Online Program Home
My Program

Abstract Details

Activity Number: 496
Type: Contributed
Date/Time: Wednesday, August 3, 2016 : 8:30 AM to 10:20 AM
Sponsor: Biometrics Section
Abstract #320284
Title: Doubly Robust Regression Trees Under Competing Risks
Author(s): Youngjoo Cho* and Robert Strawderman
Companies: University of Rochester Medical Center and University of Rochester Medical Center
Keywords: Survival analysis ; Competing risks ; Cumulative incidence function ; Regression trees ; Machine learning
Abstract:

Estimating the incidence of one disease in the presence of others, or competing risks, is a difficult task. The use of cumulative incidence curves for characterizing risk has become increasingly popular over the past decade, treated using both parametric and semi/nonparametric methods. Efforts to develop suitable extensions of machine learning methods, such as regression trees and related ensemble methods, have begun only recently. In this paper, we develop doubly robust regression trees for estimating cumulative incidence curves in a competing risks setting. Following Steingrimsson et al. (2016), the proposed methods employ augmented estimators of the Brier score risk as the primary basis for building and pruning trees. The proposed methods are easily implemented in the publically available statistical software R. Simulation studies demonstrate utility of our approach under competing risks setting.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

 
 
Copyright © American Statistical Association