[R] A question in rms package
dwinsemius at comcast.net
Fri Jan 3 08:19:31 CET 2014
On Jan 2, 2014, at 11:03 AM, Agony wrote:
> Dear all,
> Happy new year all of You and with best wishes coming to you in this new year.
> I have a problem in running a command in rms package.
> Does any can help me with my problem.
> I wanna use predab.resample command to compute bias-corrected estimates of a vector of indexes of Predictive accuracy. But I'm confused about choosing fit and measure arguments.
> Would you please help me?
> I ran this command:
> model1=lrm(Cancerous~AGE+PV+PSA,na.action=na.delete,x=TRUE, y=TRUE, data=mdat1)
> now when I want to use this one:
> predab.resample<-predab.resample(model, method=c("boot","crossvalidation",".632","randomization"),
Not spelled the same.
> bw=TRUE, B=50, pr=FALSE,
> rule="aic", type="residual", sls=.05, aics=0,
> tol=1e-12, non.slopes.in.x=TRUE)
> I received this error:
> Error in predab.resample(model, method = c("boot", "crossvalidation", :
> argument "fit" is missing, with no default
Alameda, CA, USA
More information about the R-help