[R] LOCFIT:Automatic bandwidth selection for kernel regression

[learnerminer]

Hello all!

I have recently started using the LOCFIT package, together with Clive
Loader's book.  I need to implement some method for automatic (plug-in)
bandwidth selection in a multivariate kernel regression.  From the book, and
the LOCFIT documentation,  it is not clear whether this is possible.  As far
as I can see, the only time the various automated procedures, e.g., SJPI,
are used is in the kdeb function, which is for preparing a summary
comparison of various selection methods, and only for a univariate kernel
density.  Does anyone here know of a way to implement automatic (plug-in)
bandwidth selection for kernel regression, maybe with some R function that
calls LOCFIT?  Or can you suggest a good alternative package for this,
either in R/Matlab or a stand-alone?

Thanks a lot, and regards
-- 
View this message in context: http://www.nabble.com/LOCFIT%3AAutomatic-bandwidth-selection-for-kernel-regression-tf4895390.html#a14020237
Sent from the R help mailing list archive at Nabble.com.