I suppose neural nets (e.g., in the nnet package, part of the VR bundle)
sort of fits that description: It has been shown to have the "universal
approximator" property. Read MASS4, or Pattern Recognition and Neural
Networks (Ripley, 1996) for details.
I don't know of anything as automated as you seem to want, but I can say
that if there's one, I wouldn't trust it. There's no Free Lunch
after all.
Andy
From: Michael>
> Hi all,
>
> Looking at my data, I decided to try exploring regression
> models. Is there a function in R that allows me to
> automatically explore the universe of (almost all) the
> possible regression models?
>
> Do general additive models belong to the category of
> regression models? Is there a function in R includes all of
> these and does model selection?
>
> Thanks a lot!
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
> http://www.R-project.org/posting-guide.html
>
>
