[R-sig-ME] r-square type of measure for cox PH in MCMCglmm

Farrar, David Farrar.David at epa.gov
Tue Mar 17 15:55:27 CET 2015


You might start with this FAQ:  http://glmm.wikidot.com/faq
While analogues may exist, as far as I know there is no good reason to think that an R^2 should exist which should be viewed in the same way for any sort of regression model.


-----Original Message-----
From: R-sig-mixed-models [mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf Of Van Dongen Stefan
Sent: Tuesday, March 17, 2015 9:34 AM
To: r-sig-mixed-models at r-project.org
Subject: [R-sig-ME] r-square type of measure for cox PH in MCMCglmm

Hi All,

I am running a multivariate animal model in MCMCglmm. One response variable is censored, so I use a cox proportional hazards approach for that one. The model is running OK and I get an estimate for the association of the hazard with a fixed covariate. Is there a way to get an R-square estimate for the strength of association?

many thanks

Stefan

-------------------------------------------------------------------------------------------------------------------------
Prof. Dr. Stefan Van Dongen
Evolutionary Ecology - Department of Biology - University of Antwerp (http://www.ua.ac.be/eveco) StatUA Statistics Center, University of Antwerp (http://www.ua.ac.be/statua) Groenenborgerlaan 171 - B-2020 Antwerp - Belgium

room V324b, building V, campus Groenenborger
Tel: + 32 3 265 33 36
Fax: + 32 3 265 34 74
http://www.ua.ac.be/stefan.vandongen
--------------------------------------------------------------------------------------------------------------------------


	[[alternative HTML version deleted]]

_______________________________________________
R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models



More information about the R-sig-mixed-models mailing list