On May 17, 2013, at 11:55 AM, masumeh akhgar wrote: > hi all > this command used tt function for all variables. > How can i define a different function for each variable? >> exCox.all<-coxph(Surv(SURVT,STATUS) ~ > RX+LOGWBC+SEX+tt(RX)+tt(LOGWBC)+tt(SEX), > data=rem.data,tt=function(x,t,...) log(t)*x))
After reading the help page it seems pretty clear that the 'tt' argument has a very special purpose and that it is not designed to inserting transformations of anything other than time-related transformations. Applying a tt() function to 'logwbc' or to 'SEX' makes no sense at all. In the help page the `tt` function is applied to the age variable and is intended to advance a subject's age value forward as they …. wait for it…. "age" (used now as an English verb rather than a numeric value). Other transformations would need to be built in the usual manner in the formula or by constructing transformed variables in the input dataset. -- David Winsemius Alameda, CA, USA ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.