Addressing cluster-constant covariates in mixed effects models via likelihood-based boosting techniques.

Boosting techniques from the field of statistical learning have grown to be a popular tool for estimating and selecting predictor effects in various regression models and can roughly be separated in two general approaches, namely gradient boosting and likelihood-based boosting. An extensive framewor...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Colin Griesbach, Andreas Groll, Elisabeth Bergherr
Formato:	article
Lenguaje:	EN
Publicado:	Public Library of Science (PLoS) 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/4b52185acd62433d8f95dc84ac26e9e0
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:4b52185acd62433d8f95dc84ac26e9e0
record_format	dspace
spelling	oai:doaj.org-article:4b52185acd62433d8f95dc84ac26e9e02021-12-02T20:09:21ZAddressing cluster-constant covariates in mixed effects models via likelihood-based boosting techniques.1932-620310.1371/journal.pone.0254178https://doaj.org/article/4b52185acd62433d8f95dc84ac26e9e02021-01-01T00:00:00Zhttps://doi.org/10.1371/journal.pone.0254178https://doaj.org/toc/1932-6203Boosting techniques from the field of statistical learning have grown to be a popular tool for estimating and selecting predictor effects in various regression models and can roughly be separated in two general approaches, namely gradient boosting and likelihood-based boosting. An extensive framework has been proposed in order to fit generalized mixed models based on boosting, however for the case of cluster-constant covariates likelihood-based boosting approaches tend to mischoose variables in the selection step leading to wrong estimates. We propose an improved boosting algorithm for linear mixed models, where the random effects are properly weighted, disentangled from the fixed effects updating scheme and corrected for correlations with cluster-constant covariates in order to improve quality of estimates and in addition reduce the computational effort. The method outperforms current state-of-the-art approaches from boosting and maximum likelihood inference which is shown via simulations and various data examples.Colin GriesbachAndreas GrollElisabeth BergherrPublic Library of Science (PLoS)articleMedicineRScienceQENPLoS ONE, Vol 16, Iss 7, p e0254178 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Colin Griesbach Andreas Groll Elisabeth Bergherr Addressing cluster-constant covariates in mixed effects models via likelihood-based boosting techniques.
description	Boosting techniques from the field of statistical learning have grown to be a popular tool for estimating and selecting predictor effects in various regression models and can roughly be separated in two general approaches, namely gradient boosting and likelihood-based boosting. An extensive framework has been proposed in order to fit generalized mixed models based on boosting, however for the case of cluster-constant covariates likelihood-based boosting approaches tend to mischoose variables in the selection step leading to wrong estimates. We propose an improved boosting algorithm for linear mixed models, where the random effects are properly weighted, disentangled from the fixed effects updating scheme and corrected for correlations with cluster-constant covariates in order to improve quality of estimates and in addition reduce the computational effort. The method outperforms current state-of-the-art approaches from boosting and maximum likelihood inference which is shown via simulations and various data examples.
format	article
author	Colin Griesbach Andreas Groll Elisabeth Bergherr
author_facet	Colin Griesbach Andreas Groll Elisabeth Bergherr
author_sort	Colin Griesbach
title	Addressing cluster-constant covariates in mixed effects models via likelihood-based boosting techniques.
title_short	Addressing cluster-constant covariates in mixed effects models via likelihood-based boosting techniques.
title_full	Addressing cluster-constant covariates in mixed effects models via likelihood-based boosting techniques.
title_fullStr	Addressing cluster-constant covariates in mixed effects models via likelihood-based boosting techniques.
title_full_unstemmed	Addressing cluster-constant covariates in mixed effects models via likelihood-based boosting techniques.
title_sort	addressing cluster-constant covariates in mixed effects models via likelihood-based boosting techniques.
publisher	Public Library of Science (PLoS)
publishDate	2021
url	https://doaj.org/article/4b52185acd62433d8f95dc84ac26e9e0
work_keys_str_mv	AT colingriesbach addressingclusterconstantcovariatesinmixedeffectsmodelsvialikelihoodbasedboostingtechniques AT andreasgroll addressingclusterconstantcovariatesinmixedeffectsmodelsvialikelihoodbasedboostingtechniques AT elisabethbergherr addressingclusterconstantcovariatesinmixedeffectsmodelsvialikelihoodbasedboostingtechniques
_version_	1718375058929352704

Addressing cluster-constant covariates in mixed effects models via likelihood-based boosting techniques.

Ejemplares similares