ABSTRACT
In this chapter, we study the use of empirical likelihood method for current status data, also known as type I interval censored data. We demonstrate how to use the empirical likelihood ratio to test the parameters of (1) mean, (2) the CDF at a given time point t 0, i.e., https://www.w3.org/1998/Math/MathML" display="inline"> F ( t 0 ) , and (3) integrated hazard. We also examine some regression models for the current status data, and in particular how empirical likelihood can be used there. We want to point out that for current status data, the variance of NPMLEs are usually very difficult to estimate, making the empirical likelihood (no need to estimate variance) especially attractive. The software used is mainly in a new R package emplikCS.
