[R] Cox Calibration regression test & double graphs

[Frank E Harrell Jr]

See the calibrate.* functions and the val.surv (for external validation) 
functions in the rms package.  Note especially the new continuous 
calibration curve methods using hazard spline regression.

Frank


David Winsemius wrote:
> A) Please do not highjack threads. Post new topics for new questions.
> 
> B) Calibration plots are very easy for either logistic regression or Cox 
> models when using the calibrate function in either Harrell's rms or 
> Design packages. (Not sure about how well they play with survfit 
> objects, but you ought to be able to apply the same methods.)
> 


-- 
Frank E Harrell Jr   Professor and Chairman        School of Medicine
                      Department of Biostatistics   Vanderbilt University