Likelihood ratio type two-sample tests for current status data

TL;DR

This paper develops nonparametric likelihood ratio tests for two-sample current status data, allowing different observation distributions, with bootstrap methods for critical values and simulation comparisons of test power.

Contribution

It introduces a new class of likelihood ratio tests for current status data that accommodate different observation distributions, extending previous methods.

Findings

Tests perform well in simulations with Weibull distributions

Bootstrap method effectively determines critical values

New tests show competitive power compared to existing methods

Abstract

We introduce fully nonparametric two-sample tests for testing the null hypothesis that the samples come from the same distribution if the values are only indirectly given via current status censoring. The tests are based on the likelihood ratio principle and allow the observation distributions to be different for the two samples, in contrast with earlier proposals for this situation. A bootstrap method is given for determining critical values and asymptotic theory is developed. A simulation study, using Weibull distributions, is presented to compare the power behavior of the tests with the power of other nonparametric tests in this situation.