cv.HCgglasso: cv.HCgglasso

In HCgglasso: Group-Lasso with Hierarchical Clustering

Description

k-fold cross validation for HCgglasso

Usage

cv.HCgglasso(X, y, nfolds = 5, lambda = NULL, hc = NULL,
  weightLevel = NULL, weightSizeGroup = NULL, intercept = TRUE,
  verbose = FALSE, ...)

Arguments

X	matrix of size n*p
y	vector of size n
nfolds	number of folds
lambda	lambda values for group lasso. If not provided, the function generates its own values of lambda
hc	output of hclust function. If not provided, hclust is run with ward.D2 method
weightLevel	a vector of size p for each level of the hierarchy. A zero indicates that the level will be ignored. If not provided, use 1/(height between 2 successive levels)
weightSizeGroup	a vector
intercept	should an intercept be included in the model ?
verbose	print some informations
...	Others parameters for cv.gglasso function

Value

a cv.HCgglasso object containing :

lambda

values of lambda.

cvm

the mean cross-validated error.

cvsd

estimate of standard error of cvm

cvupper

upper curve = cvm+cvsd

cvlower

lower curve = cvm-cvsd

lambda.min

The optimal value of lambda that gives minimum cross validation error cvm.

lambda.1se

The largest value of lambda such that error is within 1 standard error of the minimum.

time

computation time

Author(s)

Quentin Grimonprez

See Also

HCgglasso, stability.HCgglasso, predict.cv.gglasso, coef.cv.HCgglasso, plot.cv.HCgglasso

Examples

set.seed(42)
X = simuBlockGaussian(50,12,5,0.7)
y = drop(X[,c(2,7,12)]%*%c(2,2,-2)+rnorm(50,0,0.5))
res = cv.HCgglasso(X,y)

HCgglasso documentation built on May 2, 2019, 4:54 p.m.