Adaptive Metropolis-Hastings sampling using reversible dependent mixture proposals
From MaRDI portal
Publication:2631371
DOI10.1007/S11222-014-9509-6zbMATH Open1342.62038arXiv1305.2634OpenAlexW2121341147MaRDI QIDQ2631371FDOQ2631371
Authors: Minh-Ngoc Tran, Michael K. Pitt, Robert Kohn 
Publication date: 29 July 2016
Published in: Statistics and Computing (Search for Journal in Brave)
Abstract: This article develops a general-purpose adaptive sampler that approximates the target density by a mixture of multivariate t densities. The adaptive sampler is based on reversible proposal distributions each of which has the mixture of multivariate t densities as its invariant density. The reversible proposals consist of a combination of independent and correlated steps that allow the sampler to traverse the parameter space efficiently as well as allowing the sampler to keep moving and locally exploring the parameter space. We employ a two-chain approach, in which a trial chain is used to adapt the proposal densities used in the main chain. Convergence of the main chain and a strong law of large numbers are proved under reasonable conditions, and without imposing a Diminishing Adaptation condition. The mixtures of multivariate t densities are fitted by an efficient Variational Approximation algorithm in which the number of components is determined automatically. The performance of the sampler is evaluated using simulated and real examples. Our autocorrelated framework is quite general and can handle mixtures other than multivariate t.
Full work available at URL: https://arxiv.org/abs/1305.2634
Recommendations
Markov chain Monte Carlosimulated annealingergodic convergencevariational approximationMetropolis-within Gibbs samplingmultivariate \(t\) mixtures
Computational methods in Markov chains (60J22) Bayesian inference (62F15) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Monte Carlo methods (65C05) Strong limit theorems (60F15)
Cites Work
- Statistical Applications of the Multivariate Skew Normal Distribution
- Sequential Monte Carlo Samplers
- Strictly Proper Scoring Rules, Prediction, and Estimation
- Probability and random processes.
- An adaptive Metropolis algorithm
- Variational approximations in Bayesian model selection for finite mixture distributions
- Central limit theorem for sequential Monte Carlo methods and its application to Bayesian inference
- Inference in hidden Markov models.
- Monte Carlo strategies in scientific computing
- Coupling and Ergodicity of Adaptive Markov Chain Monte Carlo Algorithms
- On the ergodicity properties of some adaptive MCMC algorithms
- Adaptive proposal distribution for random walk Metropolis algorithm
- Estimation of a covariance matrix using the reference prior
- Title not available (Why is that?)
- A weakly informative default prior distribution for logistic and other regression models
- On some properties of Markov chain Monte Carlo simulation methods based on the particle filter
- Advanced Markov chain Monte Carlo methods. Learning from past samples.
- The Multiple-Try Method and Local Optimization in Metropolis Sampling
- Interacting multiple try algorithms with different proposal distributions
- A New Approach to the Limit Theory of Recurrent Markov Chains
- Bayesian inference for a covariance matrix
- Learn from thy neighbor: parallel-chain and regional adaptive MCMC
- Component-wise Markov chain Monte Carlo: uniform and geometric ergodicity under mixing and composition
- Tailored randomized block MCMC methods with application to DSGE models
- Likelihood-based scoring rules for comparing density forecasts in tails
- Variational approximation for heteroscedastic linear models and matching pursuit algorithms
- Extended constructions of stationary autoregressive processes
- A vine-copula based adaptive MCMC sampler for efficient inference of dynamical systems
- Adaptive independent Metropolis-Hastings
- Adaptively scaling the Metropolis algorithm using expected squared jumped distance
Cited In (15)
- Adaptive tempered reversible jump algorithm for Bayesian curve fitting
- Plateau proposal distributions for adaptive component-wise multiple-try metropolis
- Partially collapsed Gibbs sampling and path-adaptive Metropolis-Hastings in high-energy astrophysics
- On an adaptive version of the Metropolis–Hastings algorithm with independent proposal distribution
- Bayesian inference using adaptive sampling
- Striated Metropolis-Hastings sampler for high-dimensional models
- A modified conditional Metropolis-Hastings sampler
- An adaptive multiple-try Metropolis algorithm
- On adaptive Metropolis-Hastings methods
- Learn from thy neighbor: parallel-chain and regional adaptive MCMC
- Bayesian adaptively updated Hamiltonian Monte Carlo with an application to high-dimensional BEKK GARCH models
- Title not available (Why is that?)
- Markov chain Monte Carlo algorithms with sequential proposals
- Componentwise adaptation for high dimensional MCMC
- Adaptive Component-Wise Multiple-Try Metropolis Sampling
This page was built for publication: Adaptive Metropolis-Hastings sampling using reversible dependent mixture proposals
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2631371)
