User-defined Kernels

Hi,
i would like to do a local linear regression but with a user-defined
function f(x,y) as kernel.
(not the typical Gaussian or Epanechnikov kernel function but some similar
function).
More, would it be possible to do a local linear regression by adding weights
to the (x,y) data points?
Many thanks in advance.

View this message in context: http://www.nabble.com/User-defined-Kernels-tp15335660p15335660.html
Sent from the R help mailing list archive at Nabble.com.