opt.cv
Fit an M-type smoothing spline to the input data with the smoothing parameter
chosen by robust cross-validation.
DESCRIPTION:
Fit an M-type smoothing spline with the smoothing parameter chosen by robust
cross-validation as described in Eva Cantoni and Elvezio Ronchetti, Resistant
Selection of the Smoothing Paramater for Smooting Splines, 2001, Statistics and
Computing, 11, 141-146.
USAGE:
opt.cv(xx, yy, bornes.opt=c(0., 1.), ind.chuber=1.345)
REQUIRED ARGUMENTS:
- xx
-
Values of the predictor variable.
- yy
Response variable, of the same length as xx.
-
- bornes.opt
-
Interval in which the optimal smoothing parameter has to be found.
OPTIONAL ARGUMENTS:
- ind.chuber
-
Tuning constant of the Huber psi function. The default value is set to 1.345.
VALUE:
A list is returned with the following components:
lambda.opt
The optimal value of the smoothing parameter obtained by minimization of the
robust cross-validation criterion.
cv.opt
The value of the robust cross-validation for the optimal value of the smoothing parameter.
ajust.opt
The fit for the optimal value of the smoothing parameter. It is an object of type
frob(). See this function for more details.
REFERENCES:
Eva Cantoni and Elvezio Ronchetti, "Resistant Selection of the Smoothing
Paramater for Smooting Splines", 2001, Statistics and Computing, 11, 141-146.
SEE ALSO:
frob,opt.RCp
EXAMPLES:
myoptimalfit<- opt.cv(myx,myy,c(0,1))
motif()
plot(myx,myy)
lines(myoptimalfit$ajust.opt$x,myoptimalfit$ajust.opt$estim)