search for




 

Fitting semi-parametric frailty models using recent frailty R packages
Journal of the Korean Data & Information Science Society 2018;29:583-91
Published online May 31, 2018
© 2018 Korean Data and Information Science Society.

Jihoon Kwon1 · Eunyoung Park2 · Il Do Ha3

123Department of Statistics, Pukyong National University
Correspondence to: Professor, Department of Statistics, Pukyong National University, Busan 48513, Korea. E-mail: idha1353@pknu.ac.kr
Received April 16, 2018; Revised May 2, 2018; Accepted May 8, 2018.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
Semi-parametric frailty models, extensions of Cox’s proportional hazards models, have been widely used for the analysis of multivariate (or clustered) survival data where are frequently encountered in biomedical research. In this paper, we compare the estimation results from various R packages (frailtyHL, frailtyEM, frailtySurv, survival, frailtypack) which have been recently developed for fitting the semi-parametric frailty models. For this purpose we present simulation results and and example-data analysis using a well-known kidney infection data. In particular, we use two popular frailty distributions with gamma and lognormal distributions. Following simulation results, we found out that the frailtyHL method was better than the frailtyEM method in terms of the bias of the estimator of variance parameter of frailty when the cluster size is small.
Keywords : H-likelihood, marginal likelihood, random effect, semi-parametric frailty models.