NONPARAMETRIC STATISTICAL TEST FOR CENSORED DATA

Dana-Farber Cancer Institute, Boston MA

Investigators

Linked publications & trials

Abstract

DESCRIPTION (Adapted from applicant's abstract): The main theme of this project is the investigation of methodology for the censored or missing data that commonly arises in clinical trials and cohort studies in cancer research. More specifically, we will investigate: 1. Methods for analyzing the cost or resource utilization in cancer studies with incomplete follow-up. 2. Adaptive inference based on families of either regression models or rank tests when a most efficient model cannot be specified in advance. 3. Models that adjust for missing data or partial information in either the cause of failure in right censored data or in longitudinally measured response data, such as in quality of life scores. 4. Group sequential designs for monitoring survival probabilities, as opposed to hazard ratios, in cancer clinical trials.