Far East Journal of Theoretical Statistics
Volume 13, Issue 2, Pages 233 - 256
(July 2004)
|
|
A UNIFORM SEMIPARAMETRIC APPROACH FOR LONGITUDINAL DATA ANALYSIS
Xian Zhou (Hong Kong), Jianguo Sun (U. S. A.) and Liuquan Sun (P. R. China)
|
Abstract: Longitudinal data commonly occur in medical follow-up studies and epidemiological experiments. They usually include repeated measurements of the response variable and covariates at a set of distinct, irregularly spaced time points for each subject. One of the difficulties for the analysis of such data is that the set of observation times may vary from subject to subject. For their analysis, a number of methods have been proposed, but most of them were developed under specific models. In this paper, a class of general and uniform models is presented for semiparametric analysis of longitudinal data. For inference about regression parameters, a class of consistent and asymptotically normal estimators is proposed. Extensive simulation studies are conducted and an example with data from an AIDS clinical trial is presented to illustrate the proposed methodology. |
Keywords and phrases: counting process, estimating function, longitudinal study, repeated measurements. |
|
Number of Downloads: 256 | Number of Views: 622 |
|