Some suggestions on possible measures of explained variation which have appeared in the literature are considered. Following this an outline of the recommended approach is given. Leaning upon the theory of explained variation detailed in Chapter 2 and in particular 3.9 we show how a solid theory of explained variation for proportional and non-proportional hazards regression can be established. This contrasts with a substantial body of literature on this topic, almost entirely constructed around intuitive improvisations and ad-hoc modifications to sample based quantities gleaned from classical linear regression. The main reference here is the paper by O'Quigley and Flandre (1994) which showed how the Schoenfeld residuals provide the required ingredients for the task in hand. The properties of population quantities and sample based estimates have been studied thoroughly (O'Quigley and Xu 2001) and these provide the user with the necessary confidence for their practical use.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
Rights and permissions
Copyright information
© 2008 Springer Science+Business Media, LLC
About this chapter
Cite this chapter
(2008). Explained variation. In: Proportional Hazards Regression. Statistics for Biology and Health. Springer, New York, NY. https://doi.org/10.1007/978-0-387-68639-4_13
Download citation
DOI: https://doi.org/10.1007/978-0-387-68639-4_13
Publisher Name: Springer, New York, NY
Print ISBN: 978-0-387-25148-6
Online ISBN: 978-0-387-68639-4
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)