The rate of convergence for approximate Bayesian computation

DOI10.1214/15-EJS988MaRDI QIDQ2259529zbMATH OpenWikidataFDO

Authors S. Barber, Jochen Voss, Mark Webster

Publication date 4 March 2015

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1311.2038, https://projecteuclid.org/euclid.ejs/1423229751

zbMATH Keywords

approximate Bayesian computationconvergence of estimatorslikelihood-free inferenceMonte Carlo methodsrate of convergence

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Bayesian inference (62F15) Monte Carlo methods (65C05)

Abstract: Approximate Bayesian Computation (ABC) is a popular computational method for likelihood-free Bayesian inference. The term "likelihood-free" refers to problems where the likelihood is intractable to compute or estimate directly, but where it is possible to generate simulated data $X$ relatively easily given a candidate set of parameters $heta$ simulated from a prior distribution. Parameters which generate simulated data within some tolerance $delta$ of the observed data $x^{\ast }$ are regarded as plausible, and a collection of such $heta$ is used to estimate the posterior distribution $heta,|,X!=!x^{\ast }$. Suitable choice of $delta$ is vital for ABC methods to return good approximations to $heta$ in reasonable computational time. While ABC methods are widely used in practice, particularly in population genetics, study of the mathematical properties of ABC estimators is still in its infancy. We prove that ABC estimates converge to the exact solution under very weak assumptions and, under slightly stronger assumptions, quantify the rate of this convergence. Our results can be used to guide the choice of the tolerance parameter $delta$.

Recommendations

• Adaptive approximate Bayesian computation tolerance selection
• Asymptotic properties of approximate Bayesian computation
• New insights into approximate Bayesian computation
• A new approach to choose acceptance cutoff for approximate Bayesian computation
• Approximate Bayesian computational methods

Cites work

• An introduction to statistical computing. A simulation-based approach
• Approximate Bayesian computation: a nonparametric perspective
• Approximate Bayesian computational methods
• Estimation of parameters for macroparasite population evolution using approximate Bayesian computation
• Extreme Points of Moment Sets
• scientific article; zbMATH DE number 1022658 (Why is no real title available?)
• scientific article; zbMATH DE number 1158743 (Why is no real title available?)
• Inference for Stereological Extremes
• New insights into approximate Bayesian computation
• Sequential Monte Carlo without likelihoods
• The rate of convergence for approximate Bayesian computation

Cited in

(28)

• An approximate likelihood perspective on ABC methods
• Bayesian estimation of agent-based models
• Multilevel rejection sampling for approximate Bayesian computation
• A rare event approach to high-dimensional approximate Bayesian computation
• \(L^\infty\) metric criteria for convergence in Bayesian recursive inference systems
• Adaptive approximate Bayesian computation tolerance selection
• The rate of convergence for approximate Bayesian computation
• Local dimension reduction of summary statistics for likelihood-free inference
• Spectral density-based and measure-preserving ABC for partially observed diffusion processes. An illustration on Hamiltonian SDEs
• A tutorial introduction to Bayesian inference for stochastic epidemic models using approximate Bayesian computation
• Bayesian mode and maximum estimation and accelerated rates of contraction
• Mutation and selection in bacteria: modelling and calibration
• Diagnostic tools for approximate Bayesian computation using the coverage property
• Asymptotic properties of approximate Bayesian computation
• Multifidelity approximate Bayesian computation
• Information geometry for approximate Bayesian computation
• On the use of approximate Bayesian computation Markov chain Monte Carlo with inflated tolerance and post-correction
• A new approach to choose acceptance cutoff for approximate Bayesian computation
• Adaptive Gaussian Process Approximation for Bayesian Inference with Expensive Likelihood Functions
• Convergence of regression-adjusted approximate Bayesian computation
• Piecewise approximate Bayesian computation: fast inference for discretely observed Markov models using a factorised posterior distribution
• Bayesian computation: a summary of the current state, and samples backwards and forwards
• A simulated annealing approach to approximate Bayes computations
• On predictive inference for intractable models via approximate Bayesian computation
• Vector operations for accelerating expensive Bayesian computations - a tutorial guide
• Improving the Accuracy of Marginal Approximations in Likelihood-Free Inference via Localization
• Market-based insurance ratemaking: application to pet insurance
• Weak convergence of posteriors conditional on maximum pseudo-likelihood estimates and implications in ABC

Describes a project that uses

Uses Software

• R

This page was built for publication: The rate of convergence for approximate Bayesian computation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2259529)