[R] How to avoid overfitting in gam(mgcv)

Simon Wood s.wood at bath.ac.uk
Wed Oct 3 15:09:30 CEST 2007

> "Here I try to mean by "overfitting" that GCV was significantly SMALLER
> than the mean square error of prediction of the validation data, which
> was randomly selected and not used for regression."
--- so you could try increasing gamma until this is no longer the case. 

> Simon Wood, Mathematical Sciences, University of Bath, Bath, BA2 7AY UK
> +44 1225 386603  www.maths.bath.ac.uk/~sw283

More information about the R-help mailing list