Assuming independence in spatial latent variable models: consequences and implications of misspecification

Multivariate spatial data, where multiple responses are simultaneously recorded across spatially indexed observational units, are routinely collected in a wide variety of disciplines. For example, the Southern Ocean Continuous Plankton Recorder survey collects records of zooplankton communities in t...

Full description

Bibliographic Details
Published in:Biometrics
Main Authors: Hui, FKC, Hill, NA, Welsh, AH
Format: Article in Journal/Newspaper
Language:unknown
Published: Blackwell Publishing Ltd 2021
Subjects:
Online Access:https://eprints.utas.edu.au/42346/
id ftunivtasmania:oai:eprints.utas.edu.au:42346
record_format openpolar
spelling ftunivtasmania:oai:eprints.utas.edu.au:42346 2023-05-15T18:25:07+02:00 Assuming independence in spatial latent variable models: consequences and implications of misspecification Hui, FKC Hill, NA Welsh, AH 2021 https://eprints.utas.edu.au/42346/ unknown Blackwell Publishing Ltd Hui, FKC, Hill, NA orcid:0000-0001-9329-6717 and Welsh, AH 2021 , 'Assuming independence in spatial latent variable models: consequences and implications of misspecification' , Biometrics , pp. 1-15 , doi:10.1111/biom.13416 <http://dx.doi.org/10.1111/biom.13416>. joint species distribution models latent variables spatial zooplankton community ecology factor analysis loadings multivariate abundance data spatio-temporal Article PeerReviewed 2021 ftunivtasmania https://doi.org/10.1111/biom.13416 2022-02-07T23:18:41Z Multivariate spatial data, where multiple responses are simultaneously recorded across spatially indexed observational units, are routinely collected in a wide variety of disciplines. For example, the Southern Ocean Continuous Plankton Recorder survey collects records of zooplankton communities in the Indian sector of the Southern Ocean, with the aim of identifying and quantifying spatial patterns in biodiversity in response to environmental change. One increasingly popular method for modeling such data is spatial generalized linear latent variable models (GLLVMs), where the correlation across sites is captured by a spatial covariance function in the latent variables. However, little is known about the impact of misspecifying the latent variable correlation structure on inference of various parameters in such models. To address this gap in the literature, we investigate how misspecifying and assuming independence for the latent variables' correlation structure impacts estimation and inference in spatial GLLVMs. Through both theory and numerical studies, we show that performance of maximum likelihood estimation and inference on regression coefficients under misspecification depends on a combination of the response type, the magnitude of true regression coefficient, and the corresponding loadings, and, most importantly, whether the corresponding covariate is (also) spatially correlated. On the other hand, estimation and inference of truly nonzero loadings and prediction of latent variables is consistently not robust to misspecification of the latent variable correlation structure. Article in Journal/Newspaper Southern Ocean University of Tasmania: UTas ePrints Indian Southern Ocean Biometrics 78 1 85 99
institution Open Polar
collection University of Tasmania: UTas ePrints
op_collection_id ftunivtasmania
language unknown
topic joint species distribution models
latent variables
spatial
zooplankton
community ecology
factor analysis
loadings
multivariate abundance data
spatio-temporal
spellingShingle joint species distribution models
latent variables
spatial
zooplankton
community ecology
factor analysis
loadings
multivariate abundance data
spatio-temporal
Hui, FKC
Hill, NA
Welsh, AH
Assuming independence in spatial latent variable models: consequences and implications of misspecification
topic_facet joint species distribution models
latent variables
spatial
zooplankton
community ecology
factor analysis
loadings
multivariate abundance data
spatio-temporal
description Multivariate spatial data, where multiple responses are simultaneously recorded across spatially indexed observational units, are routinely collected in a wide variety of disciplines. For example, the Southern Ocean Continuous Plankton Recorder survey collects records of zooplankton communities in the Indian sector of the Southern Ocean, with the aim of identifying and quantifying spatial patterns in biodiversity in response to environmental change. One increasingly popular method for modeling such data is spatial generalized linear latent variable models (GLLVMs), where the correlation across sites is captured by a spatial covariance function in the latent variables. However, little is known about the impact of misspecifying the latent variable correlation structure on inference of various parameters in such models. To address this gap in the literature, we investigate how misspecifying and assuming independence for the latent variables' correlation structure impacts estimation and inference in spatial GLLVMs. Through both theory and numerical studies, we show that performance of maximum likelihood estimation and inference on regression coefficients under misspecification depends on a combination of the response type, the magnitude of true regression coefficient, and the corresponding loadings, and, most importantly, whether the corresponding covariate is (also) spatially correlated. On the other hand, estimation and inference of truly nonzero loadings and prediction of latent variables is consistently not robust to misspecification of the latent variable correlation structure.
format Article in Journal/Newspaper
author Hui, FKC
Hill, NA
Welsh, AH
author_facet Hui, FKC
Hill, NA
Welsh, AH
author_sort Hui, FKC
title Assuming independence in spatial latent variable models: consequences and implications of misspecification
title_short Assuming independence in spatial latent variable models: consequences and implications of misspecification
title_full Assuming independence in spatial latent variable models: consequences and implications of misspecification
title_fullStr Assuming independence in spatial latent variable models: consequences and implications of misspecification
title_full_unstemmed Assuming independence in spatial latent variable models: consequences and implications of misspecification
title_sort assuming independence in spatial latent variable models: consequences and implications of misspecification
publisher Blackwell Publishing Ltd
publishDate 2021
url https://eprints.utas.edu.au/42346/
geographic Indian
Southern Ocean
geographic_facet Indian
Southern Ocean
genre Southern Ocean
genre_facet Southern Ocean
op_relation Hui, FKC, Hill, NA orcid:0000-0001-9329-6717 and Welsh, AH 2021 , 'Assuming independence in spatial latent variable models: consequences and implications of misspecification' , Biometrics , pp. 1-15 , doi:10.1111/biom.13416 <http://dx.doi.org/10.1111/biom.13416>.
op_doi https://doi.org/10.1111/biom.13416
container_title Biometrics
container_volume 78
container_issue 1
container_start_page 85
op_container_end_page 99
_version_ 1766206338410479616