Importance Tempering

Abstract

Simulated tempering (ST) is an established Markov chain Monte Carlo (MCMC) method for sampling from a multimodal density $\pi (\theta )$. Typically, ST involves introducing an auxiliary variable $k$ taking values in a finite subset of $[0,1]$ and indexing a set of tempered distributions, say ${\pi }_k(\theta )\propto \pi (\theta {)}^k$. In this case, small values of $k$ encourage better mixing, but samples from $\pi$ are only obtained when the joint chain for $(\theta ,k)$ reaches $k=1$. However, the entire chain can be used to estimate expectations under $\pi$ of functions of interest, provided that importance sampling (IS) weights are calculated. Unfortunately this method, which we call importance tempering (IT), can disappoint. This is partly because the most immediately obvious implementation is naïve and can lead to high variance estimators. We derive a new optimal method for combining multiple IS estimators and prove that the resulting estimator has a highly desirable property related to the notion of effective sample size. We briefly report on the success of the optimal combination in two modelling scenarios requiring reversible-jump MCMC, where the naïve approach fails.

16 pages, 2 tables, significantly shortened from version 4 in response to referee comments, to appear in Statistics and Computing

Cited by in corpus (17)

• A population Monte Carlo scheme with transformed weights and its application to stochastic kinetic models
• Modified Hamiltonian Monte Carlo for Bayesian inference
• Chain ladder method: Bayesian bootstrap versus classical bootstrap
• Linear Complexity Gibbs Sampling for Generalized Labeled Multi-Bernoulli Filtering
• Joint search for isolated sources and an unresolved confusion background in pulsar timing array data
• Scalable Importance Tempering and Bayesian Variable Selection
• Ecological non-linear state space model selection via adaptive particle Markov chain Monte Carlo (AdPMCMC)
• Post-Processing of MCMC
• Variable selection and sensitivity analysis using dynamic trees, with an application to computer code performance tuning
• A Bayesian regression tree approach to identify the effect of nanoparticles' properties on toxicity profiles
• Tuning Tempered Transitions
• Computing Bayes: From Then 'Til Now'
• A Bayesian parameter estimation approach to pulsar time-of-arrival analysis
• Split Sampling: Expectations, Normalisation and Rare Events
• Revisiting the balance heuristic for estimating normalising constants
• Adaptive Importance Sampling meets Mirror Descent: a Bias-variance tradeoff
• RMCMC: A System for Updating Bayesian Models