Innovative Statistical Approaches to Modeling Multiple Data from the NSABP BCPT
Abstract
The goals of this IDEA award are two-fold, namely, to develop and refine statistical methodology for the analysis of a data set where multiple outcomes or disease incidence endpoints are of interest and to apply these methods to the analysis of a data set in prevention such as the National Surgical Adjuvant Breast and Bowel Project's Breast Cancer Prevention Trial. The software for the fitting of the B-spline models, the first type of model that was proposed in this study is now being used to fit these models and the simulation studies are also complete. Software development for the pseudospline approach is nearly complete. Extensions to the analysis of recurrent event data outcomes are now being developed.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2001
- Accession Number
- ADA403334
Entities
People
- Kiros Berhane
- Lisa A. Weissfeld
Organizations
- University of Pittsburgh