Random survival forests for competing risks

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

Research output: Contribution to journalArticle

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.

Original languageEnglish (US)
Pages (from-to)757-773
Number of pages17
JournalBiostatistics
Volume15
Issue number4
DOIs
StatePublished - 2014

Fingerprint

Competing Risks
Cumulative Incidence Function
Random Forest
Variable Selection
Acquired Immunodeficiency Syndrome
High-dimensional
HIV
Prediction
Incidence
Forests
Competing risks
Variable selection

Keywords

  • AIDS
  • Brier score
  • C-index
  • Competing risks
  • Cumulative incidence function
  • Ensemble

ASJC Scopus subject areas

  • Statistics and Probability
  • Medicine(all)
  • Statistics, Probability and Uncertainty

Cite this

Random survival forests for competing risks. / Ishwaran, Hemant; Gerds, Thomas A.; Kogalur, Udaya B.; Moore, Richard D; Gange, Stephen J; Lau, Bryan M.

In: Biostatistics, Vol. 15, No. 4, 2014, p. 757-773.

Research output: Contribution to journalArticle

Ishwaran, Hemant ; Gerds, Thomas A. ; Kogalur, Udaya B. ; Moore, Richard D ; Gange, Stephen J ; Lau, Bryan M. / Random survival forests for competing risks. In: Biostatistics. 2014 ; Vol. 15, No. 4. pp. 757-773.
@article{39211422d0394262b399359b19e0d6ab,
title = "Random survival forests for competing risks",
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",
author = "Hemant Ishwaran and Gerds, {Thomas A.} and Kogalur, {Udaya B.} and Moore, {Richard D} and Gange, {Stephen J} and Lau, {Bryan M}",
year = "2014",
doi = "10.1093/biostatistics/kxu010",
language = "English (US)",
volume = "15",
pages = "757--773",
journal = "Biostatistics",
issn = "1465-4644",
publisher = "Oxford University Press",
number = "4",

}

TY - JOUR

T1 - Random survival forests for competing risks

AU - Ishwaran, Hemant

AU - Gerds, Thomas A.

AU - Kogalur, Udaya B.

AU - Moore, Richard D

AU - Gange, Stephen J

AU - Lau, Bryan M

PY - 2014

Y1 - 2014

N2 - 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.

AB - 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.

KW - AIDS

KW - Brier score

KW - C-index

KW - Competing risks

KW - Cumulative incidence function

KW - Ensemble

UR - http://www.scopus.com/inward/record.url?scp=84985920576&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84985920576&partnerID=8YFLogxK

U2 - 10.1093/biostatistics/kxu010

DO - 10.1093/biostatistics/kxu010

M3 - Article

C2 - 24728979

AN - SCOPUS:84985920576

VL - 15

SP - 757

EP - 773

JO - Biostatistics

JF - Biostatistics

SN - 1465-4644

IS - 4

ER -