Cox Proportional Hazards Model

Introduction

Cox Proportional Hazards Model

The PHREG procedure in SAS/STAT software performs regression analysis of survival or duration data based on the Cox proportional hazards model. Cox's semiparametric model is widely used in the analysis of survival time, failure time, or other duration data to explain the effect of exogenous explanatory variables.

The population under study may consist of a number of subpopulations, each of which has its own baseline hazard function. PROC PHREG performs a stratified analysis to adjust for such subpopulation differences. The Cox model also allows time-dependent explanatory variables. A time-dependent variable is one whose value for any given individual can change over time. Time-dependent variables have many useful applications in survival analysis.

PROC PHREG includes the following features:

tests of linear hypotheses about the regression parameters
conditional logistic regression analysis for matched case-control studies
creates an output SAS data set containing survivor function estimates and residuals
creates a SAS data set containing estimates of the survivor function at all event times for a given realization of the explanatory variables
four model selection methods: forward selection, backward selection, stepwise selection, and best subset selection
several methods for handling ties

PROC PHREG is also useful for the econometric analysis of discrete choice models.

Chapter Contents
Previous
Next
Top