Understanding the PH Regression Models with Time Change Covariates

Mai Zhou

Abstract:

Proportional hazards regression model, or Cox regression model is widely used today. It is a cornerstone of modern survival analysis. But proportional hazards models with time change covariates are not easy to understand/visualize. We offer a simple and easy to understand interpretation of the (arbitrary) baseline and time-change covariate. This also offers a way to simulate variables that follow a proportional hazards model with arbitrary baseline and time-change covariate.

Here is the full postscript text for this technical report. It is 95437 bytes long.