R help - May 2007 - comparing fit of cubic spline

I want to compare the fit of a quadratic model to continuous data, with that
of a cubic spline fit. Is there a way of computing AIC from for e.g. a GAM
with a smoothing spine, and comparing this to AIC from a quadratic model?

 

Cheers

 

******************************************

Tom Reed
PhD Student
Institute of Evolutionary Biology
102 Ashworth Laboratories
Kings Buildings
University of Edinburgh
Edinburgh EH9 3JT



tom.reed@ed.ac.uk


Tel. 00 44 (0)131 6505462
http://homepages.ed.ac.uk/ecunning/tom.html
http://homepages.ed.ac.uk/loeske/tom.html

******************************************

 


	[[alternative HTML version deleted]]

Thomas Reed <tom.reed <at> ed.ac.uk> writes:
> 
> I want to compare the fit of a quadratic model to continuous data, with
that
> of a cubic spline fit. Is there a way of computing AIC from for e.g. a GAM
> with a smoothing spine, and comparing this to AIC from a quadratic model?
  
  Why don't you try AIC() on each of the fitted models and
see what happens?  In general R can compute log-likelihoods
and AICs for most fitted model objects ...  If it doesn't work,
you can post your simple example to the list and get more
advice then ...

  Ben Bolker

On Monday 21 May 2007 16:17, Thomas Reed wrote:
> I want to compare the fit of a quadratic model to continuous data, with
> that of a cubic spline fit. Is there a way of computing AIC from for e.g. a
> GAM with a smoothing spine, and comparing this to AIC from a quadratic
> model?
library(mgcv)
AIC(gam(y~s(x))) ## a `generalized aic' since effective df used 
AIC(gam(y~x+I(x^2)))

>
>
>
> Cheers
>
>
>
> ******************************************
>
> Tom Reed
> PhD Student
> Institute of Evolutionary Biology
> 102 Ashworth Laboratories
> Kings Buildings
> University of Edinburgh
> Edinburgh EH9 3JT
>
>
>
> tom.reed at ed.ac.uk
>
>
> Tel. 00 44 (0)131 6505462
> http://homepages.ed.ac.uk/ecunning/tom.html
> http://homepages.ed.ac.uk/loeske/tom.html
>
> ******************************************
>
>
>
>
> 	[[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 and provide commented, minimal,
> self-contained, reproducible code.
-- 
> Simon Wood, Mathematical Sciences, University of Bath, Bath, BA2 7AY UK
> +44 1225 386603  www.maths.bath.ac.uk/~sw283