Multilevel rejection sampling for approximate Bayesian computation

DOI10.1016/J.CSDA.2018.02.009zbMATH Open1469.62160arXiv1702.03126OpenAlexW2616545773WikidataQ130185824 ScholiaQ130185824MaRDI QIDQ1662859FDOQ1662859

David J. Warne, Ruth E. Baker, Matthew J. Simpson

Publication date: 20 August 2018

Published in: Computational Statistics and Data Analysis (Search for Journal in Brave)

Abstract: Likelihood-free methods, such as approximate Bayesian computation, are powerful tools for practical inference problems with intractable likelihood functions. Markov chain Monte Carlo and sequential Monte Carlo variants of approximate Bayesian computation can be effective techniques for sampling posterior distributions in an approximate Bayesian computation setting. However, without careful consideration of convergence criteria and selection of proposal kernels, such methods can lead to very biased inference or computationally inefficient sampling. In contrast, rejection sampling for approximate Bayesian computation, despite being computationally intensive, results in independent, identically distributed samples from the approximated posterior. An alternative method is proposed for the acceleration of likelihood-free Bayesian inference that applies multilevel Monte Carlo variance reduction techniques directly to rejection sampling. The resulting method retains the accuracy advantages of rejection sampling while significantly improving the computational efficiency.

Full work available at URL: https://arxiv.org/abs/1702.03126

Recommendations

• Efficient acquisition rules for model-based approximate Bayesian computation
• Improved approximate Bayesian computation methods via empirical likelihood
• Approximate Bayesian computation with differential evolution
• Fast approximate Bayesian computation for estimating parameters in differential equations
• Approximate Bayesian computation: a nonparametric perspective

zbMATH Keywords

Bayesian inferenceapproximate Bayesian computationlikelihood-free methodsrejection samplingmultilevel Monte Carlo

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Bayesian inference (62F15) Monte Carlo methods (65C05)

Cites Work

• Sequential Monte Carlo Samplers
• Title not available (Why is that?)
• Constructing Summary Statistics for Approximate Bayesian Computation: Semi-Automatic Approximate Bayesian Computation
• Quantifying uncertainty in parameter estimates for stochastic models of collective cell spreading using approximate Bayesian computation
• Sequential Monte Carlo without likelihoods
• Monte Carlo sampling methods using Markov chains and their applications
• Approximate Bayesian computational methods
• Equation of State Calculations by Fast Computing Machines
• General state space Markov chains and MCMC algorithms
• Adaptive approximate Bayesian computation
• ABC likelihood-free methods for model choice in Gibbs random fields
• Approximate Bayesian computation by modelling summary statistics in a quasi-likelihood framework
• Estimation of Parameters for Macroparasite Population Evolution Using Approximate Bayesian Computation
• Multilevel Monte Carlo Path Simulation
• An adaptive sequential Monte Carlo method for approximate Bayesian computation
• Title not available (Why is that?)
• Multilevel Monte Carlo for continuous time Markov chains, with applications in biochemical kinetics
• On the asymptotic behavior of the stochastic and deterministic models of an epidemic
• Multilevel approximate Bayesian approaches for flows in highly heterogeneous porous media and their applications
• Extending the multi-level method for the simulation of stochastic biological systems
• On irregular functionals of SDEs and the Euler scheme
• Multilevel Markov Chain Monte Carlo Method for High-Contrast Single-Phase Flow Problems
• A Hierarchical Multilevel Markov Chain Monte Carlo Algorithm with Applications to Uncertainty Quantification in Subsurface Flow
• Multilevel sequential Monte Carlo samplers
• Multilevel Monte Carlo Approximation of Distribution Functions and Densities
• Inferring parameters for a lattice-free model of cell migration and proliferation using experimental data
• The rate of convergence for approximate Bayesian computation
• Multilevel ensemble transform particle filtering
• Bayesian computation: a summary of the current state, and samples backwards and forwards

Cited In (13)

• Multifidelity multilevel Monte Carlo to accelerate approximate Bayesian parameter inference for partially observed stochastic processes
• Multifidelity Approximate Bayesian Computation
• Nested sampling for general Bayesian computation
• Vector operations for accelerating expensive Bayesian computations - a tutorial guide
• Title not available (Why is that?)
• Rapid Bayesian Inference for Expensive Stochastic Models
• Efficient multifidelity likelihood-free Bayesian inference with adaptive computational resource allocation
• Using experimental data and information criteria to guide model selection for reaction-diffusion problems in mathematical biology
• Multifidelity Approximate Bayesian Computation with Sequential Monte Carlo Parameter Sampling
• Multilevel Markov Chain Monte Carlo
• A Bayesian sequential learning framework to parameterise continuum models of melanoma invasion into human skin
• Multilevel Monte Carlo with numerical smoothing for robust and efficient computation of probabilities and densities
• Multilevel Monte Carlo in approximate Bayesian computation

This page was built for publication: Multilevel rejection sampling for approximate Bayesian computation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1662859)