On the stability and ergodicity of adaptive scaling Metropolis algorithms

DOI10.1016/J.SPA.2011.08.006zbMATH Open1234.65017arXiv0903.4061OpenAlexW2144035709WikidataQ109998355 ScholiaQ109998355MaRDI QIDQ645599FDOQ645599

Authors: Matti Vihola 

Publication date: 10 November 2011

Published in: Stochastic Processes and their Applications (Search for Journal in Brave)

Abstract: The stability and ergodicity properties of two adaptive random walk Metropolis algorithms are considered. The both algorithms adjust the scaling of the proposal distribution continuously based on the observed acceptance probability. Unlike the previously proposed forms of the algorithms, the adapted scaling parameter is not constrained within a predefined compact interval. The first algorithm is based on scale adaptation only, while the second one incorporates also covariance adaptation. A strong law of large numbers is shown to hold assuming that the target density is smooth enough and has either compact support or super-exponentially decaying tails.

Full work available at URL: https://arxiv.org/abs/0903.4061

Recommendations

• On adaptive Markov chain Monte Carlo algorithms
• Adaptive optimal scaling of Metropolis-Hastings algorithms using the Robbins-Monro process
• On the ergodicity of the adaptive Metropolis algorithm on unbounded domains
• Weak convergence and optimal scaling of random walk Metropolis algorithms
• An adaptive version for the Metropolis adjusted Langevin algorithm with a truncated drift

zbMATH Keywords

ergodicityrandom walkconvergencestabilityMetropolis algorithmstochastic approximationlaw of large numbersMonte Carlo methodscalingadaptive Markov chain

Mathematics Subject Classification ID

Computational methods in Markov chains (60J22) Monte Carlo methods (65C05) Numerical analysis or methods applied to Markov chains (65C40) Sums of independent random variables; random walks (60G50) Continuous-time Markov processes on discrete state spaces (60J27)

Cites Work

• Weak convergence and optimal scaling of random walk Metropolis algorithms
• An adaptive Metropolis algorithm
• Title not available (Why is that?)
• Optimal scaling for various Metropolis-Hastings algorithms.
• Geometric ergodicity of Metropolis algorithms
• Title not available (Why is that?)
• MC's for MCMC'ists
• Stability of Stochastic Approximation under Verifiable Conditions
• Coupling and Ergodicity of Adaptive Markov Chain Monte Carlo Algorithms
• Limit theorems for some adaptive MCMC algorithms with subgeometric kernels
• On the ergodicity of the adaptive Metropolis algorithm on unbounded domains
• Can the adaptive Metropolis algorithm collapse without the covariance lower bound?
• On adaptive Markov chain Monte Carlo algorithms
• On the ergodicity properties of some adaptive MCMC algorithms
• General state space Markov chains and MCMC algorithms
• Title not available (Why is that?)
• Adaptive Markov Chain Monte Carlo through Regeneration
• Trace bounds on the solution of the algebraic matrix Riccati and Lyapunov equation
• Optimal scaling of the random walk Metropolis on elliptically symmetric unimodal targets
• Optimal acceptance rates for Metropolis algorithms: Moving beyond 0.234

Cited In (15)

• Adaptive optimal scaling of Metropolis-Hastings algorithms using the Robbins-Monro process
• On the stability of some controlled Markov chains and its applications to stochastic approximation with Markovian dynamic
• Adaptive Metropolis algorithm using variational Bayesian adaptive Kalman filter
• An adaptive multiple-try Metropolis algorithm
• Grapham: graphical models with adaptive random walk Metropolis algorithms
• Adaptive Gibbs samplers and related MCMC methods
• Robust adaptive Metropolis algorithm with coerced acceptance rate
• Conditional particle filters with diffuse initial distributions
• A framework for adaptive MCMC targeting multimodal distributions
• Accelerating adaptation in the adaptive Metropolis–Hastings random walk algorithm
• Markovian stochastic approximation with expanding projections
• On the ergodicity of the adaptive Metropolis algorithm on unbounded domains
• Can the adaptive Metropolis algorithm collapse without the covariance lower bound?
• On adaptive Markov chain Monte Carlo algorithms
• On the ergodicity properties of some adaptive MCMC algorithms

This page was built for publication: On the stability and ergodicity of adaptive scaling Metropolis algorithms

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q645599)