[R] Comparing gam models fitted using ti()

[Bert Gunter]

I wonder if this might be better asked on a statistical list -- e.g.
stats.stackexchange.com
 -- as this seems to involve complex statistical model comparison
issues, which are normally OT here.

-- Bert

Bert Gunter
Genentech Nonclinical Biostatistics
(650) 467-7374

"Data is not information. Information is not knowledge. And knowledge
is certainly not wisdom."
Clifford Stoll




On Mon, Feb 16, 2015 at 10:13 AM, Arnaud Mosnier <a.mosnier at gmail.com> wrote:
> Dear R-Help list,
>
> I want to compare gam models including interaction with simpler models.
> For interaction models, I used gam(Y~ti(X1) + ti(X2) + ti(X1,X2))
> removing the interaction, the models end as Y~ti(X1) + ti(X2)
>
> How those models compare with models with the form Y ~ s(X1) + s(X2)
>
> In my case, the former (i.e. ti() ) have a better AIC value than those with
> s().
>
> Should I keep the ti() form or use only the s() form when I have no
> interaction ?
>
> Thanks for your help !
>
> Arnaud
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.