A small sample study of occurrence/exposure rates for rare events

Abstract
This paper is an investigation of the small sample properties of various confidence procedures based on the asymptotic theory for occurrence/exposure rates using censored exponential survival as the underlying model. The following types of censoring are considered: fixed censoring, uniform censoring and a mixture of fixed and uniform censoring. The study is restricted to small values of the intensity and the investigation is carried out by computer simulations. The large sample approximations seem reasonable if the intensity multiplied by the sample size and by the length of the observational period exceeds ten.

This publication has 3 references indexed in Scilit: