[R] User request to ppr (projection pursuit regression, library(modreg))

Markus-Hermann Koch kochmn at cip.mathematik.uni-stuttgart.de
Wed May 29 17:37:24 CEST 2002


Dear R-Help-Supporters,

I am working on a regression problem using a set of trainig data (X|Y) and
the R-function ppr() in the modreg library. I have succeeded in having
ppr()
create an object B which contains the vectors B$alpha and coefficients
B$beta for the ridge functions. But now i am stuck fast. Could you please
either give me an adress to someone who knows ppr() or, if that
happens to be yourself, help me with the following two questions:

* The ridge functions ought to be of the form f_i(alpha_i*X) where the f_i
  are univariate regression estimates, i.e. a nearest neighbour estimate.
  Which type of estimate did you choose for your program?

* After calling B<-ppr(...) and after choosing some x<-c(x1,...,xd). How
  can i make R give me a prediction for y on x?

All the best is sent to you by
        Markus Koch

E-Mail: kochmn at cip.mathematik.uni-stuttgart.de
Tel.  : 0049 7191 / 68 68 110 (Germany)(assuming you are in suisse)

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list