Department of Statistics Seminar
North Carolina State University

presents

Dr. Debajyoti Sinha

Medical University of South Carolina

"Bayesian Methods for cancer prevention trials"

ABSTRACT

We propose methods for Bayesian inference for a class of survival models for cancer relapse data. Specifically, we propose a modeling cancer relapse via a two-stage process. The scope of this model in cancer prevention clinical trails is investigated. Several properties of the proposed models and behavior of their posteriors under a class of noninformative prior are discussed. Also, the properties of a class of informative priors based on historical data are investigated. Several case studies involving melanoma clinical trials are discussed in detail to demonstrate the proposed methodologies

Friday, October 06, 2000

3:35 - 4:35 pm

206 Cox Hall

Refreshments will be served on the second floor of Dabney Hall (left of Room 222) at 3:00 pm.