Addressing the selection uncertainty in association structure in the joint modelling framework
Conference
65th ISI World Statistics Congress 2025
Format: IPS Abstract - WSC 2025
Keywords: aggregation, association, joint models
Session: IPS 757 - New Developments and Insights in Joint Modeling of Longitudinal and Survival Outcomes
Thursday 9 October 10:50 a.m. - 12:30 p.m. (Europe/Amsterdam)
Abstract
There has been an increasing interest in applying joint models to related longitudinal and survival outcome data in clinical studies. In the joint modelling framework, the choice of association structure that links the longitudinal and survival outcome sub-models is of fundamental importance. Often, information criterion such as AIC are applied to identify an appropriate joint model specification from a range of possible association structures. Under this approach, a single (best fit) joint model is used to draw the inference from estimated parameters, which overlooks the issue of model uncertainty entirely.
To allow for the association structure uncertainty, we developed a novel methodology which can collate parameter information from multiple joint models with different association structures. The proposed approaches have been investigated through simulation studies in both frequentist and Bayesian settings of the joint model, and illustrated with real-world clinical data.