[R] coefficients of each local polynomial from locfit

[Liaw, Andy]

I believe the prediction is done some some sort of grid, then
interpolated to fill in the rest.  This is, however, purely for
computational reason, and not for any threoretical reasons.  The formal
definition of local polynomials is to do a weighted fit of polynomial at
each point.

Andy 

> -----Original Message-----
> From: r-help-bounces at r-project.org 
> [mailto:r-help-bounces at r-project.org] On Behalf Of David Winsemius
> Sent: Tuesday, December 08, 2009 12:45 PM
> To: David Grabiner
> Cc: r-help at r-project.org
> Subject: Re: [R] coefficients of each local polynomial from locfit
> 
> 
> On Dec 8, 2009, at 12:33 PM, David Grabiner wrote:
> 
> > Hi list,
> >
> > This was asked a couple of years ago but I can't find a 
> resolution.   
> > Is
> > there any way to get the coefficients from one of the local  
> > polynomial fits
> > in locfit.  I realize that locfit only constructs polynomials at a  
> > handful
> > of intelligently selected points and uses interpolation to predict  
> > any other
> > points.
> 
> That is not my understanding of what locfit does. I could be 
> wrong and  
> am interested in correcting my misunderstanding if that is 
> the case. I  
> thought a polynomial was fit at _each_ point.  What pages in 
> Loader's  
> book should I be reading.
> 
> 
> >  I would like to know the terms of the polynomials at these points.
> > It seems likely that they are stored in [locfit 
> object]$eva$coef but  
> > I'm
> > having trouble making sense of that list.  That list is the second  
> > argument
> > passed into the C function spreplot called from preplot.locfit.raw  
> > if that
> > helps.
> >
> > Thanks,
> > David
> --
> 
> David Winsemius, MD
> Heritage Laboratories
> West Hartford, CT
> 
> ______________________________________________
> R-help at r-project.org 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.
> 
Notice:  This e-mail message, together with any attachme...{{dropped:10}}