Random survival forests

TL;DR

This paper presents random survival forests, a novel method for analyzing right-censored survival data, introducing new splitting rules, missing data algorithms, and an interpretable mortality measure, with implementation in R.

Contribution

It introduces random survival forests with new survival splitting rules, a missing data algorithm, and an ensemble mortality measure, advancing survival analysis techniques.

Findings

Effective in analyzing right-censored data

Provides interpretable mortality predictions

Implemented in accessible R package

Abstract

We introduce random survival forests, a random forests method for the analysis of right-censored survival data. New survival splitting rules for growing survival trees are introduced, as is a new missing data algorithm for imputing missing data. A conservation-of-events principle for survival forests is introduced and used to define ensemble mortality, a simple interpretable measure of mortality that can be used as a predicted outcome. Several illustrative examples are given, including a case study of the prognostic implications of body mass for individuals with coronary artery disease. Computations for all examples were implemented using the freely available R-software package, randomSurvivalForest.