[R] Survival Analysis - Cox Regression - Confidence Limits for HR based on Likelihood function

Wilmar Igl wilmar.igl at mail.uni-wuerzburg.de
Mon Mar 20 16:03:25 CET 2006


Does anybody know how to make R calculate confidence limits for the 
Hazard Ratio based on the likelihood function when doing Cox 
Regression? As the p Values and the confidence intervals of Wald 
statistics differ from the p Value of the likelihood function 
sometimes "likelihood"-p is <.05 while the 95%-CI is still including 1.

Any help is appreciated.


Dipl.-Psych. Wilmar Igl
c/o Institut für Psychotherapie u. Medizinische Psychologie
Arbeitsbereich Rehabilitationswissenschaften
Marcusstrasse 9-11 (R. 409), 97070 Würzburg
Telefon: 0931/31-2573, FAX: 0931/31-2078
E-mail: wilmar.igl at mail.uni-wuerzburg.de
URL: http://www.psychotherapie.uni-wuerzburg.de/mitarbeiter/igl.html

More information about the R-help mailing list