Title: Extending joint models in terms of boosting algorithms
Authors: Colin Griesbach - FAU Erlangen-Nuernberg (Germany) [presenting]
Elisabeth Waldmann - Friedrich-Alexander-Universitaet Erlangen-Nuernberg (Germany)
Andreas Mayr - University of Bonn (Germany)
Abstract: Joint models turned out to be a powerful approach to analysing data where event times are measured alongside a longitudinal outcome. The idea is to combine a longitudinal and a survival model via a shared predictor used in both original models while a parameter quantifies their relation. To fit a basic joint model efficiently, a gradient boosting algorithm has been presented based on statistical boosting methods for longitudinal data in multiple dimensions. The aim is to extend that algorithm by incorporating a predictor solely for survival data, hence a set of covariates, which are independent of the longitudinal structure, is added.