EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Bayesian Model Selection for Generalized Linear Mixed Models

Shuangshuang Xu, Marco A R Ferreira iD, Erica M Porter, Christopher T Franck

Biometrics · 2023

DOI: 10.1111/biom.13896 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biometrics, 2023. Verfasst von Shuangshuang Xu und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistics, Virginia Tech, Blacksburg, Virginia, USA. Hauptthemen laut MEDLINE: Genome-Wide Association Study, Models, Statistical. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We propose a Bayesian model selection approach for generalized linear mixed models (GLMMs). We consider covariance structures for the random effects that are widely used in areas such as longitudinal studies, genome-wide association studies, and spatial statistics. Since the random effects cannot be integrated out of GLMMs analytically, we approximate the integrated likelihood function using a pseudo-likelihood approach. Our Bayesian approach assumes a flat prior for the fixed effects and includes both approximate reference prior and half-Cauchy prior choices for the variances of random effects. Since the flat prior on the fixed effects is improper, we develop a fractional Bayes factor approach to obtain posterior probabilities of the several competing models. Simulation studies with Poisson GLMMs with spatial random effects and overdispersion random effects show that our approach performs favorably when compared to widely used competing Bayesian methods including deviance information criterion and Watanabe–Akaike information criterion. We illustrate the usefulness and flexibility of our approach with three case studies including a Poisson longitudinal model, a Poisson spatial model, and a logistic mixed model. Our proposed approach is implemented in the R package GLMMselect that is available on CRAN.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Genome-Wide Association StudyModels, StatisticalBayes TheoremLikelihood FunctionsLinear ModelsComputer Simulation

Stichwörter der Autor:innen

approximate reference priorfractional Bayes factorgeneralized linear mixed modelmodel selectionpseudo-likelihood method

Autor:innen und Institutionen

  • Department of Statistics, Virginia Tech, Blacksburg, Virginia, USA.

ORCID-Profile: Marco A R Ferreira

Förderung

  • National Science Foundation

Zitieren

APA

Xu, S., Ferreira, M. A. R., Porter, E. M., & Franck, C. T. (2023). Bayesian Model Selection for Generalized Linear Mixed Models. Biometrics. https://doi.org/10.1111/biom.13896

Vancouver

Xu S, Ferreira MAR, Porter EM, Franck CT. Bayesian Model Selection for Generalized Linear Mixed Models. Biometrics. 2023. doi:10.1111/biom.13896

BibTeX
@article{Xu2023,
  author = {Xu, Shuangshuang and Ferreira, Marco A R and Porter, Erica M and Franck, Christopher T},
  title = {Bayesian Model Selection for Generalized Linear Mixed Models},
  journal = {Biometrics},
  year = {2023},
  doi = {10.1111/biom.13896}
}
Kostenlos starten →