Dynamic estimation with random forests for discrete-time survival data

被引：8

作者：

Moradian, Hoora ^{[1
]}

Yao, Weichi ^{[2
]}

Larocque, Denis ^{[1
]}

Simonoff, Jeffrey S. ^{[2
]}

Frydman, Halina ^{[2
]}

机构：

[1] HEC Montreal, Dept Decis Sci, Montreal, PQ, Canada

[2] NYU, Stern Sch Business, Dept Technol Operat & Stat, New York, NY USA

来源：

CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE | 2022年 / 50卷 / 02期

基金：

加拿大自然科学与工程研究理事会;

关键词：

Discrete-time survival analysis; landmark analysis; random forests; survival forests; time-varying covariates; TREES; PREDICTION; REGRESSION; RECOVERY;

D O I：

10.1002/cjs.11639

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

Time-varying covariates are often available in survival studies, and estimation of the hazard function needs to be updated as new information becomes available. In this article, we investigate several different easy-to-implement ways that random forests can be used for dynamic estimation of the survival or hazard function from discrete-time survival data. Results from a simulation study indicate that all methods can perform well, and that none dominates the others. In general, situations that are more difficult from an estimation point of view (such as weaker signals and less data) favour a global fit, pooling over all time points, while situations that are easier from an estimation point of view (such as stronger signals and more data) favour local fits.

引用

页码：533 / 548

页数：16

共 42 条

[1] ANALYSIS OF SURVIVAL BY TUMOR RESPONSE [J].