A multiple-try Metropolis-Hastings algorithm with tailored proposals

DOI10.1007/S00180-019-00878-YzbMATH Open1505.62261arXiv1807.01914OpenAlexW2810256321WikidataQ128246492 ScholiaQ128246492MaRDI QIDQ2319482FDOQ2319482

Authors: Xin Luo, Håkon Tjelmeland 

Publication date: 19 August 2019

Published in: Computational Statistics (Search for Journal in Brave)

Abstract: We present a new multiple-try Metropolis-Hastings algorithm designed to be especially beneficial when a tailored proposal distribution is available. The algorithm is based on a given acyclic graph $G$, where one of the nodes in $G$, $k$ say, contains the current state of the Markov chain and the remaining nodes contain proposed states generated by applying the tailored proposal distribution. The Metropolis-Hastings algorithm alternates between two types of updates. The first update type is using the tailored proposal distribution to generate new states in all nodes in $G$ except in node $k$. The second update type is generating a new value for $k$, thereby changing the value of the current state. We evaluate the effectiveness of the proposed scheme in an example with previously defined target and proposal distributions.

Full work available at URL: https://arxiv.org/abs/1807.01914

Recommendations

• An efficient proposal distribution for Metropolis-Hastings using a \(B\)-splines technique
• On an adaptive version of the Metropolis–Hastings algorithm with independent proposal distribution
• On the flexibility of the design of multiple try Metropolis schemes
• Adaptively scaling the Metropolis algorithm using expected squared jumped distance
• On adaptive Metropolis-Hastings methods

zbMATH Keywords

Markov chain Monte Carloacyclic graphcustomized proposalsdependent proposalsGibbs updates

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Monte Carlo methods (65C05) Numerical analysis or methods applied to Markov chains (65C40)

Cites Work

• Reversible jump Markov chain Monte Carlo computation and Bayesian model determination
• Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
• Title not available (Why is that?)
• Markov chain Monte Carlo. Stochastic simulation for Bayesian inference.
• Monte Carlo sampling methods using Markov chains and their applications
• Title not available (Why is that?)
• Equation of state calculations by fast computing machines
• Mode jumping proposals in MCMC
• A multi-point Metropolis scheme with generic weight functions
• Multipoint Metropolis method with application to hybrid Monte Carlo
• A double Metropolis–Hastings sampler for spatial models with intractable normalizing constants
• Advanced Markov chain Monte Carlo methods. Learning from past samples.
• Title not available (Why is that?)
• On estimation in binary autologistic spatial models
• On the flexibility of the design of multiple try Metropolis schemes
• The Multiple-Try Method and Local Optimization in Metropolis Sampling
• Interacting multiple try algorithms with different proposal distributions
• Image analysis with partially ordered Markov models.
• On the flexibility of Metropolis-Hastings acceptance probabilities in auxiliary variable proposal generation
• Classification of binary random patterns
• Scaling analysis of multiple-try MCMC methods
• Facies modeling using a Markov mesh model specification
• Tailored randomized block MCMC methods with application to DSGE models
• Approximate computations for binary Markov random fields and their use in Bayesian models
• A generalized multiple-try version of the reversible jump algorithm
• Issues in the multiple try Metropolis mixing
• Prior specification for binary Markov mesh models

Cited In (8)

• Plateau proposal distributions for adaptive component-wise multiple-try metropolis
• Issues in the multiple try Metropolis mixing
• Generating MCMC proposals by randomly rotating the regular simplex
• A Quantum Parallel Markov Chain Monte Carlo
• A Repelling–Attracting Metropolis Algorithm for Multimodality
• Computing Bayes: from then `til now
• A multi-point Metropolis scheme with generic weight functions
• An efficient proposal distribution for Metropolis-Hastings using a \(B\)-splines technique

Uses Software

• GSLIB

This page was built for publication: A multiple-try Metropolis-Hastings algorithm with tailored proposals

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2319482)