On independence testing: a randomly censored data perspective

Abstract

Here we consider the problem of testing independence in the presence of randomly censored data. In the first part, we focus on the recent adaptations of tests related to Kendall's τ, examining their asymptotic behavior and performance on small sample sizes (see [1]). The second part of the talk explores extensions of these tests within mixture cure models. We will also present some preliminary results and outline directions for future research.