Statistics and Its Interface

Volume 12 (2019)

Number 2

Estimation of the additive hazards model with current status data in the presence of informative censoring

Pages: 321 – 330



Huiqiong Li (Department of Statistics, Yunnan University, Kunming, China)

Han Zhang (Department of Statistics, University of Missouri, Columbia, Mo., U.S.A.)

Jianguo Sun (Department of Statistics, University of Missouri, Columbia, Mo., U.S.A.)


The additive hazards model is one of the most commonly used regression models in the analysis of failure time data and many methods have been developed for its inference under various situations. This paper discusses the situation where one faces current status data and also there exists informative censoring or when the failure time of interest and the observation process are correlated. Several authors have discussed the problem and in particular, Zhang et al. (2005) and Zhao et al. (2015) proposed an estimating equation-based approach and a copula model-based method, respectively. However, the former may not be efficient and the latter needs some restrictive assumptions. To address these, we propose a sieve maximum likelihood estimation approach that can be more efficient and also does not require the assumption above. For the implementation of the method, an EM algorithm is developed and the asymptotic properties of the resulting estimators are established. The numerical results suggest that the proposed method works well in practical situations and an application is provided.


current status data, EM algorithm, informative censoring

Received 13 March 2018

Published 11 March 2019