Identifiability conditions for covariate effects model on survival times under informative censoring

摘要

We provide the identifiability conditions for the covariate effects modeling of bivariate survival data when one survival time is dependently censored by the other survival time. The covariate effects are specified through three components of the copula decomposition. Many commonly used copula families are shown to satisfy the identifiable condition. A condition for causal interpretation is also provided.