[ADMB Users] NLMM Model Selection

[H. Skaug]

Hi,

It does not matter if you do the intergration analytically or by
numerical integration.
For model selection it is the number of init_number parameters that matters.

It does however make sense to ask in a spatial glmm "how many degrees
of freedom does my fitted surface have" which is another question
(related to what Ben says),
and will be lower than the number of elements in random_effects_vector.

In ADMB this distinction is clear. You make a decission when you write
your program.
In BUGS it is not so clear, even after you written the program.

Hans




On Sun, Feb 6, 2011 at 8:47 PM, dave fournier <davef at otter-rsch.com> wrote:
> Something that has always bothered me about this random effects stuff is
> that if
> I fit a model with a neg bin dist it is just a parametric model with one
> more parameter than
> a Poisson with a Poisson at the end. I can do standard LR tests and random
> effects never come up.
> But that is just because one can do the integration analytically so that the
> RE nature or
> interpretation never comes up. How can that be?  Why are other RE models
> different?
> _______________________________________________
> Users mailing list
> Users at admb-project.org
> http://lists.admb-project.org/mailman/listinfo/users
>