Home > Research > Publications & Outputs > Particle Metropolis-adjusted Langevin algorithms

Electronic data

  • Updated version of Particle Metropolis-adjusted Langevin algorithms

    Rights statement: This is a pre-copy-editing, author-produced PDF of an article accepted for publication in Biometrika following peer review. The definitive publisher-authenticated version Christopher Nemeth, Chris Sherlock, and Paul Fearnhead Particle Metropolis-adjusted Langevin algorithms Biometrika (2016) 103 (3): 701-717 first published online August 24, 2016 doi 10.1093/biomet/asw020 is available online at: http://biomet.oxfordjournals.org/content/103/3/701.abstract

    Accepted author manuscript, 586 KB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

Text available via DOI:

View graph of relations

Particle Metropolis-adjusted Langevin algorithms

Research output: Contribution to journalJournal articlepeer-review

Published

Standard

Particle Metropolis-adjusted Langevin algorithms. / Nemeth, Christopher; Sherlock, Christopher; Fearnhead, Paul.

In: Biometrika, Vol. 103, No. 3, 09.2016, p. 701-717.

Research output: Contribution to journalJournal articlepeer-review

Harvard

APA

Vancouver

Author

Bibtex

@article{9902f832eb2d4e5e8c2ecdc46a2ca0bb,
title = "Particle Metropolis-adjusted Langevin algorithms",
abstract = "This paper proposes a new sampling scheme based on Langevin dynamics that is applicable within pseudo-marginal and particle Markov chain Monte Carlo algorithms. We investigate this algorithm's theoretical properties under standard asymptotics, which correspond to an increasing dimension of the parameters, $n$. Our results show that the behaviour of the algorithm depends crucially on how accurately one can estimate the gradient of the log target density.If the error in the estimate of the gradient is not sufficiently controlled as dimension increases, then asymptotically there will be no advantage over the simpler random-walk algorithm. However, if the error is sufficiently well-behaved, then the optimal scaling of this algorithm will be $O(n^{-1/6})$ compared to $O(n^{-1/2})$ for the random walk. Our theory also gives guidelines on how to tune the number of Monte Carlo samples in the likelihood estimate and the proposal step-size.",
keywords = "stat.ME, stat.CO, stat.ML",
author = "Christopher Nemeth and Christopher Sherlock and Paul Fearnhead",
note = "This is a pre-copy-editing, author-produced PDF of an article accepted for publication in Biometrika following peer review. The definitive publisher-authenticated version Christopher Nemeth, Chris Sherlock, and Paul Fearnhead Particle Metropolis-adjusted Langevin algorithms Biometrika (2016) 103 (3): 701-717 first published online August 24, 2016 doi 10.1093/biomet/asw020 is available online at: http://biomet.oxfordjournals.org/content/103/3/701.abstract",
year = "2016",
month = sep,
doi = "10.1093/biomet/asw020",
language = "English",
volume = "103",
pages = "701--717",
journal = "Biometrika",
issn = "0006-3444",
publisher = "Oxford University Press",
number = "3",

}

RIS

TY - JOUR

T1 - Particle Metropolis-adjusted Langevin algorithms

AU - Nemeth, Christopher

AU - Sherlock, Christopher

AU - Fearnhead, Paul

N1 - This is a pre-copy-editing, author-produced PDF of an article accepted for publication in Biometrika following peer review. The definitive publisher-authenticated version Christopher Nemeth, Chris Sherlock, and Paul Fearnhead Particle Metropolis-adjusted Langevin algorithms Biometrika (2016) 103 (3): 701-717 first published online August 24, 2016 doi 10.1093/biomet/asw020 is available online at: http://biomet.oxfordjournals.org/content/103/3/701.abstract

PY - 2016/9

Y1 - 2016/9

N2 - This paper proposes a new sampling scheme based on Langevin dynamics that is applicable within pseudo-marginal and particle Markov chain Monte Carlo algorithms. We investigate this algorithm's theoretical properties under standard asymptotics, which correspond to an increasing dimension of the parameters, $n$. Our results show that the behaviour of the algorithm depends crucially on how accurately one can estimate the gradient of the log target density.If the error in the estimate of the gradient is not sufficiently controlled as dimension increases, then asymptotically there will be no advantage over the simpler random-walk algorithm. However, if the error is sufficiently well-behaved, then the optimal scaling of this algorithm will be $O(n^{-1/6})$ compared to $O(n^{-1/2})$ for the random walk. Our theory also gives guidelines on how to tune the number of Monte Carlo samples in the likelihood estimate and the proposal step-size.

AB - This paper proposes a new sampling scheme based on Langevin dynamics that is applicable within pseudo-marginal and particle Markov chain Monte Carlo algorithms. We investigate this algorithm's theoretical properties under standard asymptotics, which correspond to an increasing dimension of the parameters, $n$. Our results show that the behaviour of the algorithm depends crucially on how accurately one can estimate the gradient of the log target density.If the error in the estimate of the gradient is not sufficiently controlled as dimension increases, then asymptotically there will be no advantage over the simpler random-walk algorithm. However, if the error is sufficiently well-behaved, then the optimal scaling of this algorithm will be $O(n^{-1/6})$ compared to $O(n^{-1/2})$ for the random walk. Our theory also gives guidelines on how to tune the number of Monte Carlo samples in the likelihood estimate and the proposal step-size.

KW - stat.ME

KW - stat.CO

KW - stat.ML

U2 - 10.1093/biomet/asw020

DO - 10.1093/biomet/asw020

M3 - Journal article

VL - 103

SP - 701

EP - 717

JO - Biometrika

JF - Biometrika

SN - 0006-3444

IS - 3

ER -