Title: Time dependent association for bivariate interval censored data
Authors: Yang-Jin Kim - Sookmyung Women University (Korea, South) [presenting]
Abstract: The aim is to suggest a time dependent association measure for bivariate interval censored data. There are many statistical methods to consider the dependency between two failure time variables. Most approaches are global measures with shortage not reflecting a local dependency and based on the marginal survival function and joint survival function. However, the estimation of joint survival function under bivariate interval censored data seems to be difficult. Pseudo partial likelihood is extended to bivariate interval censored data. A two-stage procedure is proposed for the estimation. Simulation studies are conducted to assess the finite sample properties of the presented estimates. Real data from dairy cow udder infection time is analyzed for illustration.