Efficient Monte Carlo sampling by parallel marginalization

Jonathan Weare

Research output: Contribution to journalArticle

Abstract

Markov chain Monte Carlo sampling methods often suffer from long correlation times. Consequently, these methods must be run for many steps to generate an independent sample. In this paper, a method is proposed to overcome this difficulty. The method utilizes information from rapidly equilibrating coarse Markov chains that sample marginal distributions of the full system. This is accomplished through exchanges between the full chain and the auxiliary coarse chains. Results of numerical tests on the bridge sampling and filtering/smoothing problems for a stochastic differential equation are presented.

Original languageEnglish (US)
Pages (from-to)12657-12662
Number of pages6
JournalProceedings of the National Academy of Sciences of the United States of America
Volume104
Issue number31
DOIs
StatePublished - Jul 31 2007

Fingerprint

Markov Chains
Monte Carlo Method

Keywords

  • Filtering
  • Markov chain Monte Carlo
  • Multi-grid
  • Parameter estimation
  • Renormalization

ASJC Scopus subject areas

  • General

Cite this

Efficient Monte Carlo sampling by parallel marginalization. / Weare, Jonathan.

In: Proceedings of the National Academy of Sciences of the United States of America, Vol. 104, No. 31, 31.07.2007, p. 12657-12662.

Research output: Contribution to journalArticle

@article{6b503ff2307944ecbff0e04efd5c8f60,
title = "Efficient Monte Carlo sampling by parallel marginalization",
abstract = "Markov chain Monte Carlo sampling methods often suffer from long correlation times. Consequently, these methods must be run for many steps to generate an independent sample. In this paper, a method is proposed to overcome this difficulty. The method utilizes information from rapidly equilibrating coarse Markov chains that sample marginal distributions of the full system. This is accomplished through exchanges between the full chain and the auxiliary coarse chains. Results of numerical tests on the bridge sampling and filtering/smoothing problems for a stochastic differential equation are presented.",
keywords = "Filtering, Markov chain Monte Carlo, Multi-grid, Parameter estimation, Renormalization",
author = "Jonathan Weare",
year = "2007",
month = "7",
day = "31",
doi = "10.1073/pnas.0705418104",
language = "English (US)",
volume = "104",
pages = "12657--12662",
journal = "Proceedings of the National Academy of Sciences of the United States of America",
issn = "0027-8424",
number = "31",

}

TY - JOUR

T1 - Efficient Monte Carlo sampling by parallel marginalization

AU - Weare, Jonathan

PY - 2007/7/31

Y1 - 2007/7/31

N2 - Markov chain Monte Carlo sampling methods often suffer from long correlation times. Consequently, these methods must be run for many steps to generate an independent sample. In this paper, a method is proposed to overcome this difficulty. The method utilizes information from rapidly equilibrating coarse Markov chains that sample marginal distributions of the full system. This is accomplished through exchanges between the full chain and the auxiliary coarse chains. Results of numerical tests on the bridge sampling and filtering/smoothing problems for a stochastic differential equation are presented.

AB - Markov chain Monte Carlo sampling methods often suffer from long correlation times. Consequently, these methods must be run for many steps to generate an independent sample. In this paper, a method is proposed to overcome this difficulty. The method utilizes information from rapidly equilibrating coarse Markov chains that sample marginal distributions of the full system. This is accomplished through exchanges between the full chain and the auxiliary coarse chains. Results of numerical tests on the bridge sampling and filtering/smoothing problems for a stochastic differential equation are presented.

KW - Filtering

KW - Markov chain Monte Carlo

KW - Multi-grid

KW - Parameter estimation

KW - Renormalization

UR - http://www.scopus.com/inward/record.url?scp=34547907797&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547907797&partnerID=8YFLogxK

U2 - 10.1073/pnas.0705418104

DO - 10.1073/pnas.0705418104

M3 - Article

C2 - 17640896

AN - SCOPUS:34547907797

VL - 104

SP - 12657

EP - 12662

JO - Proceedings of the National Academy of Sciences of the United States of America

JF - Proceedings of the National Academy of Sciences of the United States of America

SN - 0027-8424

IS - 31

ER -