Speaker: 段博文，云南大学

Inviter: 万林

Title: Two-Stage estimation method in bivariate copula regression models for semi-competing risks data

Language: Chinese 

Time & Venue: 2024.12.11 15:40-16:20  思源楼415

Abstract: In medical study, the semi-competing risk data is omnipresent, where a terminal event can censor a non-terminal event, but not vice versa. The two events are related and often affected by some other covariates respectively, and for some reason, the covariates data are often unobserved. In this article, we combine the copula-based regression model with Multiple Imputation by Chained Equation (MICE) method to estimate the association parameters and the regression coefficient. The performance of proposed methods is assessed through simulation studies are motivated by and illustrated with an analysis of data from Amsterdam Cohort Study on HIV/AIDS.