A lognormal central limit theorem for particle approximations of normalizing constants

DOI10.1214/EJP.V19-3428MaRDI QIDQ2514286zbMATH OpenFDO

Authors Jean Bérard, Pierre Del Moral, Arnaud Doucet

Publication date 3 February 2015

Published in Electronic Journal of Probability (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1307.0181

zbMATH Keywords

central limit theoremsFeynman-Kac modelsmean field interacting particle systemsnonlinear filteringparticle absorption modelsquasi-invariant measures

Mathematics Subject Classification ID

Central limit and other weak theorems (60F05) Applications of Brownian motions and diffusion theory (population genetics, absorption problems, etc.) (60J70) Monte Carlo methods (65C05) Stochastic particle methods (65C35) Interacting particle systems in time-dependent statistical mechanics (82C22)

Abstract: This paper deals with the numerical approximation of normalizing constants produced by particle methods, in the general framework of Feynman-Kac sequences of measures. It is well-known that the corresponding estimates satisfy a central limit theorem for a fixed time horizon $n$ as the number of particles $N$ goes to infinity. Here, we study the situation where both $n$ and $N$ go to infinity in such a way that $li{m}_{nightarrowinfty}\mathrm{\%}n/N=alpha>0$. In this context, Pitt et al. cite{pitt2012} recently conjectured that a lognormal central limit theorem should hold. We formally establish this result here, under general regularity assumptions on the model. We also discuss special classes of models (time-homogeneous environment and ergodic random environment) for which more explicit descriptions of the limiting bias and variance can be obtained.

Recommendations

• A central limit theorem for measurements on the logarithmic scale and its application to dimension estimates
• The law of the iterated logarithm and central limit theorem for L-statistics
• Central limit theorems for \(L_ p\)-norms of density estimators
• An almost sure central limit theorem for NA and LNQD random variables
• The central limit theorem and the law of iterated logarithm for empirical processes under local conditions
• On the mean normal measures of a particle process
• A central limit type theorem for a class of particle filters
• On the Norming Constants in the Feller-Khintchine-Lévy Central Limit Theorem

Cited in

(13)

• Accelerating pseudo-marginal MCMC using Gaussian processes
• Efficiency of delayed-acceptance random walk metropolis algorithms
• Comparison of Markov chains via weak Poincaré inequalities with application to pseudo-marginal MCMC
• Augmented pseudo-marginal Metropolis-Hastings for partially observed diffusion processes
• Unbiased Markov chain Monte Carlo for intractable target distributions
• On the efficiency of pseudo-marginal random walk Metropolis algorithms
• Correlated pseudo-marginal schemes for time-discretised stochastic kinetic models
• Adaptive particle allocation in iterated sequential Monte Carlo via approximating meta-models
• Optimal scaling for the pseudo-marginal random walk Metropolis: insensitivity to the noise generating mechanism
• A stable particle filter for a class of high-dimensional state-space models
• Accelerating inference for stochastic kinetic models
• On the stability of positive semigroups
• Scalable simulation-based inference for implicitly defined models using a metamodel for Monte Carlo log-likelihood estimator

This page was built for publication: A lognormal central limit theorem for particle approximations of normalizing constants

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2514286)