AD GRANT NUMBER DAMD17-94-J-4332 TITLE: Statistical Methods for Analyzing Time-Dependent Events in Breast Cancer Chemoprevention Studies PRINCIPAL INVESTIGATOR:

Interval-censored data consist of n pairs of observations (k,ri), i =1, ...,n, where h < r{. We either observe the exact survival time X if Zj = r* or only knowX G (k,ri) otherwise. We established the asymptotic normality of the nonparametric MLEof a survival function S(t) (= Pr(X > t) with such interval-censored data and presentan estimate of the asymptotic variance of the MLE. We show that the convergence ratein distribution is in ^/n. Simulation study also supports our result. An application to thecancer research is presented.Paper presented in person, contributed paper. (2) Institute of Mathematical Statistics Meeting no. 247 4-8 August 1996 Chicago, Illinois TITLE: ESTIMATION OF A SURVIVAL FUNCTION WITH CASE 1 INTERVAL-CENSORED DATAQiqing /Yu , Anton /Schick, Linxiong /Li and George Y. C. /WongSUNY at Binghamton, University of New Orleans and Strang Cancer Preventive Insti-tute ABSTRACT: Case 1 interval censored data consists of either right-censored data orleft-censored data but not exact observations. Let F(x) and G(y) be the distribution funcCase 1 interval censored data consists of either right-censored data orleft-censored data but not exact observations. Let F(x) and G(y) be the distribution func-