Improving SMC sampler estimate by recycling all past simulated particles

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

doi:10.1109/SSP.2014.6884589

Improving SMC sampler estimate by recycling all past simulated particles

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

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Citations (Scopus)

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.

Original language	English
Title of host publication	2014 IEEE Workshop on Statistical Signal Processing (SSP)
Publisher	IEEE
Pages	117-120
Number of pages	4
ISBN (Print)	9781479949755
DOIs	https://doi.org/10.1109/SSP.2014.6884589
Publication status	Published - 28 Aug 2014
Event	17th IEEE Workshop on Statistical Signal Processing 2014 - Gold Coast, Australia Duration: 29 Jun 2014 → 2 Jul 2014

Conference

Conference	17th IEEE Workshop on Statistical Signal Processing 2014
Abbreviated title	SSP 2014
Country/Territory	Australia
City	Gold Coast
Period	29/06/14 → 2/07/14

Keywords

Bayesian Inference
Recycling scheme
Sequential Monte Carlo sampler
Variance reduction

ASJC Scopus subject areas

Electrical and Electronic Engineering
Applied Mathematics
Signal Processing
Computer Science Applications

Access to Document

10.1109/SSP.2014.6884589

Cite this

@inproceedings{12253fee23fc4489923a6006fc1797eb,

title = "Improving SMC sampler estimate by recycling all past simulated particles",

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 = "Bayesian Inference, Recycling scheme, Sequential Monte Carlo sampler, Variance reduction",

author = "Nguyen, {Thi Le Thu} and Fran{\c c}ois Septier and Peters, {Gareth W.} and Yves Delignon",

year = "2014",

month = aug,

day = "28",

doi = "10.1109/SSP.2014.6884589",

language = "English",

isbn = "9781479949755",

pages = "117--120",

booktitle = "2014 IEEE Workshop on Statistical Signal Processing (SSP)",

publisher = "IEEE",

address = "United States",

note = "17th IEEE Workshop on Statistical Signal Processing 2014, SSP 2014 ; Conference date: 29-06-2014 Through 02-07-2014",

}

TY - GEN

T1 - Improving SMC sampler estimate by recycling all past simulated particles

AU - Nguyen, Thi Le Thu

AU - Septier, François

AU - Peters, Gareth W.

AU - Delignon, Yves

PY - 2014/8/28

Y1 - 2014/8/28

N2 - 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.

AB - 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.

KW - Bayesian Inference

KW - Recycling scheme

KW - Sequential Monte Carlo sampler

KW - Variance reduction

UR - http://www.scopus.com/inward/record.url?scp=84907392760&partnerID=8YFLogxK

U2 - 10.1109/SSP.2014.6884589

DO - 10.1109/SSP.2014.6884589

M3 - Conference contribution

AN - SCOPUS:84907392760

SN - 9781479949755

SP - 117

EP - 120

BT - 2014 IEEE Workshop on Statistical Signal Processing (SSP)

PB - IEEE

T2 - 17th IEEE Workshop on Statistical Signal Processing 2014

Y2 - 29 June 2014 through 2 July 2014

ER -

Improving SMC sampler estimate by recycling all past simulated particles

Abstract

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this