[R] non linear regression

Olivier Martin olivier.martin at inrialpes.fr
Thu Jul 12 16:20:22 CEST 2001


Hi all,

Let x and y be two real predictor variables and z the real variable
reponse such that
z=f(x,y)+e.  Could you give some methods to estimate f and the R
packages if there are available.
f is unknown and i would be interested by kernel regression methods  or
methods based on projection to estimate it.
In the one dimension case, my problem can be solved with the function
lowess.
So, what should be the solution(s) in two dimensions ?

Thanks,
Olivier

--
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
Olivier MARTIN               phone: (33) 04 76 61 53 55
Projet IS2                               06 08 67 93 42
INRIA Rhone-Alpes            fax  : (33) 04 76 61 54 77
655, Av. de l'Europe
Montbonnot                  e-mail:olivier.martin at inrialpes.fr
38334 Saint Ismier cedex
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-


-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://stat.ethz.ch/pipermail/r-help/attachments/20010712/65c913b9/attachment.html


More information about the R-help mailing list