EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Non-boundary covariance matrix estimation in generalized linear mixed effects models using data augmentation priors

Tina Košuta iD, Erik Langerholc, Rok Blagus iD

Biometrics · 2026

DOI: 10.1093/biomtc/ujag013 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biometrics, 2026. Verfasst von Tina Košuta und 2 weiteren Autor:innen. Beteiligte Einrichtung: Institute for Biostatistics and Medical Informatics, University of Ljubljana, 1000 Ljubljana, Slovenia. Hauptthemen laut MEDLINE: Likelihood Functions, Linear Models. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Slovene National Research and Innovation Agency.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Boundary estimates of random effects covariance matrices commonly arise when using maximum likelihood (ML) estimation in generalized linear mixed effects models, leading to numerical challenges and affecting statistical inference. To mitigate this, we introduce penalties to the likelihood function derived from conditionally conjugate priors for the covariance or precision matrices of the random effects. Our choice of penalties (priors) allows representation through pseudo-observations, enabling implementation of the proposed penalized estimator within the existing ML software by augmenting the original data. We derive a procedure for constructing these pseudo-observations, a non-trivial task because their likelihood contribution must match the functional form of the penalty and depend only on the covariance or precision matrix of the random effects. Our method includes penalty parameters that can be set using existing prior knowledge or, when no reliable prior information is available, via a novel fully data-driven procedure that eliminates the need for prior specification. Through simulation studies under realistic scenarios, we illustrate that the proposed approach can provide improved estimates of random-effects covariance matrices compared with competing methods in the settings considered. The approach is further illustrated on real-world data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Likelihood FunctionsLinear ModelsAlgorithmsBiometryComputer SimulationData Interpretation, StatisticalModels, Statistical

Stichwörter der Autor:innen

(inverse) Wishart priordata-driven priorsmaximum a posteriori estimatepenalized maximum likelihoodpseudo-observations

Autor:innen und Institutionen

  • Institute for Biostatistics and Medical Informatics, University of Ljubljana, 1000 Ljubljana, Slovenia.
  • Faculty of Mathematics, Natural Sciences and Information Technologies, University of Primorska, 6000 Koper, Slovenia.

ORCID-Profile: Tina Košuta, Rok Blagus

Förderung

  • Slovene National Research and Innovation Agency

Zitieren

APA

Košuta, T., Langerholc, E., & Blagus, R. (2026). Non-boundary covariance matrix estimation in generalized linear mixed effects models using data augmentation priors. Biometrics. https://doi.org/10.1093/biomtc/ujag013

Vancouver

Košuta T, Langerholc E, Blagus R. Non-boundary covariance matrix estimation in generalized linear mixed effects models using data augmentation priors. Biometrics. 2026. doi:10.1093/biomtc/ujag013

BibTeX
@article{Kouta2026,
  author = {Košuta, Tina and Langerholc, Erik and Blagus, Rok},
  title = {Non-boundary covariance matrix estimation in generalized linear mixed effects models using data augmentation priors},
  journal = {Biometrics},
  year = {2026},
  doi = {10.1093/biomtc/ujag013}
}
Kostenlos starten →