Lan Kong
Doctoral student
Department of Biostatistics
Analysis of survival data from case-cohort studies
with semiparametric transformation models
Prentice (1986) proposed a case-cohort design to reduce the cost and effort of
large cohort studies of rare diseases. Under this design, the covariate information
is ascertained for only a random sample called subcohort from the entire study
population and additional cases outside the subcohort. The Cox model has been
the most popular model for survival data analysis. There are several statistical
methods available for fitting the Cox model to the case-cohort data. However,
the proportional hazard assumption in the Cox model may not always be true in
some applications or one may be interested in modeling the association from
different aspects. In this research, we consider a so-called semiparametric
transformation model for the case-cohort data. The fruitful class of semiparametric
transformation models includes the Cox model and a variety of non-proportional
hazard models. The estimating methods for regression parameters and survival
function under such models are proposed for case-cohort studies. We theoretically
derive the asymptotic properties of the proposed estimators. The finite sample
properties of the proposed estimators, as well as the efficiency relative to the full
cohort estimators are assessed via simulation studies. A data set from the Atherosclerosis Risk in Communities (ARIC) study is used to illustrate the proposed
methods.