304-4149
Publication in research information system
MetadataShow full item record
Related funder(s)Academy of Finland
Funding program(s)Academy Research Fellow, AoF; Research costs of Academy Research Fellow, AoF
Additional information about fundingSupport has been provided for JF and MV from the Academy of Finland (grants 274740, 284513 and 312605), and for JF from The Alan Turing Institute. JF thanks the organisers of the 2017 SMC course and workshop in Uppsala.
Showing items with similar title or keywords.
Importance sampling type estimators based on approximate marginal Markov chain Monte Carlo Vihola, Matti; Helske, Jouni; Franks, Jordan (Wiley-Blackwell, 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 ...
Coupled conditional backward sampling particle filter Lee, Anthony; Singh, Sumeetpal S.; Vihola, Matti (Institute of Mathematical Statistics, 2020)The conditional particle filter (CPF) is a promising algorithm for general hidden Markov model smoothing. Empirical evidence suggests that the variant of CPF with backward sampling (CBPF) performs well even with long time ...
On the use of approximate Bayesian computation Markov chain Monte Carlo with inflated tolerance and post-correction Vihola, Matti; Franks, Jordan (Oxford University Press, 2020)Approximate Bayesian computation enables inference for complicated probabilistic models with intractable likelihoods using model simulations. The Markov chain Monte Carlo implementation of approximate Bayesian computation ...
Statistical analysis of life sequence data Helske, Satu (University of Jyväskylä, 2016)
On resampling schemes for particle filters with weakly informative observations Chopin, Nicolas; Singh, Sumeetpal S.; Soto, Tomás; Vihola, Matti (Institute of Mathematical Statistics, 2022)We consider particle filters with weakly informative observations (or ‘potentials’) relative to the latent state dynamics. The particular focus of this work is on particle filters to approximate time-discretisations of ...