Ensemble methods for survival function estimation with time-varying covariates

被引:12
作者
Yao, Weichi [1 ]
Frydman, Halina [1 ]
Larocque, Denis [2 ]
Simonoff, Jeffrey S. [1 ]
机构
[1] NYU, New York, NY 10012 USA
[2] HEC Montreal, Montreal, PQ, Canada
关键词
Survival forests; time-varying covariates; survival curve estimate; dynamic estimation; left-truncated right-censored survival data; TREES; FORESTS; INFERENCE; MODELS; ERROR;
D O I
10.1177/09622802221111549
中图分类号
R19 [保健组织与事业(卫生事业管理)];
学科分类号
摘要
Survival data with time-varying covariates are common in practice. If relevant, they can improve on the estimation of a survival function. However, the traditional survival forests-conditional inference forest, relative risk forest and random survival forest-have accommodated only time-invariant covariates. We generalize the conditional inference and relative risk forests to allow time-varying covariates. We also propose a general framework for estimation of a survival function in the presence of time-varying covariates. We compare their performance with that of the Cox model and transformation forest, adapted here to accommodate time-varying covariates, through a comprehensive simulation study in which the Kaplan-Meier estimate serves as a benchmark, and performance is compared using the integrated L-2 difference between the true and estimated survival functions. In general, the performance of the two proposed forests substantially improves over the Kaplan-Meier estimate. Taking into account all other factors, under the proportional hazard setting, the best method is always one of the two proposed forests, while under the non-proportional hazard setting, it is the adapted transformation forest. K-fold cross-validation is used as an effective tool to choose between the methods in practice.
引用
收藏
页码:2217 / 2236
页数:20
相关论文
共 42 条
[1]   COX REGRESSION-MODEL FOR COUNTING-PROCESSES - A LARGE SAMPLE STUDY [J].
ANDERSEN, PK ;
GILL, RD .
ANNALS OF STATISTICS, 1982, 10 (04) :1100-1120
[2]  
[Anonymous], 2020, TRTF TRANSFORMATION
[3]   Generating survival times to simulate Cox proportional hazards models with time-varying covariates [J].
Austin, Peter C. .
STATISTICS IN MEDICINE, 2012, 31 (29) :3946-3958
[4]  
Bacchetti P, 1995, Lifetime Data Anal, V1, P35, DOI 10.1007/BF00985256
[5]   Adjusting for time-varying confounding in survival analysis [J].
Barber, JS ;
Murphy, SA ;
Verbitsky, N .
SOCIOLOGICAL METHODOLOGY, 2004, VOL 34, 2004, 34 :163-192
[6]   Discrete-time survival trees and forests with time-varying covariates: application to bankruptcy data [J].
Bou-Hamad, Imad ;
Larocque, Denis ;
Ben-Ameur, Hatem .
STATISTICAL MODELLING, 2011, 11 (05) :429-446
[7]   Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[8]  
COX DR, 1972, J R STAT SOC B, V34, P187
[9]   COVARIANCE ANALYSIS OF HEART-TRANSPLANT SURVIVAL DATA [J].
CROWLEY, J ;
HU, M .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1977, 72 (357) :27-36
[10]   PROGNOSIS IN PRIMARY BILIARY-CIRRHOSIS - MODEL FOR DECISION-MAKING [J].
DICKSON, ER ;
GRAMBSCH, PM ;
FLEMING, TR ;
FISHER, LD ;
LANGWORTHY, A .
HEPATOLOGY, 1989, 10 (01) :1-7