Page 43 - Special Topic Session (STS) - Volume 1
P. 43
STS353 H. Zhao et al.
Regression analysis of clustered interval-
censored failure time data with linear
transformation models in the presence of
informative cluster size
2
2
2
Hui Zhao , Chenchen Ma , Junlong Li , Jianguo Sun
1
1 Central China Normal University, Wuhan 430079, P.R.China
2 Department of Statistics, University of Missouri, Columbia, Missouri 65211, U.S.A.
Abstract
This paper discusses regression analysis of clustered interval-censored failure
time data, which often occur in medical follow-up studies among other areas.
For such data, sometimes the failure time may be related to the cluster size,
the number of subjects within each cluster or we have informative cluster
sizes. For the problem, we present a within-cluster resampling method for
the situation where the failure time of interest can be described by a class of
linear transformation models. In addition to the establishment of the
asymptotic properties of the proposed estimators of regression parameters,
an extensive simulation study is conducted for the assessment of the finite
sample properties of the proposed method and suggests that it works well
in practical situations. An application to the example that motivated this
study is also provided.
Keywords
Clustered data; Interval-censoring; Informative cluster size; Linear
transformation models; Within-cluster resampling
1. Introduction
This paper discusses regression analysis of clustered interval-censored
failure time data, which often occur in medical follow-up studies among other
areas (Williamson et al., 2003; Zhang and Sun, 2010). For such data, the failure
times of interest are clustered into small groups instead of being
independent and also are known only to lie within certain intervals instead of
being observed exactly or right-censored. In these situations, sometimes the
failure time may be related to the cluster size, the number of subjects within
each cluster, too. In other words, in addition to clustering and interval
censoring, we may also face or have to deal with informative cluster sizes. In
the following, a semiparametric inference procedure is presented for the
problem.
Clustered failure time data arise in a failure time study when some failure
times of interest are dependent with each other. An example of such data is
given by randomized multi-center clinical trials where patients are recruited
and grouped by study centers. In these situations, the patients from the same
32 | I S I W S C 2 0 1 9