CMStatistics 2018: Start Registration
View Submission - CMStatistics
B1379
Title: Joint model for bivariate zero inflated recurrent event data with terminal event Authors:  Yang-Jin Kim - Sookmyung Women University (Korea, South) [presenting]
Abstract: Multivariate recurrent event arises when a subject has experienced several types of event repeatedly over time. These observations can be stopped by a terminal event which may be related with the recurrent event. Furthermore, there exist a substantial portion of subjects without any recurrent events during a fairly long follow-up time which results in a zero-inflated nature of data. For simultaneously considering both zero inflation and terminal event in a context of multivariate recurrent event data, a joint model is applied to model several dependencies through frailty effects; Correlation between recurrent events, one between recurrent event and terminal event and one between cure fraction and terminal event. Diverse simulation studies are performed to evaluate the suggested models. Infection data from AML (acute myeloid leukemia) patients are analyzed as an application.