Fast and Accurate Binary Response Mixed Model Analysis via Expectation  Propagation

Hall, P; Johnstone, IM; Ormerod, JT; Wand, MP; Yu, JCF

Fast and Accurate Binary Response Mixed Model Analysis via Expectation Propagation

Hall, P Johnstone, IM Ormerod, JT Wand, MP Yu, JCF

Permalink

Publisher:: Taylor and Francis
Publication Type:: Journal Article
Citation:: Journal of the American Statistical Association, 2020, 115, (532), pp. 1902-1916
Issue Date:: 2020

Closed Access

	Filename	Description	Size
	Hall20.pdf	Published version	2.65 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Hall, P
dc.contributor.author	Johnstone, IM
dc.contributor.author	Ormerod, JT
dc.contributor.author	Wand, MP
dc.contributor.author	Yu, JCF
dc.date.accessioned	2021-03-08T07:49:21Z
dc.date.available	2021-03-08T07:49:21Z
dc.date.issued	2020
dc.identifier.citation	Journal of the American Statistical Association, 2020, 115, (532), pp. 1902-1916
dc.identifier.issn	0162-1459
dc.identifier.issn	1537-274X
dc.identifier.uri	http://hdl.handle.net/10453/146926
dc.description.abstract	Expectation propagation is a general prescription for approximation of integrals in statistical inference problems. Its literature is mainly concerned with Bayesian inference scenarios. However, expectation propagation can also be used to approximate integrals arising in frequentist statistical inference. We focus on likelihood-based inference for binary response mixed models and show that fast and accurate quadrature-free inference can be realized for the probit link case with multivariate random effects and higher levels of nesting. The approach is supported by asymptotic theory in which expectation propagation is seen to provide consistent estimation of the exact likelihood surface. Numerical studies reveal the availability of fast, highly accurate and scalable methodology for binary mixed model analysis.
dc.language	English
dc.publisher	Taylor and Francis
dc.relation	http://purl.org/au-research/grants/arc/DP180100597
dc.relation.ispartof	Journal of the American Statistical Association
dc.relation.isbasedon	10.1080/01621459.2019.1665529
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0104 Statistics, 1403 Econometrics, 1603 Demography
dc.subject.classification	Statistics & Probability
dc.title	Fast and Accurate Binary Response Mixed Model Analysis via Expectation Propagation
dc.type	Journal Article
utslib.citation.volume	115
utslib.for	0104 Statistics
utslib.for	1403 Econometrics
utslib.for	1603 Demography
pubs.organisational-group	/University of Technology Sydney/Faculty of Science
pubs.organisational-group	/University of Technology Sydney/Faculty of Science/School of Mathematical and Physical Sciences
pubs.organisational-group	/University of Technology Sydney
utslib.copyright.status	closed_access	*
pubs.consider-herdc	true
dc.date.updated	2021-03-08T07:49:19Z
pubs.issue	532
pubs.publication-status	Published
pubs.volume	115
utslib.citation.issue	532

Abstract:

Expectation propagation is a general prescription for approximation of integrals in statistical inference problems. Its literature is mainly concerned with Bayesian inference scenarios. However, expectation propagation can also be used to approximate integrals arising in frequentist statistical inference. We focus on likelihood-based inference for binary response mixed models and show that fast and accurate quadrature-free inference can be realized for the probit link case with multivariate random effects and higher levels of nesting. The approach is supported by asymptotic theory in which expectation propagation is seen to provide consistent estimation of the exact likelihood surface. Numerical studies reveal the availability of fast, highly accurate and scalable methodology for binary mixed model analysis.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/146926