A flexible copula model for bivariate survival data with dependent censoring

This article introduces a flexible copula model for analyzing bivariate survival data with dependent censoring, addressing a significant limitation in traditional statistical methods. The model uses the Joe-Hu copula to capture complex dependence structures, allowing for more accurate estimates of survival functions. The article also discusses the importance of accounting for dependent censoring and provides a detailed comparison of different copula families and their tail dependence properties. Through extensive simulation studies, the authors demonstrate the superior performance of their proposed model compared to naive estimators that assume independent censoring. Additionally, the model is applied to a real dataset from twin prostate cancer studies, highlighting its practical utility. The article concludes with a discussion on future research directions, including the extension to allow for negative dependence and the development of goodness-of-fit test procedures.