The proposed research involves six specific subject areas: (1) growth curves analyses are appropriate for laboratory or clinical longitudinal studies when response data may be heterogeneous or missing; (2) assessing multivariate normality is valuable when characterizing multivariate data sets and determining whether common parametric multivariate data-analytic procedures may safely be applied to them; (3) the development of mathematical models for monoclonal antibody-tumor systems is of importance to immunologists and oncologists, and will be applicable to the staging of therapeutic monoclonal antibody preparations in various malignant settings; (4) the development of mathematical methods and algorithms for improving the efficiency of sequence comparison operations that are computationally intensive will further understanding of structure and function relationships in nucleic acids, proteins, and other biologically important molecules, in order ultimately to help prevent, diagnose, and treat human disorders; (5) statistical considerations in the design and analysis of in vitro drug testing techniques will be addressed, in order to enhance the applicability of in vitro assays in the clinical setting; (6) the extension of group sequential methods to k (greater than two) armed trials should increase the efficiency of therapeutic evaluations in cancer clinical trials; and, the development of a nonparametric approach to the evaluation of multiple endpoints in clinical trials should provide a useful complement to parametric strategies previously proposed.
Showing the most recent 10 out of 12 publications
