Model Selection and Model Averaging for Mixed-Effects Models with Crossed Random Effects for Subjects and Items

Martínez-Huertas, José Ángel, Olmosa, Ricardo y Ferrer, Emilio . (2021) Model Selection and Model Averaging for Mixed-Effects Models with Crossed Random Effects for Subjects and Items. Multivariate Behavioral Research

Ficheros (Some files may be inaccessible until you login with your e-spacio credentials)
Nombre Descripción Tipo MIME Size
Martinez-Huertas_JA_MEMsCRmodel.pdf Martinez-Huertas_JA_MEMsCRmodel.pdf application/pdf 619.90KB

Título Model Selection and Model Averaging for Mixed-Effects Models with Crossed Random Effects for Subjects and Items
Autor(es) Martínez-Huertas, José Ángel
Olmosa, Ricardo
Ferrer, Emilio
Materia(s) Psicología
Abstract A good deal of experimental research is characterized by the presence of random effects on subjects and items. A standard modeling approach that includes such sources of variability is the mixed-effects models (MEMs) with crossed random effects. However, under-parameterizing or over-parameterizing the random structure of MEMs bias the estimations of the Standard Errors (SEs) of fixed effects. In this simulation study, we examined two different but complementary perspectives: model selection with likelihood-ratio tests, AIC, and BIC; and model averaging with Akaike weights. Results showed that true model selection was constant across the different strategies examined (including ML and REML estimators). However, sample size and variance of random slopes were found to explain true model selection and SE bias of fixed effects. No relevant differences in SE bias were found for model selection and model averaging. Sample size and variance of random slopes interacted with the estimator to explain SE bias. Only the within-subjects effect showed significant underestimation of SEs with smaller number of items and larger item random slopes. SE bias was higher for ML than REML, but the variability of SE bias was the opposite. Such variability can be translated into high rates of unacceptable bias in many replications.
Palabras clave mixed-effects models
crossed random effects
random slopes
model selection
model averaging
ML
REML
Editor(es) Taylor and Francis Group
Routledge
Fecha 2021-02-26
Formato application/pdf
Identificador bibliuned:DptoMCC-FPSI-Articulos-Jamartinez-0022
http://e-spacio.uned.es/fez/view/bibliuned:DptoMCC-FPSI-Articulos-Jamartinez-0022
DOI - identifier https://doi.org/10.1080/00273171.2021.1889946
ISSN - identifier 0027-3171; eISSN 1532-7906
Nombre de la revista Multivariate Behavioral Research
Número de Volumen 57
Número de Issue 4
Página inicial 603-619
Página final 603-619
Publicado en la Revista Multivariate Behavioral Research
Idioma eng
Versión de la publicación acceptedVersion
Tipo de recurso Article
Derechos de acceso y licencia http://creativecommons.org/licenses/by-nc-nd/4.0
info:eu-repo/semantics/openAccess
Tipo de acceso Acceso abierto
Notas adicionales La versión registrada de este artículo, publicado por primera vez en Multivariate Behavioral Research (2021) 57-4, p. 603-619, está disponible en línea en el sitio web del editor: Taylor and Francis Group / Routledge; https://doi.org/10.1080/00273171.2021.1889946
Notas adicionales The registered version of this article, first published in Multivariate Behavioral Research (2021) 57-4, p. 603-619, is available online at the publisher's website: Taylor and Francis Group/Routledge; https://doi.org/10.1080/00273171.2021.1889946

 
Versiones
Versión Tipo de filtro
Contador de citas: Google Scholar Search Google Scholar
Estadísticas de acceso: 31 Visitas, 13 Descargas  -  Estadísticas en detalle
Creado: Fri, 01 Mar 2024, 18:58:27 CET