Improving SMC sampler estimate by recycling all past simulated particles

Author

Thi Le Thu Nguyen ; Septier, Francois ; Peters, Gareth W. ; Delignon, Yves

Author_Institution

LAGIS, Inst. Mines-Telecom, Lille, France

fYear

2014

fDate

June 29 2014-July 2 2014

Firstpage

117

Lastpage

120

Abstract

Sequential Monte Carlo (SMC) methods are not only a popular tool in the analysis of state-space models, but offer a powerful alternative to Markov chain Monte Carlo (MCMC) in situations where static Bayesian inference must be performed via simulation. In this paper, we propose a recycling scheme of all past simulated particles in the SMC sampler in order to reduce the variance of the final estimator. We demonstrate how the proposed approach outperforms the classical strategy in two challenging models.

Keywords

Bayes methods; Markov processes; Monte Carlo methods; inference mechanisms; signal sampling; state-space methods; Markov chain Monte Carlo; SMC sampler estimate improvement; all past simulated particle recycling scheme; sequential Monte Carlo sampler; signal processing; state-space model analysis; static Bayesian inference; variance reduction; Bayes methods; Conferences; Kernel; Monte Carlo methods; Recycling; Signal processing; Signal processing algorithms; Bayesian Inference; Recycling scheme; Sequential Monte Carlo sampler; Variance reduction;

fLanguage

English

Publisher

ieee

Conference_Titel

Statistical Signal Processing (SSP), 2014 IEEE Workshop on

Conference_Location

Gold Coast, VIC

Type

conf

DOI

10.1109/SSP.2014.6884589

Filename

6884589