Wang, Yishu (2012) Covariate-adjusted response-adaptive design with longitudinal responses. Masters thesis, Memorial University of Newfoundland.
PDF (Migrated (PDF/A Conversion) from original format: (application/pdf))
- Accepted Version
Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.
Response-adaptive designs have been extensively studied and applied in clinical trials. However, few research is for longitudinal data and less is known on limiting allocation properties when response adaptive designs are used in a longitudinal setting. Zhang et. al. (2007) proposed a general covariate-adjusted response-adaptive (CARA) design for non-longitudinal clinical trials and explored its asymptotic properties. The objectives of this research are to extend the general CARA design to clinical trials with longitudinal responses, and to study the asymptotic properties of the parameter estimators and the allocation proportion. The explicit expressions for the limiting allocation proportions of the extended design are obtained. The generalized estimating equations and martingale theories are used to develop the asymptotic properties of regression parameters and allocation proportions. This research is also the first study on covariate-adjusted response-adaptive design with longitudinal clinical responses and with more than two competing treatments.
|Item Type:||Thesis (Masters)|
|Additional Information:||Includes bibliographical references (leaves 46-53).|
|Department(s):||Science, Faculty of > Mathematics and Statistics|
|Library of Congress Subject Heading:||Clinical trials--Mathematical models; Clinical trials--Longitudinal studies; Stochastic models; Analysis of covariance|
Actions (login required)