[R] Akaike or BIC criteria for feed-forward neural nets

[Prof Brian Ripley]

On Mon, 30 Jun 2008, J. Miguel Marin wrote:

> Hello all,
>
> I would like to know whether there exists any way to calculate the AIC or the 
> BIC criteria for a nnet object (not only in the multinomial case).
> Commands like stepAIC, AIC or extractAIC do not work with a feed forward 
> neural net and I cannot find any command that deals with nnet objects.

They are not defined for most nnet fits -- they need a maximized 
likelihood, and no neural-net fitting package gives you that.

>
> Best
>
> jm~
>
> _______________________________
>
>       J. Miguel Marin
>
> http://www.est.uc3m.es/jmmarin
>
>   Dep. of Statistics
> University Carlos III of Madrid
>       Spain (E.U.)
>
> ______________________________________________
> R-help at 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.
>

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595