[R] Another question on locfit
Ralf.Kloessinger at student.fh-nuernberg.de
Thu Apr 4 00:50:56 CEST 2002
Thank you all for your suggestion on my last question about locfit.
Now I have another question:
I would like to change the evalution structure ev in locfit() to a vector/matrix.
I created a vector and a matrix:
> vec <- mat.or.vec(101, 1)
> vec <- c(0:100)
> mat <- matrix(data = vec, nrow = 101, ncol = 1)
and I tried to change the ev parameter:
> fit <- locfit(~imposter, kern="gauss", renorm=T, xlim=c(1e-20,100), ev=vec)
This works good but when I tried to get the density points I get an error, this happens also when I use the matrix (mat) for ev:
> fhat <- fitted(fit)
Error in fitted.locfit(fit) : NA/NaN/Inf in foreign function call (arg 10)
I think arg 10 means "dp" and there I got the following parameters:
nnalph fixh adpen cut lk df1 df2 rv
0.7 0.0 0.0 0.8 0.0 0.0 0.0 1.0
When I tried locfit with ev="tree", lk, df1, df2, rsc are changed, so I think there is the problem!!
If I use the predict() function, I get also an error message:
> pre <- predict.locfit(fit, mat)
Error in preplot.locfit.raw(object, newdata, where, what, band)
NA/NaN/Inf in foreign function call (arg 2)
But here the problem could be the newdata parameter. I don't know an other way to get around of this problems to get what I need.
Also when I change the xlim to c(0,100) I get for the fitted() and predict() function only 1s in each row. It would be nice to know, why this happen.
Thank you for your time and your help
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