Skip to content

lowess C code

4 messages · pawel, cstrato@EUnet.at, Brian Ripley

Original
pawel
# 
Hi,
I understand that the core of lowess() function is written in C.
Is the C source code available and if yes, how can I get to it?

I want to write a multidimensional extension of lowess() if you know of
anybody who has implemented this already I would greatly appreciate any
references or pointers.
thank you very much,
pawel


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Brian Ripley
#
On Tue, 23 Oct 2001, pawel wrote:

            

      
      
      
    

        
It's in the R sources in src/appl/lowess.c

            

      
      
      
    

        
Look at loess in package modreg for one such.

            

      
      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      cstrato@EUnet.at
    

    

      
      #
    

  


  

      
Hi,

It would be great to have a C function extending lowess to multiple
dimensions, i.e. about 100 dimensions. It is not quite clear to me
whether this is possible.
As far as I understand, loess can support only up to 14 dimensions.
Is this correct?

Best regards
Christian Stratowa
----------------------------------
C.h.r.i.s.t.i.a.n  S.t.r.a.t.o.w.a
V.i.e.n.n.a,  A.u.s.t.r.i.a

        
Prof Brian Ripley wrote:

        

            

      
      
      
    

        
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

  
  


  


      

    

    

      
      

        


  

        

      Brian Ripley
    

    

      
      #
    

  


  

      
On Wed, 24 Oct 2001, cstrato at EUnet.at wrote:

        

            

      
      
      
    

        
It would most probably be statistical nonsense.  Local smoothing just
isn't local in high-dimensional spaces without enormous volumes of data.
Indeed, with 100-dimensional inputs you are going to be hard-pressed to
make any non-linear method work well without needing far more data than R
could cope with.

            

      
      
      
    

        
You have the sources ....