S. Huet et A. Kaddour, MAXIMUM-LIKELIHOOD-ESTIMATION IN SURVIVAL ANALYSIS WITH GROUPED DATA ON CENSORED INDIVIDUALS AND CONTINUOUS DATA ON FAILURES, Applied Statistics, 43(2), 1994, pp. 325-333
In grouped survival data, the study of the variation in the relative r
isk as a function of covariates depends on the chosen classes for the
covariates. We present a method which allows us to take into account k
nowledge of the individual values of the covariates, for failed indivi
duals only. The method is applied to the study of the relationship bet
ween the relative risk for breast cancer and the dose of ionizing radi
ation received at Nagasaki.