Moritz Berger, Matthias Schmid
Subdistribution hazard models are a popular tool for competing risks analysis. The classical approach in discrete time consists of fitting parametric models, which focuses on main effects. An alternative tree-based method is proposed that allows for more flexibility, in particular when interactions between the covariates are present. The method is illustrated by an analysis of age-related macular degeneration among elderly people.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados