[Rd] Possible bug in summary.survfit - 'scale' argument ignored?

[Marc Schwartz]

Thomas,

I worked on the plane home last night and have a modified version that  
restores the original behavior vis-s-vis the interaction between the  
'times' and 'scale' arguments, so that 'times' are in the 'scale'  
transformed range.

I am attaching both a patch file and the complete function code. The  
patch is against the current version in R-Forge. If you should try to  
run the full code file to test this, note that survmean() is not  
exported in the survival NAMESPACE, hence if run 'as is' there will be  
an error message about this function not being found.

I would of course advise appropriate review and testing before  
considering committing this.

Regards,

Marc



-------------- next part --------------