On the auxiliary particle filter

zbMATH Open1176.62092MaRDI QIDQ3633243FDOQ3633243

Authors Randal Douc, Jimmy Olsson, Eric Moulines

Publication date 18 June 2009

Full work available at URL https://arxiv.org/abs/0709.3448

zbMATH Keywords

sequential Monte Carlostate space modelcentral limit theoremstratified samplingARCH modelsauxiliary particle filterGaussian modelstwo-stage samplingadjustment multiplier weight

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Inference from stochastic processes and prediction (62M20)

Abstract: In this article we study asymptotic properties of weighted samples produced by the auxiliary particle filter (APF) proposed by pitt and shephard (1999). Besides establishing a central limit theorem (CLT) for smoothed particle estimates, we also derive bounds on the Lp error and bias of the same for a finite particle sample size. By examining the recursive formula for the asymptotic variance of the CLT we identify first-stage importance weights for which the increase of asymptotic variance at a single iteration of the algorithm is minimal. In the light of these findings, we discuss and demonstrate on several examples how the APF algorithm can be improved.

Recommendations

• A note on auxiliary particle filters
• Optimal SIR algorithm vs. fully adapted auxiliary particle filter: a non asymptotic analysis
• On some properties of Markov chain Monte Carlo simulation methods based on the particle filter
• Variance estimation in the particle filter
• The auxiliary iterated extended Kalman particle filter

Cited in

(23)

• Nudging the particle filter
• Optimal potential functions for the interacting particle system method
• Four encounters with system identification
• Particle learning for Bayesian semi-parametric stochastic volatility model
• Direct, prediction- and smoothing-based Kalman and particle filter algorithms
• Auxiliary particle filtering with lookahead support for univariate state space models
• Twisted particle filters
• Optimality guarantees for particle belief approximation of POMDPs
• Long-term stability of sequential Monte Carlo methods under verifiable conditions
• A note on auxiliary particle filters
• Efficient inference for nonlinear state space models: an automatic sample size selection rule
• Particle-based likelihood inference in partially observed diffusion processes using generalised Poisson estimators
• An iterated extended Kalman auxiliary particle filter and analysis of algorithm performance
• Optimal SIR algorithm vs. fully adapted auxiliary particle filter: a non asymptotic analysis
• Sequential Monte Carlo smoothing for general state space hidden Markov models
• On some properties of Markov chain Monte Carlo simulation methods based on the particle filter
• Antithetic sampling for sequential Monte Carlo methods with application to state-space models
• The auxiliary iterated extended Kalman particle filter
• Adaptive sequential Monte Carlo by means of mixture of experts
• A survey of sequential Monte Carlo methods for economics and finance
• Negative association, ordering and convergence of resampling methods
• Ergodicity and accuracy of optimal particle filters for Bayesian data assimilation
• Particle filters and Bayesian inference in financial econometrics

This page was built for publication: On the auxiliary particle filter

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3633243)