Dear R-users,
I have used the nomogram function from the rms package for a logistic
regresison model made with lrm(). Everything works perfectly (r version
2.15.1 on a mac). My question is this: if my final model is not the one
created by lrm, but I internally validated the model and 'shrunk' the
regression coefficients and computed a new intercept, how can I build a
nomogram using that object? Please see the simplified code for details:
library(rms)
x1<-rnorm(100,1,1)
x2<-rnorm(100,1,1)
y<-rbinom(100,1,0.5)
d<-data.frame(x1,x2,y)
attach(d)
ddist<-datadist(d)
options(datadist='ddist')
model<-lrm(y~x1+x2, x=TRUE, y=TRUE, data=d)
plot(nomogram(model))
##Nomogram is printed, as expected
##Now the model is internally validated, and regression coefficients are
penalized
bootstrap<-validate(model, bw=FALSE, B=100)
shrinkage<-round(bootstrap[4,5],2)
final<-round(model$coef*shrinkage, 3)
final.lp<-cbind(model$x)%*%final[-1]
final["Intercept"]<-round(lrm.fit(y=d$y, offset=final.lp)$coef,3)
final.lp<-final[1]+model$x%*%final[-1]
##The object 'final' now contains all model parameters, yet in a different
fashion than the lrm-created model. Does anyone have an idea how to make
this compatible again with nomogram()?
Thanks in advance for any thoughts on it.
Best wishes,
Sander
[[alternative HTML version deleted]]
______________________________________________
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.
