[ADMB Users] Likelihood profiling in ADMB-RE?

[r.millar at auckland.ac.nz]

Dear Hans and Mark,

Thanks for that.

FYI, I implemented the manual approach (sequence of values within a for
loop in R, read in by the admb executable each time) with the subject
variance component and it worked really well, and was quick.

Cheers,

Russell


> Russell,
>
>
>
> I enquired about this a year or so ago and it was not implemented then.
>
>
>
> Mark
>
>
>
>
>
> ________________________________
>
> From: users-bounces at admb-project.org
> [mailto:users-bounces at admb-project.org] On Behalf Of H. Skaug
> Sent: Tuesday, March 17, 2009 10:09 PM
> To: Russell Millar
> Cc: users at admb-project.org
> Subject: Re: [ADMB Users] Likelihood profiling in ADMB-RE?
>
>
>
> Russel,
>
> I do not think the profile likelihood functionality is available
> in ADMB-RE (but I could be wrong). I would argue that it is not so
>
> much needed with there, because the nuisance parameters
>
> would typically be treated as random effects. If you
>
> still need a profile likelihood, you can of course always
>
> get that manually by fixing your interest parameter at different
>
> values, while fitting the model.
>
> Hans
>
>
>
> On Tue, Mar 17, 2009 at 7:00 PM, Russell Millar
> <r.millar at auckland.ac.nz> wrote:
>
> Dear ADMB'ers again,
>
> I just wanted to ask whether the profile likelihood functionality was
> available
> in ADMB-RE? (I'm getting a runtime error when using the -lprof options).
>
> Once again, many thanks for making this software freely available and
> accessible through R. I'm really enjoying getting more familiar with it,
> and
> putting it through its paces.
>
> Regards,
>
> Russell
>
> _______________________________________________
>
>