On the auxiliary particle filter

zbMATH Open1176.62092arXiv0709.3448MaRDI QIDQ3633243FDOQ3633243

Jimmy Olsson, Eric Moulines, Randal Douc

Publication date: 18 June 2009

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.

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

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

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)

Cited In (20)

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

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)