Variance reduction for additive functionals of Markov chains via martingale representations

DOI10.1007/S11222-021-10073-ZMaRDI QIDQ2114045zbMATH OpenOpenAlexFDO

Authors D. Belomestny, Eric Moulines, S. P. Samsonov

Publication date 14 March 2022

Published in Statistics and Computing (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1903.07373

zbMATH Keywords

MCMCvariance reductionMarkov chainsmartingale representations

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08)

Abstract: In this paper we propose an efficient variance reduction approach for additive functionals of Markov chains relying on a novel discrete time martingale representation. Our approach is fully non-asymptotic and does not require the knowledge of the stationary distribution (and even any type of ergodicity) or specific structure of the underlying density. By rigorously analyzing the convergence properties of the proposed algorithm, we show that its cost-to-variance product is indeed smaller than one of the naive algorithm. The numerical performance of the new method is illustrated for the Langevin-type Markov Chain Monte Carlo (MCMC) methods.

Recommendations

• Variance reduction for Markov chains with application to MCMC
• Variance reduction for dependent sequences with applications to stochastic gradient MCMC
• Variance reduction for Metropolis-Hastings samplers
• Variance reduction for discretised diffusions via regression
• Variance reduction for diffusions

Cites work

• scientific article; zbMATH DE number 1999206 (Why is no real title available?)
• scientific article; zbMATH DE number 3999929 (Why is no real title available?)
• A Multivariate Faa di Bruno Formula with Applications
• A distribution-free theory of nonparametric regression
• Adaptive simulation using perfect control variates
• An adaptive scheme for the approximation of dissipative systems
• Asymptotic behaviour of the stochastic Lotka-Volterra model.
• Concentration inequalities. A nonasymptotic theory of independence
• Control Variates for Estimation Based on Reversible Markov Chain Monte Carlo Samplers
• Control functionals for Monte Carlo integration
• Dynamic analysis of stochastic Lotka-Volterra predator-prey model with discrete delays and feedback control
• Ergodicity for SDEs and approximations: locally Lipschitz vector fields and degenerate noise.
• Markov Chains
• Monte Carlo Methods for Applied Scientists
• Monte Carlo complexity of parametric integration
• Monte-Carlo methods and stochastic processes. From linear to non-linear
• Nonasymptotic convergence analysis for the unadjusted Langevin algorithm
• Preliminary control variates to improve empirical regression methods
• Rates of convergence of the Hastings and Metropolis algorithms
• Recursive computation of the invariant distribution of a diffusion
• Simulation and the Monte Carlo method
• Theoretical Guarantees for Approximate Sampling from Smooth and Log-Concave Densities
• Variance reduction for Markov chains with application to MCMC
• Variance reduction for discretised diffusions via regression
• Weighted multilevel Langevin simulation of invariant measures
• Zero variance Markov chain Monte Carlo for Bayesian estimators

Cited in

(3)

• Variance reduction for dependent sequences with applications to stochastic gradient MCMC
• Variance reduction for Markov chains with application to MCMC
• Variational principles for asymptotic variance of general Markov processes

This page was built for publication: Variance reduction for additive functionals of Markov chains via martingale representations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2114045)