Regression Trees and Ensembles for Cumulative Incidence Functions

• Activity Number: 522 - Life Science Applications of Data Science
• Type: Contributed
• Date/Time: Thursday, August 11, 2022 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Learning and Data Science
• Author(s): Youngjoo Cho* and Annette Molinaro and Chen Hu and Robert L Strawderman
• Companies: Konkuk University and University of California San Francisco and Johns Hopkins University and University of Rochester
• Keywords: CART; Random Forests; Brier score; Competing risks; Fine and Gray model; Sub-distribution function

Abstract

The use of cumulative incidence functions for characterizing the risk of one type of event in the presence of others has become increasingly popular over the past decade. The problems of modeling, estimation and inference have been treated using parametric, nonparametric and semi-parametric methods. Efforts to develop suitable extensions of machine learning methods, such as regression trees and related ensemble methods, have begun comparatively recently. In this paper, we propose a novel approach to estimating cumulative incidence curves in a competing risks setting using regression trees and associated ensemble estimators. The proposed methods employ augmented estimators of the Brier score risk as the primary basis for building and pruning trees, and lead to methods that are easily implemented using existing R packages. Data from the Radiation Therapy Oncology Group (trial 9410) is used to illustrate these new methods. Moreover, we discuss extension of proposed random forests to causal inference.