Valid and approximately valid confidence intervals for current status data

Abstract We introduce a new approach for creating pointwise confidence intervals the distribution of event times current status data. Existing methods are based on asymptotics. Our is binomial properties and motivates that very simple to apply valid guarantee nominal coverage. Although these necessarily conservative small sample sizes, asymptotically their coverage rate approaches one. This also approximately intervals, simulations show approximate generally have rates closer level with shorter length than existing such as interval likelihood ratio test. Unlike previous asymptotic require different distributions continuous or grid-based assessment, can be applied either type assessment distribution.