Rethinking the Effective Sample Size

DOI10.1111/INSR.12500MaRDI QIDQ6067598zbMATH OpenOpenAlexFDO

Authors Víctor Elvira, Luca Martino, Christian Robert

Publication date 14 December 2023

Published in International Statistical Review (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1809.04129

zbMATH Keywords

Bayesian inferenceeffective sample sizeimportance samplingMonte Carlo methods

Mathematics Subject Classification ID

Sampling theory, sample surveys (62D05) Bayesian inference (62F15) Monte Carlo methods (65C05)

Abstract: The effective sample size (ESS) is widely used in sample-based simulation methods for assessing the quality of a Monte Carlo approximation of a given distribution and of related integrals. In this paper, we revisit the approximation of the ESS in the specific context of importance sampling (IS). The derivation of this approximation, that we will denote as $widehatextESS$, is partially available in Kong (1992). This approximation has been widely used in the last 25 years due to its simplicity as a practical rule of thumb in a wide variety of importance sampling methods. However, we show that the multiple assumptions and approximations in the derivation of $widehatextESS$, makes it difficult to be considered even as a reasonable approximation of the ESS. We extend the discussion of the $widehatextESS$ in the multiple importance sampling (MIS) setting, we display numerical examples, and we discuss several avenues for developing alternative metrics. This paper does not cover the use of ESS for MCMC algorithms.

Recommendations

• The effective sample size
• Determining the Effective Sample Size of a Parametric Prior
• Consistent estimation of the accuracy of importance sampling using regenerative simulation
• Importance Sampling Via the Estimated Sampler
• Quasi-reliable estimates of effective sample size

Cites work

• Another look at rejection sampling through importance sampling
• Blind Deconvolution via Sequential Imputations
• Compressed Monte Carlo with application in particle filtering
• Generalized multiple importance sampling
• scientific article; zbMATH DE number 5919872 (Why is no real title available?)
• scientific article; zbMATH DE number 107482 (Why is no real title available?)
• scientific article; zbMATH DE number 1124118 (Why is no real title available?)
• scientific article; zbMATH DE number 2117879 (Why is no real title available?)
• Importance Gaussian Quadrature
• Importance sampling the union of rare events with an application to power systems analysis
• Importance sampling: intrinsic dimension and computational cost
• Introducing Monte Carlo Methods with R
• Introduction to rare event simulation.
• Layered adaptive importance sampling
• Methods of reducing sample size in Monte Carlo computations
• Multiple Importance Sampling for Symbol Error Rate Estimation of Maximum-Likelihood Detectors in MIMO Channels
• On the role of interaction in sequential Monte Carlo algorithms
• Sequential Imputations and Bayesian Missing Data Problems
• Sequential Monte Carlo Methods in Practice

Cited in

(13)

• scientific article; zbMATH DE number 7234286 (Why is no real title available?)
• scientific article; zbMATH DE number 1222374 (Why is no real title available?)
• The Myth of Continuity-Corrected Sample Size Formulae
• On the effective geographic sample size
• Annealed adaptive importance sampling method in PINNs for solving high dimensional partial differential equations
• An adaptive mixture view of particle filters
• Global convergence of optimized adaptive importance samplers
• An overview of differentiable particle filters for data-adaptive sequential Bayesian inference
• Effective sample size approximations as entropy measures
• Persistent sampling: enhancing the efficiency of sequential Monte Carlo
• Annealing strategies for variance reduction in balance heuristic estimators
• Doubly Adaptive Importance Sampling
• Mitigating mode collapse in normalizing flows by annealing with an adaptive schedule: application to parameter estimation

This page was built for publication: Rethinking the Effective Sample Size

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6067598)