A1152
Title: Analysis of errors-in-variables competing risks data in discrete time
Authors: Chi-Chung Wen - Tamkang University (Taiwan) [presenting]
Abstract: Analysis of competing risk data has been an important topic in survival analysis due to the need to account for the dependence among competing events. Also, event times are often recorded on discrete time scales, rendering the models tailored for discrete-time nature useful in the practice of survival analysis. Regression analysis with discrete-time competing risks data and the errors-in-variables issue where the covariates are prone to measurement errors are considered. Without assuming a distribution for the true covariate, we develop robust sufficient score methods for the cause-specific and subdistribution hazards models. TEfficient computation algorithms can implement the proposed estimators, and the associated large sample theories can be simply obtained.
