Author Country (or Countries)



This paper addresses regression analysis of partly interval censored data. Partly interval censored failure time data consist of both exact observed and interval censored observations on the survival time of interest. Furthermore, there may exist a cured subgroup, indicating that a proportion of study subjects are not susceptible to the failure event of interest. For the problem, we assume a logistic model for the cure probability and that the failure times of the uncured group come from a wide class of transformation models, which includes proportional hazards and proportional odds models as special cases. For the determination of the proposed estimators, an EM algorithm based on some subject-specific independent Poisson variables is developed to calculate the maximum likelihood estimators. Extensive simulation studies are conducted and indicate that the proposed method works well for practical situations. A motivating application from NASA’s Hypobaric decompression sickness experiment is also provided.

Suggested Reviewers


Digital Object Identifier (DOI)