cggd (version 0.8)

cv.cggd: Computes K-fold cross-validated error curve for cggd

Description

Computes the K-fold cross-validated mean squared prediction error for cggd.

Usage

cv.cggd(x, y, nfolds = 6, kmax=40 ,  trace = FALSE, plot.it = TRUE, se=TRUE, ...)

Arguments

x

Input to cggd

y

Input to cggd

nfolds

Number of folds

kmax

Max number of iterations per model

trace

Show computations

plot.it

Plot it

se

Include standard error bands

...

Additional arguments to cggd

Value

cv

The CV loss curve at each value of k

cv.error

The standard error of the CV curve

References

Cun-Hui Zhang (2007) "Continuous Generalized Gradient Descent" Journal of Computational and Graphical Statistics ; see also http://stat.rutgers.edu/~cunhui/software/CGGD.html.

Examples

data(Wine)
attach(Wine)
cv.cggd(x.learning,y.learning,kmax=40,fctr=1e3)
detach(Wine)