[R-sig-eco] glm-model evaluation

[Kingsford Jones]

On Fri, May 30, 2008 at 3:25 PM, Kingsford Jones
<kingsfordjones at gmail.com> wrote:

>  Asymptotically, AIC is equivalent to leave-one-out
> cross validation, Mallow's Cp, and some other methods for model
> selection.

I forgot to qualify this -- I believe it's true for standard linear
models -- not for the negative binomial glms discussed in this threat.



Kingsford