lossGCV: Generalized cross-validation loss
In kernDeepStackNet: Kernel Deep Stacking Networks

Description Usage Arguments Value Author(s) References See Also Examples

Calculates the generalized cross-validation score given number of observations, model deviance, effective degrees of freedom and gamma.

1	lossGCV(n, Dev, trA, gammaPar=1)

`n`	Number of observations (integer scalar).
`Dev`	Deviance of the kernel ridge regression (numeric scalar).
`trA`	Effective degrees of freedom (numeric scalar).
`gammaPar`	Weighting parameter (numeric scalar), which specifies how the generalized cross-validation score is penalized by the effective degrees of freedom. Default value is 1.

Generalized cross-validation loss (numeric scalar).

Thomas Welchowski welchow@imbie.meb.uni-bonn.de

Simon N. Wood, (2006), Generalized Additive Models: An Introduction with R, Taylor \& Francis Group LLC

lossApprox, lossCvKDSN, lossSharedCvKDSN, lossSharedTestKDSN

# Simple example based on help pages of mgcv::gam
# GCV Score is the same as used in function mgcv::mgcv
library(mgcv)
dat <- gamSim(1, n=100, dist="normal", scale=2)
gamMod <- gam(y~s(x0)+s(x1)+s(x2)+s(x3),data=dat)
calcGCV <- lossGCV(n=100, Dev=deviance(gamMod), trA=sum(gamMod$edf), gammaPar=1)
all.equal(calcGCV, gamMod$gcv.ubre) # TRUE