Gaining more flexibility in Cox proportional hazards regression models with cubic spline functions

被引：259

作者：

Heinzl, H

Kaider, A

机构：

[1] Dept. of Medical Computer Sciences, University of Vienna, A-1090 Vienna

来源：

COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE | 1997年 / 54卷 / 03期

关键词：

proportional hazards regression model; nonlinear functional relationship; time-by-covariate interaction; binary time-dependent covariate; restricted cubic spline function; SAS macro;

D O I：

10.1016/S0169-2607(97)00043-6

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

The Cox proportional hazards model is the most popular model for the analysis of survival data. The use of cubic spline functions allows investigation of non-linear effects of continuous covariates and flexible assessment of time-by-covariate interactions. Two main advantages are provided-no particular functional form has to be specified and standard computer software packages like SAS or BMDP can be used. A SAS macro which implements the method is presented. (C) 1997 Elsevier Science Ireland Ltd.

引用

页码：201 / 208

页数：8

共 6 条

[1]

COX DR, 1975, BIOMETRIKA, V62, P269, DOI 10.1093/biomet/62.2.269

[2]

COX DR, 1972, J R STAT SOC B, V34, P187

[3] FLEXIBLE REGRESSION-MODELS WITH CUBIC-SPLINES [J].