Skip to content

function cv.glm in library 'boot'

2 messages · Jun Ding, Brian Ripley

Jun Ding
# 
Hi, everyone, 
I have a question regarding function cv.glm in library
'boot'. 
Basically cv.glm can  calculate the estimated K-fold
cross-validation prediction error for generalized
linear models. My question is this: if I am fitting a
logit model, what kind of threshold will it use to
calculate the prediction error (saved in 'delta')? It
will use 0.5 as the threshold or pick a threshold such
that the prediction error reaches its minimum?

Thank you!

Best wishes,
Jun
Brian Ripley
# 
Please do RTFM.  It uses the cost function given by its 'cost' argument.

Using your suggestion to choose the threshold is not honest (in the 
technical sense of the word).
On Thu, 29 Dec 2005, Jun Ding wrote:

            

      
      
      
    

        

      
      
    

        
PLEASE do take note of that, as you asked a question which would have 
been answered by doing the homework asked for there.