Home > Research > Publications & Outputs > Bayesian inference for diffusion-driven, mixed-...

Electronic data

  • 1507.06807v2

    Rights statement: c 2017 International Society for Bayesian Analysis

    Accepted author manuscript, 1.2 MB, PDF document

Links

Text available via DOI:

View graph of relations

Bayesian inference for diffusion-driven, mixed-effects models

Research output: Contribution to Journal/MagazineJournal articlepeer-review

E-pub ahead of print

Standard

Bayesian inference for diffusion-driven, mixed-effects models. / Whitaker, Gavin A.; Golightly, Andrew; Boys, Richard J et al.
In: Bayesian Analysis, Vol. 12, No. 2, 30.09.2016, p. 435-463.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Whitaker, GA, Golightly, A, Boys, RJ & Sherlock, CG 2016, 'Bayesian inference for diffusion-driven, mixed-effects models', Bayesian Analysis, vol. 12, no. 2, pp. 435-463. https://doi.org/10.1214/16-BA1009

APA

Whitaker, G. A., Golightly, A., Boys, R. J., & Sherlock, C. G. (2016). Bayesian inference for diffusion-driven, mixed-effects models. Bayesian Analysis, 12(2), 435-463. Advance online publication. https://doi.org/10.1214/16-BA1009

Vancouver

Whitaker GA, Golightly A, Boys RJ, Sherlock CG. Bayesian inference for diffusion-driven, mixed-effects models. Bayesian Analysis. 2016 Sept 30;12(2):435-463. Epub 2016 Sept 30. doi: 10.1214/16-BA1009

Author

Whitaker, Gavin A. ; Golightly, Andrew ; Boys, Richard J et al. / Bayesian inference for diffusion-driven, mixed-effects models. In: Bayesian Analysis. 2016 ; Vol. 12, No. 2. pp. 435-463.

Bibtex

@article{43c31d6f64284436ae7ac318fee791c1,
title = "Bayesian inference for diffusion-driven, mixed-effects models",
abstract = "Stochastic differential equations (SDEs) provide a natural framework for modelling intrinsic stochasticity inherent in many continuous-time physical processes. When such processes are observed in multiple individuals or experimental units, SDE driven mixed-effects models allow the quantification of between (as well as within) individual variation. Performing Bayesian inference for such models, using discrete time data that may be incomplete and subject to measurement error is a challenging problem and is the focus of this paper. We extend a recently proposed MCMC scheme to include the SDE driven mixed-effects framework. Fundamental to our approach is the development of a novel construct that allows for efficient sampling of conditioned SDEs that may exhibit nonlinear dynamics between observation times. We apply the resulting scheme to synthetic data generated from a simple SDE model of orange tree growth, and real data consisting of observations on aphid numbers recorded under a variety of different treatment regimes. In addition, we provide a systematic comparison of our approach with an inference scheme based on a tractable approximation of the SDE, that is, the linear noise approximation.",
keywords = "Stochastic differential equation, mixed-effects, Markov chain Monte Carlo, modified innovation scheme, linear noise approximation",
author = "Whitaker, {Gavin A.} and Andrew Golightly and Boys, {Richard J} and Sherlock, {Christopher Gerrard}",
note = "c 2017 International Society for Bayesian Analysis ",
year = "2016",
month = sep,
day = "30",
doi = "10.1214/16-BA1009",
language = "English",
volume = "12",
pages = "435--463",
journal = "Bayesian Analysis",
issn = "1936-0975",
publisher = "Carnegie Mellon University",
number = "2",

}

RIS

TY - JOUR

T1 - Bayesian inference for diffusion-driven, mixed-effects models

AU - Whitaker, Gavin A.

AU - Golightly, Andrew

AU - Boys, Richard J

AU - Sherlock, Christopher Gerrard

N1 - c 2017 International Society for Bayesian Analysis

PY - 2016/9/30

Y1 - 2016/9/30

N2 - Stochastic differential equations (SDEs) provide a natural framework for modelling intrinsic stochasticity inherent in many continuous-time physical processes. When such processes are observed in multiple individuals or experimental units, SDE driven mixed-effects models allow the quantification of between (as well as within) individual variation. Performing Bayesian inference for such models, using discrete time data that may be incomplete and subject to measurement error is a challenging problem and is the focus of this paper. We extend a recently proposed MCMC scheme to include the SDE driven mixed-effects framework. Fundamental to our approach is the development of a novel construct that allows for efficient sampling of conditioned SDEs that may exhibit nonlinear dynamics between observation times. We apply the resulting scheme to synthetic data generated from a simple SDE model of orange tree growth, and real data consisting of observations on aphid numbers recorded under a variety of different treatment regimes. In addition, we provide a systematic comparison of our approach with an inference scheme based on a tractable approximation of the SDE, that is, the linear noise approximation.

AB - Stochastic differential equations (SDEs) provide a natural framework for modelling intrinsic stochasticity inherent in many continuous-time physical processes. When such processes are observed in multiple individuals or experimental units, SDE driven mixed-effects models allow the quantification of between (as well as within) individual variation. Performing Bayesian inference for such models, using discrete time data that may be incomplete and subject to measurement error is a challenging problem and is the focus of this paper. We extend a recently proposed MCMC scheme to include the SDE driven mixed-effects framework. Fundamental to our approach is the development of a novel construct that allows for efficient sampling of conditioned SDEs that may exhibit nonlinear dynamics between observation times. We apply the resulting scheme to synthetic data generated from a simple SDE model of orange tree growth, and real data consisting of observations on aphid numbers recorded under a variety of different treatment regimes. In addition, we provide a systematic comparison of our approach with an inference scheme based on a tractable approximation of the SDE, that is, the linear noise approximation.

KW - Stochastic differential equation

KW - mixed-effects

KW - Markov chain Monte Carlo

KW - modified innovation scheme

KW - linear noise approximation

U2 - 10.1214/16-BA1009

DO - 10.1214/16-BA1009

M3 - Journal article

VL - 12

SP - 435

EP - 463

JO - Bayesian Analysis

JF - Bayesian Analysis

SN - 1936-0975

IS - 2

ER -