Statistics and Its Interface

Volume 8 (2015)

Number 3

Efficient estimation for the additive hazards model in the presence of left-truncation and interval censoring

Pages: 391 – 402



Peijie Wang (Institute of Mathematics, Jilin University, Changchun, China)

Xingwei Tong (School of Mathematical Sciences, Beijing Normal University, Beijing, China)

Shishun Zhao (Institute of Mathematics, Jilin University, Changchun, China)

Jianguo Sun (Department of Statistics, University of Missouri, Columbia, Missouri, U.S.A.; and Institute of Mathematics, Jilin University, Changchun, China)


The additive hazards model is one of the most commonly used regression models in failure time data analysis and many authors have discussed its inference under various situations (Lin and Ying, 1994; Lin et al., 1998; Zeng et al., 2006; Wang et al., 2010). In this paper, we consider it when one faces left-truncated and interval-censored data, which often occur in, for example, epidemiological and medical follow-up studies. For inference, an efficient sieve maximum likelihood estimation procedure is developed and assessed by simulation studies, which indicate that the proposed method works well in practical situations. An illustrative example is also provided.


additive hazards model, interval-censoring, left-truncation, sieve maximum likelihood estimation

Full Text (PDF format)