Test the robustess of the cross-validation procedure
This function calculates the estimated K-fold cross-validation for different
values of K.
testCrossValidation(model,Kfold=c(2,5,10,20,30,40,dim(model$data$X)[1]),N=10)
model |
a fitted model from |
Kfold |
a vector containing the values to test (default corresponds to 2,5,10,20,30,40 and the number of observations for leave-one-out procedure) |
N |
an integer given the number of times the K-fold cross-validation is performed for each value of K |
a matrix of all the values obtained by K-fold cross-validation
For each value of K, the cross-validation procedure is repeated N times in
order to get an idea of the dispersion of the Q2 criterion and of the RMSE by K-fold cross-validation.
D. Dupuy
## Not run:
rm(list=ls())
# A 2D example
Branin <- function(x1,x2) {
x1 <- x1*15-5
x2 <- x2*15
(x2 - 5/(4*pi^2)*(x1^2) + 5/pi*x1 - 6)^2 + 10*(1 - 1/(8*pi))*cos(x1) + 10
}
# a 2D uniform design and the value of the response at these points
n <- 50
X <- matrix(runif(n*2),ncol=2,nrow=n)
Y <- Branin(X[,1],X[,2])
mod <- modelFit(X,Y,type="Linear",formula=formulaLm(X,Y))
out <- testCrossValidation(mod,N=20)
## End(Not run)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.