Home > Research > Publications & Outputs > Consistent and fast inference in compartmental ...

Research

Associated organisational unit

Text available via DOI:

Keywords

View graph of relations

Consistent and fast inference in compartmental models of epidemics using Poisson Approximate Likelihoods

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Consistent and fast inference in compartmental models of epidemics using Poisson Approximate Likelihoods. / Whitehouse, Michael; Whiteley, Nick; Rimella, Lorenzo.
In: Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 85, No. 4, 29.09.2023, p. 1173-1203.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Whitehouse, M, Whiteley, N & Rimella, L 2023, 'Consistent and fast inference in compartmental models of epidemics using Poisson Approximate Likelihoods', Journal of the Royal Statistical Society: Series B (Statistical Methodology), vol. 85, no. 4, pp. 1173-1203. https://doi.org/10.1093/jrsssb/qkad065

APA

Whitehouse, M., Whiteley, N., & Rimella, L. (2023). Consistent and fast inference in compartmental models of epidemics using Poisson Approximate Likelihoods. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 85(4), 1173-1203. https://doi.org/10.1093/jrsssb/qkad065

Vancouver

Whitehouse M, Whiteley N, Rimella L. Consistent and fast inference in compartmental models of epidemics using Poisson Approximate Likelihoods. Journal of the Royal Statistical Society: Series B (Statistical Methodology). 2023 Sept 29;85(4):1173-1203. Epub 2023 Jul 3. doi: 10.1093/jrsssb/qkad065

Author

Whitehouse, Michael ; Whiteley, Nick ; Rimella, Lorenzo. / Consistent and fast inference in compartmental models of epidemics using Poisson Approximate Likelihoods. In: Journal of the Royal Statistical Society: Series B (Statistical Methodology). 2023 ; Vol. 85, No. 4. pp. 1173-1203.

Bibtex

@article{db1efa35dd9a4a998f1f18442035d5c1,
title = "Consistent and fast inference in compartmental models of epidemics using Poisson Approximate Likelihoods",
abstract = "Addressing the challenge of scaling-up epidemiological inference to complex and heterogeneous models, we introduce Poisson approximate likelihood (PAL) methods. In contrast to the popular ordinary differential equation (ODE) approach to compartmental modelling, in which a large population limit is used to motivate a deterministic model, PALs are derived from approximate filtering equations for finite-population, stochastic compartmental models, and the large population limit drives consistency of maximum PAL estimators. Our theoretical results appear to be the first likelihood-based parameter estimation consistency results which apply to a broad class of partially observed stochastic compartmental models and address the large population limit. PALs are simple to implement, involving only elementary arithmetic operations and no tuning parameters, and fast to evaluate, requiring no simulation from the model and having computational cost independent of population size. Through examples we demonstrate how PALs can be used to: fit an age-structured model of influenza, taking advantage of automatic differentiation in Stan; compare over-dispersion mechanisms in a model of rotavirus by embedding PALs within sequential Monte Carlo; and evaluate the role of unit-specific parameters in a meta-population model of measles.",
keywords = "Statistics, Probability and Uncertainty, Statistics and Probability",
author = "Michael Whitehouse and Nick Whiteley and Lorenzo Rimella",
year = "2023",
month = sep,
day = "29",
doi = "10.1093/jrsssb/qkad065",
language = "English",
volume = "85",
pages = "1173--1203",
journal = "Journal of the Royal Statistical Society: Series B (Statistical Methodology)",
issn = "1369-7412",
publisher = "Wiley-Blackwell",
number = "4",

}

RIS

TY - JOUR

T1 - Consistent and fast inference in compartmental models of epidemics using Poisson Approximate Likelihoods

AU - Whitehouse, Michael

AU - Whiteley, Nick

AU - Rimella, Lorenzo

PY - 2023/9/29

Y1 - 2023/9/29

N2 - Addressing the challenge of scaling-up epidemiological inference to complex and heterogeneous models, we introduce Poisson approximate likelihood (PAL) methods. In contrast to the popular ordinary differential equation (ODE) approach to compartmental modelling, in which a large population limit is used to motivate a deterministic model, PALs are derived from approximate filtering equations for finite-population, stochastic compartmental models, and the large population limit drives consistency of maximum PAL estimators. Our theoretical results appear to be the first likelihood-based parameter estimation consistency results which apply to a broad class of partially observed stochastic compartmental models and address the large population limit. PALs are simple to implement, involving only elementary arithmetic operations and no tuning parameters, and fast to evaluate, requiring no simulation from the model and having computational cost independent of population size. Through examples we demonstrate how PALs can be used to: fit an age-structured model of influenza, taking advantage of automatic differentiation in Stan; compare over-dispersion mechanisms in a model of rotavirus by embedding PALs within sequential Monte Carlo; and evaluate the role of unit-specific parameters in a meta-population model of measles.

AB - Addressing the challenge of scaling-up epidemiological inference to complex and heterogeneous models, we introduce Poisson approximate likelihood (PAL) methods. In contrast to the popular ordinary differential equation (ODE) approach to compartmental modelling, in which a large population limit is used to motivate a deterministic model, PALs are derived from approximate filtering equations for finite-population, stochastic compartmental models, and the large population limit drives consistency of maximum PAL estimators. Our theoretical results appear to be the first likelihood-based parameter estimation consistency results which apply to a broad class of partially observed stochastic compartmental models and address the large population limit. PALs are simple to implement, involving only elementary arithmetic operations and no tuning parameters, and fast to evaluate, requiring no simulation from the model and having computational cost independent of population size. Through examples we demonstrate how PALs can be used to: fit an age-structured model of influenza, taking advantage of automatic differentiation in Stan; compare over-dispersion mechanisms in a model of rotavirus by embedding PALs within sequential Monte Carlo; and evaluate the role of unit-specific parameters in a meta-population model of measles.

KW - Statistics, Probability and Uncertainty

KW - Statistics and Probability

U2 - 10.1093/jrsssb/qkad065

DO - 10.1093/jrsssb/qkad065

M3 - Journal article

VL - 85

SP - 1173

EP - 1203

JO - Journal of the Royal Statistical Society: Series B (Statistical Methodology)

JF - Journal of the Royal Statistical Society: Series B (Statistical Methodology)

SN - 1369-7412

IS - 4

ER -