[despammed] RE: [R] multidimensional function fitting

[Liaw, Andy]

If something like the second-order function does not fit your data well, it
may well be that the data do not admit a simple structure that you can
easily code in C.

If you expect the structure of the function to be simple, tell gam() so by
specifying a small dimensional basis (via the k= argument in s()).  This
will probably ease the computational burden.

HTH,
Andy

> -----Original Message-----
> From: RenE J.V. Bertin [mailto:rjvbertin at despammed.com]
> Sent: Thursday, February 27, 2003 3:42 PM
> To: Wiener, Matthew
> Cc: r-help at stat.math.ethz.ch
> Subject: Re: [despammed] RE: [R] multidimensional function fitting
> 
> 
> On Thu, 27 Feb 2003 13:52:50 -0500, "Wiener, Matthew" 
> <matthew_wiener at merck.com> wrote regarding
> "[despammed] RE: [R] multidimensional function fitting"
> 
> 8-) Take a look at package mgcv.  Hope this helps.  --Matt
> 8-) 
> 
> Thank you, I just did. It may indeed be what I'm looking for 
> (I haven't quite understood everything about it...), but:
> 
> 1) The best fits I obtain with a formula like z~s(x,y) ; but 
> this I cannot possibly transport into the C programme where I 
> need it! Maybe I wasn't clear on this aspect?
> 
> 2) It is very memory hungry, esp. when using the s() 
> function: I have 192Mb with 256Mb swap (not a lot, but 
> reasonable I'd say), and I've never had to kill R as often as 
> when trying gam()...
> 
> R.B.
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> http://www.stat.math.ethz.ch/mailman/listinfo/r-help
> 

------------------------------------------------------------------------------