Regression Analysis for Clustered Current Status Data Under Additive Transformation Models
摘要
This paper discusses regression analysis of clustered case I interval-censored data under a class of additive transformation risk models. We consider the situation that cluster size is informative about correlated failure times from the same cluster. A weighted estimating equation approach is constructed and the asymptotic properties are established. Finite-sample performance of the proposed method is assessed through a series of simulation studies. The results indicate that the proposed procedures work well and the weighted estimating equation method performs well in both the cases that cluster size is informative and noninformative. The proposed method is applied to analyze a real data set from a lung tumorigenicity study.