Random survival forests for competing risks

Hemant Ishwaran; Thomas A Gerds; Udaya B Kogalur; Richard D Moore; Stephen J Gange; Bryan M Lau

doi:10.1093/biostatistics/kxu010

Random survival forests for competing risks

Biostatistics. 2014 Oct;15(4):757-73. doi: 10.1093/biostatistics/kxu010. Epub 2014 Apr 11.

Authors

Hemant Ishwaran¹, Thomas A Gerds², Udaya B Kogalur³, Richard D Moore⁴, Stephen J Gange⁵, Bryan M Lau⁵

Affiliations

¹ Division of Biostatistics, University of Miami, Miami, FL 33136, USA hemant.ishwaran@gmail.com.
² Department of Biostatistics, University of Copenhagen, 1014 Copenhagen, Denmark.
³ Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH 44195, USA.
⁴ Department of Medicine, School of Medicine, Johns Hopkins University, Baltimore, MD 21205, USA.
⁵ Department of Epidemiology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD 21205, USA.

Abstract

We introduce a new approach to competing risks using random forests. Our method is fully non-parametric and can be used for selecting event-specific variables and for estimating the cumulative incidence function. We show that the method is highly effective for both prediction and variable selection in high-dimensional problems and in settings such as HIV/AIDS that involve many competing risks.

Keywords: AIDS; Brier score; C-index; Competing risks; Cumulative incidence function; Ensemble.

Publication types

Research Support, N.I.H., Extramural
Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

Data Interpretation, Statistical*
HIV Infections / drug therapy
HIV Infections / mortality
Humans
Models, Statistical*
Risk*
Survival Analysis*

Abstract

Publication types

MeSH terms

Grants and funding