Particle-kernel estimation of the filter density in state-space models

DOI10.3150/13-BEJ545zbMATH Open1346.60112arXiv1111.5866OpenAlexW3122092251MaRDI QIDQ470055FDOQ470055

Authors: Dan Crisan, Joaquín Míguez 

Publication date: 11 November 2014

Published in: Bernoulli (Search for Journal in Brave)

Abstract: Sequential Monte Carlo (SMC) methods, also known as particle filters, are simulation-based recursive algorithms for the approximation of the a posteriori probability measures generated by state-space dynamical models. At any given time $t$, a SMC method produces a set of samples over the state space of the system of interest (often termed "particles") that is used to build a discrete and random approximation of the posterior probability distribution of the state variables, conditional on a sequence of available observations. One potential application of the methodology is the estimation of the densities associated to the sequence of a posteriori distributions. While practitioners have rather freely applied such density approximations in the past, the issue has received less attention from a theoretical perspective. In this paper, we address the problem of constructing kernel-based estimates of the posterior probability density function and its derivatives, and obtain asymptotic convergence results for the estimation errors. In particular, we find convergence rates for the approximation errors that hold uniformly on the state space and guarantee that the error vanishes almost surely as the number of particles in the filter grows. Based on this uniform convergence result, we first show how to build continuous measures that converge almost surely (with known rate) toward the posterior measure and then address a few applications. The latter include maximum a posteriori estimation of the system state using the approximate derivatives of the posterior density and the approximation of functionals of it, for example, Shannon's entropy. This manuscript is identical to the published paper, including a gap in the proof of Theorem 4.2. The Theorem itself is correct. We provide an {em erratum} at the end of this document with a complete proof and a brief discussion.

Full work available at URL: https://arxiv.org/abs/1111.5866

Recommendations

• Sequential Monte Carlo smoothing for general state space hidden Markov models
• Central limit theorem for sequential Monte Carlo methods and its application to Bayesian inference
• Sequential Monte Carlo with Highly Informative Observations
• Particle filters
• On convergence of kernel density estimates in particle filtering.

zbMATH Keywords

density estimationstate-space modelsalmost sure convergenceparticle filteringstochastic filteringsequential Monte Carlo methodMarkov systems

Mathematics Subject Classification ID

Computational methods in Markov chains (60J22) Density estimation (62G07) Markov processes: estimation; hidden Markov models (62M05) Monte Carlo methods (65C05) Numerical analysis or methods applied to Markov chains (65C40) Filtering in stochastic control theory (93E11) Signal detection and filtering (aspects of stochastic processes) (60G35) Markov chains (discrete-time Markov processes on discrete state spaces) (60J10) Strong limit theorems (60F15) Discrete-time Markov processes on general state spaces (60J05) Stochastic particle methods (65C35)

Cites Work

• Minimizing multimodal functions of continuous variables with the “simulated annealing” algorithm—Corrigenda for this article is available here
• Title not available (Why is that?)
• Title not available (Why is that?)
• Smoothing methods in statistics
• Title not available (Why is that?)
• Title not available (Why is that?)
• Stability and uniform approximation of nonlinear filters using the Hilbert metric and application to particle filters
• Title not available (Why is that?)
• Title not available (Why is that?)
• Sequential Monte Carlo Methods for Dynamic Systems
• Cross-validation Bandwidth Matrices for Multivariate Kernel Density Estimation
• A Basic Convergence Result for Particle Filtering
• A survey of convergence results on particle filtering methods for practitioners
• Recursive Monte Carlo filters: algorithms and theoretical analysis
• Title not available (Why is that?)
• Fundamentals of stochastic filtering
• On Accelerated Random Search
• Parameter estimation for hidden Markov models with intractable likelihoods
• Derivative-free filter simulated annealing method for constrained continuous global optimization
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Expectation maximization algorithms for MAP estimation of jump Markov linear systems
• A Bayesian approach to bandwidth selection for multivariate kernel density estimation
• Edgeworth Approximation of Multivariate Differential Entropy
• Maximum a posteriori sequence estimation using Monte Carlo particle filters
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• On the Estimation of Differential Entropy From Data Located on Embedded Manifolds
• Uniform approximations of discrete-time filters
• On the asymptotic properties of a simple estimate of the Mode
• On the convergence of two sequential Monte Carlo methods for maximum a posteriori sequence estimation and stochastic global optimization
• Bootstrapping nonparametric density estimators with empirically chosen bandwidths.
• Title not available (Why is that?)

Cited In (18)

• Parallel sequential Monte Carlo for stochastic gradient-free nonconvex optimization
• Convergence rates of kernel density estimates in particle filtering
• Efficient particle filtering for stochastic Korteweg-de Vries equations
• Convergence of the SMC implementation of the PHD filter
• A nonlinear population Monte Carlo scheme for the Bayesian estimation of parameters of \(\alpha\)-stable distributions
• Stochastic filtering methods in electronic trading
• Convergence rates for optimised adaptive importance samplers
• Inferences from optimal filtering equation
• Convergence of Regularized Particle Filters for Stochastic Reaction Networks
• A statistical approach to estimate state variables in flow-accelerated corrosion problems
• Continuous‐time threshold autoregressions with jumps: Properties, estimation, and application to electricity markets
• On the convergence of two sequential Monte Carlo methods for maximum a posteriori sequence estimation and stochastic global optimization
• Stochastic filtering for multiscale stochastic reaction networks based on hybrid approximations
• Sequential Monte Carlo with kernel embedded mappings: the mapping particle filter
• Bandwidth selection in pre-smoothed particle filters
• Parallelizing particle filters with butterfly interactions
• On convergence of kernel density estimates in particle filtering.
• Biased online parameter inference for state-space models

Uses Software

• KernSmooth
• simannf90
• pyuvdata

This page was built for publication: Particle-kernel estimation of the filter density in state-space models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q470055)