Skip to main content English


The measures of predictive accuracy and explained variation in Cox regression presented in Schemper and Henderson 2000 and also dealt with in Schemper 2003 can be computed by means of a SAS macro. By courtesy of Janez Stare, University of Ljubljana, Slovenia, and Lara Lusa, National Cancer Institute, Milan, Italy, independently developed versions in R and S-Plus are also made available below. Finally, this page provides a link to more powerful and recently developed R and S-Plus programs by Lara Lusa.


SAS code:
The ZIP archive contains the files (the SAS macro), surev.pdf (a short documentation of macro options), and (a SAS file containing the myeloma data set analyzed by Schemper and Henderson 2000, and an example macro call).

R code for R version 2.x.x:
A ZIP archive contains an R package developed by L. Lusa, Miceli and Mariani in 2007 (available as R source package or pre-compiled R package for Windows) as well as R functions provided by Janez Stare.

R function for R version 3.x.x:
The function f.surev() from the above R package has been updated by Hana Sinkovec in 2019.


Lusa, L., Miceli, R., Mariani, L. (2007): "Estimation of predictive accuracy in survival analysis using R and S-Plus", Computer Methods and Programs in Biomedicine 87, 132 - 137
Dunkler D., Michiels S, Schemper M. (2007): "Gene expression profiling: Does it add predictive accuracy to clinical characteristics in cancer prognosis?" European Journal of Cancer 2007; 43(4): 745-751.
Schemper, M. (2003): "Predictive accuracy and explained variation", Statistics in Medicine 22, 2299 - 2308
Schemper, M., Henderson, R. (2000): "Predictive Accuracy and Explained Variation in Cox Regression", Biometrics 56(1), 249 - 255
Schemper, M. (1990): "The explained variation in proportional hazards regression", Biometrika 77, 216 - 218 (and "Correction" 1994 in: Biometrika 81, 631)


SAS code
  R code for R 2.xx
  R function for R 3.x.x