Chapter

Extending the Cox Regression Model

Judith D. Singer and John B. Willett

in Applied Longitudinal Data Analysis

Published in print May 2003 | ISBN: 9780195152968
Published online September 2009 | e-ISBN: 9780199864980 | DOI: http://dx.doi.org/10.1093/acprof:oso/9780195152968.003.0015
Extending the Cox Regression Model

More Like This

Show all results sharing this subject:

  • Public Health and Epidemiology

GO

Show Summary Details

Preview

This chapter begins by describing how to include time-varying predictors in the Cox regression model. It then introduces two methods for relaxing the proportionality assumption. Section 15.2 presents the stratified Cox regression model, which stipulates that while the effects of each predictor are identical across strata, the baseline hazard functions can differ. Section 15.3 presents an alternative strategy that closely mirrors the approach used in discrete time: the inclusion of interactions with time as predictors in the model. Section 15.4 introduces a range of regression diagnostics useful for examining the underlying assumptions of the Cox model. Section 15.5 discusses what to do when modeling “competing risks”—multiple events that compete to terminate an individual's lifetime. Section 15.6 concludes by describing what to do when you have not observed the beginning of time for everyone in your sample and there are so-called late entrants to the risk set.

Keywords: time-varying predictors; Cox regression models; hazard models time; baseline hazard; discrete time

Chapter.  23499 words.  Illustrated.

Subjects: Public Health and Epidemiology

Full text: subscription required

How to subscribe Recommend to my Librarian

Buy this work at Oxford University Press »

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.