Statistics and Its Interface

Volume 8 (2015)

Number 3

Regression analysis of interval-censored failure time data with the additive hazards model in the presence of informative censoring

Pages: 367 – 377

DOI: http://dx.doi.org/10.4310/SII.2015.v8.n3.a10

Authors

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

Tao Hu (School of Mathematical Sciences & BCMIIS, Capital Normal University, Beijing, China)

Ling Ma (Department of Statistics, University of Missouri, Columbia, Missouri, U.S.A.)

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

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

Abstract

Regression analysis of interval-censored failure time data has recently attracted a great deal of attention and many procedures have been developed (Chen et al., 2012; Finkelstein, 1986; Sun, 2006; Sun and Li, 2013). However, most of the established procedures are for noninformative censoring, meaning that the censoring mechanism is independent of the underlying failure time of interest. In this paper, we discuss a more general situation where the censoring mechanism and the failure time of interest may be related and a sieve maximum likelihood estimation procedure is proposed for such data arising from the additive hazards model. In the method, the copula model is employed to model the correlation. The asymptotic properties of the resulting estimators are established and a simulation study is conducted, which indicates that the presented approach works well for the situations considered. An illustrative example is also provided.

Keywords

additive hazards model, copula models, informative censoring, regression analysis

Full Text (PDF format)