On nonlinear Markov chain Monte Carlo

DOI10.3150/10-BEJ307zbMATH Open1241.60037arXiv1107.3046MaRDI QIDQ638765FDOQ638765

Authors: Christophe Andrieu, Ajay Jasra, Arnaud Doucet, Pierre Del Moral 

Publication date: 14 September 2011

Published in: Bernoulli (Search for Journal in Brave)

Abstract: Let $mathscrP(E)$ be the space of probability measures on a measurable space $(E,mathcalE)$. In this paper we introduce a class of nonlinear Markov chain Monte Carlo (MCMC) methods for simulating from a probability measure $piinmathscrP(E)$. Nonlinear Markov kernels (see [Feynman--Kac Formulae: Genealogical and Interacting Particle Systems with Applications (2004) Springer]) $K:mathscrP(E)imesEightarrowmathscrP(E)$ can be constructed to, in some sense, improve over MCMC methods. However, such nonlinear kernels cannot be simulated exactly, so approximations of the nonlinear kernels are constructed using auxiliary or potentially self-interacting chains. Several nonlinear kernels are presented and it is demonstrated that, under some conditions, the associated approximations exhibit a strong law of large numbers; our proof technique is via the Poisson equation and Foster--Lyapunov conditions. We investigate the performance of our approximations with some simulations.

Full work available at URL: https://arxiv.org/abs/1107.3046

Recommendations

• Non-linear Markov Chain Monte Carlo
• Convergence of adaptive and interacting Markov chain Monte Carlo algorithms
• scientific article; zbMATH DE number 1560252
• Interacting Markov chain Monte Carlo methods for solving nonlinear measure-valued equations
• Collective proposal distributions for nonlinear MCMC samplers: mean-field theory and fast implementation

zbMATH Keywords

interacting Markov chainsPoisson equationFoster-Lyapunov conditionnonlinear Markov kernels

Mathematics Subject Classification ID

Computational methods in Markov chains (60J22) Monte Carlo methods (65C05) Numerical analysis or methods applied to Markov chains (65C40)

Cites Work

• Sequential Monte Carlo Methods in Practice
• Sequential Monte Carlo Samplers
• Title not available (Why is that?)
• Mixing: Properties and examples
• Markov Chains and Stochastic Stability
• Title not available (Why is that?)
• An adaptive Metropolis algorithm
• Title not available (Why is that?)
• Geometric ergodicity of Metropolis algorithms
• Geometric convergence and central limit theorems for multidimensional Hastings and Metropolis algorithms
• Equi-energy sampler with applications in statistical inference and statistical mechanics
• Coupling and Ergodicity of Adaptive Markov Chain Monte Carlo Algorithms
• On the ergodicity properties of some adaptive MCMC algorithms
• A Liapounov bound for solutions of the Poisson equation
• Computable bounds for geometric convergence rates of Markov chains
• Convergence rates for U-statistics and related statistics
• Polynomial ergodicity of Markov transition kernels.
• Non-linear Markov Chain Monte Carlo
• Maximal coupling
• Strong laws for 𝐿- and 𝑢-statistics
• Sequentially interacting Markov chain Monte Carlo methods
• A cautionary tale on the efficiency of some adaptive Monte Carlo schemes
• Convergence of adaptive and interacting Markov chain Monte Carlo algorithms
• A Note on Convergence of the Equi-Energy Sampler
• On convergence of chains with occupational self–interactions
• Two convergence properties of hybrid samplers
• Resampling from the past to improve on MCMC algorithms

Cited In (20)

• A comparative study of nonlinear Markov chain models for conditional simulation of multinomial classes from regular samples
• Discretizing nonlinear, non-Gaussian Markov processes with exact conditional moments
• Collective proposal distributions for nonlinear MCMC samplers: mean-field theory and fast implementation
• A central limit theorem for adaptive and interacting Markov chains
• Markov Kernels Local Aggregation for Noise Vanishing Distribution Sampling
• Empirical measure large deviations for reinforced chains on finite spaces
• Noisy Monte Carlo: convergence of Markov chains with approximate transition kernels
• Markov chain simulation for multilevel Monte Carlo
• Fluctuations of interacting Markov chain Monte Carlo methods
• On the stability of sequential Monte Carlo methods in high dimensions
• Non-reversible Metropolis-Hastings
• Convergence of adaptive and interacting Markov chain Monte Carlo algorithms
• Some remarks on MCMC estimation of spectra of integral operators
• Non-linear Markov Chain Monte Carlo
• Non-reversible guided Metropolis kernel
• Limit theorems for sequential MCMC methods
• Inference for a class of partially observed point process models
• Markov-chain Monte-Carlo methods and non-identifiabilities
• A new rate of convergence estimate for homogeneous discrete-time nonlinear Markov chains
• Markov chain Monte Carlo estimation of nonlinear dynamics from time series

This page was built for publication: On nonlinear Markov chain Monte Carlo

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q638765)