Inference on the Occurrence/Exposure Rate with Mixed Censoring Models

Abstract

ABSTRACT:

There are many kinds of measures of risk rate for a disease. In this paper we consider one of them known as the occurrence/exposure rate. We use a sample from a mixed censoring model to estimate this risk rate. The asymptotic properties of the empirical risk rate is discussed. The classical asymptotic and bootstrap methods are employed to construct confidence intervals for the theoretical risk rates. The results of this paper are useful in medical research.

AMS Subject Classfications: 62E20, 62GOS, 62G15, 62P10.