Statistics and Its Interface

Volume 11 (2018)

Number 3

A new copula model-based method for regression analysis of dependent current status data

Pages: 463 – 471

DOI: http://dx.doi.org/10.4310/SII.2018.v11.n3.a9

Authors

Qi Cui (School of Mathematics, Jilin University, Changchun, China)

Hui Zhao (School of Mathematics and Statistics & Hubei Key Laboratory of Mathematical Sciences, Central China Normal Universitym Wuhan, China)

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

Abstract

This paper discusses regression analysis of current status data, which arise when the occurrence of the failure event of interest is observed only once or the occurrence time is either left- or right-censored [5, 11]. Many authors have investigated the problem, however, most of the existing methods are parametric or apply only to limited situations such that the failure time and the observation time have to be independent. In particular, Ma et al. [7] recently proposed a copula-based procedure for the situation where the failure time and the observation time are allowed to be dependent but their association needs to be known. To address this restriction, we present a new two-step estimation procedure that allows one to estimate the association parameter in addition to estimation of other unknown parameters. The asymptotic properties of the resulting estimators are established and a simulation study is conducted and suggests that the proposed method performs well for practical situations. Also an illustrative example is provided.

Keywords

copula model, current status data, informative censoring, proportional hazards model

Full Text (PDF format)

This work was partly supported by the National Nature Science Foundation of China Grant Nos. 11471135, 11571133, 11731011, 11671168, and the self-determined research funds of CCNU from the college’s basic research of MOE (CCNU15ZD011, CCNU16JCZX11).

Received 2 July 2017