EUVIMEDEUVIMED entdecken
Crossref

A shared frailty regression model for clustered survival data

Gilbert Kiprotich, Diego Ignacio Gallardo iD, Pedro Luiz Ramos iD, Thomas Augustin

Statistical Methods in Medical Research · 2025

DOI: 10.1177/09622802251338984 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Statistical Methods in Medical Research, 2025. Verfasst von Gilbert Kiprotich und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistics, Ludwig Maximilian University Munich, Munich, Germany. Hauptthemen laut MEDLINE: Models, Statistical. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 77 Literaturangaben. Gefördert von: Deutscher Akademischer Austauschdienst.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In this article, we propose a new frailty model based on a mixture of inverse Gaussian distributions for multivariate lifetimes. This approach provides an advantage over previous models, as the weights are directly determined through parameterization of the mixture, removing the need for arbitrary guesswork in the weighting process. Moreover, the closed-form Laplace transform of the model facilitates the quantification of Kendall’s tau measure of dependence. The frailty model’s parametric and flexible parametric variants are examined. For parameter estimation, the expectation-maximization technique is employed, taking advantage of the hierarchical representation of the frailty distribution, providing a simpler and more stable method than directly maximizing the observed log-likelihood function. The performance of the estimators is assessed numerically using Monte Carlo simulations. We apply our methodology to two medical datasets on cancer. The results indicate the benefits of the proposed model over existing frailty models in the literature. The implementation of the procedure is added to the R package extrafrail .

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Models, StatisticalHumansMonte Carlo MethodSurvival AnalysisLikelihood FunctionsRegression AnalysisNeoplasmsFrailtyCluster AnalysisNormal Distribution

Stichwörter der Autor:innen

Clustered survival dataexpectation-maximization algorithmfinite mixturefrailtyinverse Gaussian distribution

Autor:innen und Institutionen

  • Department of Statistics, Ludwig Maximilian University Munich, Munich, Germany.
  • Departamento de Estadística, Facultad de Ciencias, Universidad del Bío-Bío, Concepción, Chile.
  • Facultad de Matemáticas, Pontificia Universidad Catòlica de Chile, Santiago, Chile.

ORCID-Profile: Diego Ignacio Gallardo, Pedro Luiz Ramos

Förderung

  • Deutscher Akademischer Austauschdienst

Zitieren

APA

Kiprotich, G., Gallardo, D. I., Ramos, P. L., & Augustin, T. (2025). A shared frailty regression model for clustered survival data. Statistical Methods in Medical Research. https://doi.org/10.1177/09622802251338984

Vancouver

Kiprotich G, Gallardo DI, Ramos PL, Augustin T. A shared frailty regression model for clustered survival data. Statistical Methods in Medical Research. 2025. doi:10.1177/09622802251338984

BibTeX
@article{Kiprotich2025,
  author = {Kiprotich, Gilbert and Gallardo, Diego Ignacio and Ramos, Pedro Luiz and Augustin, Thomas},
  title = {A shared frailty regression model for clustered survival data},
  journal = {Statistical Methods in Medical Research},
  year = {2025},
  doi = {10.1177/09622802251338984}
}
Kostenlos starten →