I have problems getting predict.smooth.Pspline (part of the pspline package) to work. Here's an example on artificial data:

########start#########
library (Pspline)
tt <- seq (0,1,length=20)
xt <- tt^3

fit <- smooth.Pspline (tt, xt, norder=3,df=4, method=2)

fit <- smooth.Pspline (tt, xt, norder=3,spar=0.0001, method=1)
plot (tt,xt)
lines(fit)       # so far everything works fine - looks fine at least

predict.smooth.Pspline (fit, tt, nderiv=0)[,1]

####### end ##########

This produces only 20x NaN instead of something similar to xt


At the end I'm trying to get 2nd derivatives of quintic spline fits. Does anybody know how to get predict.smooth.Pspline or is there another way to do this? Any help would be greatly appreciated. Cheers,


Martin

______________________________________________
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help

Reply via email to