Sequential Markov Chain Monte Carlo

Abstract

We propose a sequential Markov chain Monte Carlo (SMCMC) algorithm to sample from a sequence of probability distributions, corresponding to posterior distributions at different times in on-line applications. SMCMC proceeds as in usual MCMC but with the stationary distribution updated appropriately each time new data arrive. SMCMC has advantages over sequential Monte Carlo (SMC) in avoiding particle degeneracy issues. We provide theoretical guarantees for the marginal convergence of SMCMC under various settings, including parametric and nonparametric models. The proposed approach is compared to competitors in a simulation study. We also consider an application to on-line nonparametric regression.

Cited by in corpus (6)

• On the evaluation of uncertainties for state estimation with the Kalman filter
• Bayesian Conditional Density Filtering
• Fast Bayesian Record Linkage for Streaming Data Contexts
• Bayesian Dynamic Quantile Model Averaging
• Apple Tasting Revisited: Bayesian Approaches to Partially Monitored Online Binary Classification
• $k$-means: Fighting against Degeneracy in Sequential Monte Carlo with an Application to Tracking