A frailty model approach for regression analysis of multivariate current status data |
| |
Authors: | Man‐Hua Chen Xingwei Tong Jianguo Sun |
| |
Affiliation: | 1. Department of Statistics, Tamkang University, Tamsui 25137, Taiwan;2. School of Mathematical Sciences, Beijing Normal University, Beijing 100875, People's Republic of China;3. Department of Statistics, University of Missouri, 146 Middlebush Hall, Columbia, MO 65211, U.S.A. |
| |
Abstract: | This paper discusses regression analysis of multivariate current status failure time data (The Statistical Analysis of Interval‐censoring Failure Time Data. Springer: New York, 2006), which occur quite often in, for example, tumorigenicity experiments and epidemiologic investigations of the natural history of a disease. For the problem, several marginal approaches have been proposed that model each failure time of interest individually (Biometrics 2000; 56 :940–943; Statist. Med. 2002; 21 :3715–3726). In this paper, we present a full likelihood approach based on the proportional hazards frailty model. For estimation, an Expectation Maximization (EM) algorithm is developed and simulation studies suggest that the presented approach performs well for practical situations. The approach is applied to a set of bivariate current status data arising from a tumorigenicity experiment. Copyright © 2009 John Wiley & Sons, Ltd. |
| |
Keywords: | EM algorithm frailty model interval censoring maximum likelihood estimate multivariate failure time data |
|
|