Importance sampling correction versus standard averages of reversible MCMCs in terms of the asymptotic variance
arXiv:1706.09873 · doi:10.1016/j.spa.2020.05.006
Abstract
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.
arXiv comment27 pages
References in corpus (20)
- The pseudo-marginal approach for efficient Monte Carlo computations
- On the ergodicity properties of some adaptive MCMC algorithms
- Batch means and spectral variance estimators in Markov chain Monte Carlo
- On the utility of graphics cards to perform massively parallel simulation of advanced Monte Carlo methods
- Renewal theory and computable convergence rates for geometrically ergodic Markov chains
- Harris recurrence of Metropolis-within-Gibbs and trans-dimensional Markov chains
- Explicit error bounds for Markov chain Monte Carlo
- Piecewise-Deterministic Markov Chain Monte Carlo
- Convergence properties of pseudo-marginal Markov chain Monte Carlo algorithms
- Scalable posterior approximations for large-scale Bayesian inverse problems via likelihood-informed parameter and state reduction
- Delayed acceptance particle MCMC for exact inference in stochastic kinetic models
- Speeding Up MCMC by Delayed Acceptance and Data Subsampling
- A Noisy Monte Carlo Algorithm
- Comparison of asymptotic variances of inhomogeneous Markov chains with application to Markov chain Monte Carlo methods
- A vanilla Rao--Blackwellization of Metropolis--Hastings algorithms
- Importance sampling type estimators based on approximate marginal MCMC
- Approximations of Geometrically Ergodic Reversible Markov Chains
- On the utility of Metropolis-Hastings with asymmetric acceptance ratio
- Optimal approximating Markov chains for Bayesian inference
- A Hierarchical Multilevel Markov Chain Monte Carlo Algorithm with Applications to Uncertainty Quantification in Subsurface Flow
Cited by in corpus (4)
- Importance sampling type estimators based on approximate marginal MCMC
- bssm: Bayesian Inference of Non-linear and Non-Gaussian State Space Models in R
- Accelerating delayed-acceptance Markov chain Monte Carlo algorithms
- Sampling algorithms in statistical physics: a guide for statistics and machine learning