parcor (version 0.2-6)

adalasso: Adaptive Lasso

Description

This function computes the lasso and adaptive lasso solution based on k-fold cross-validation. The initial weights for adaptive lasso are computed from a lasso fit.

Usage

adalasso(X, y, k = 10, use.Gram = TRUE,both=TRUE,intercept=TRUE)

Arguments

X

matrix of input observations. The rows of X contain the samples, the columns of X contain the observed variables

y

vector of responses. The length of y must equal the number of rows of X

k

the number of splits in k-fold cross-validation. The same k is used for the estimation of the weights and the estimation of the penalty term for adaptive lasso. Default is k=10.

use.Gram

When the number of variables is very large, you may not want LARS to precompute the Gram matrix. Default is use.Gram=TRUE.

both

Logical. If both=FALSE, only the lasso coefficients are computed. Default is both=TRUE.

intercept

Should an intercept be included? Default is intercept=TRUE.

Value

intercept.lasso

intercept for lasso. If intercept=FALSE was specified, the intercept is set to 0.

intercept.adalasso

intercept for adaptive lasso. If intercept=FALSE was specified, the intercept is set to 0.

coefficients.adalasso

regression coefficients for adaptive lasso.

coefficients.lasso

regression coefficients for lasso.

cv.lasso

cv error for the optimal lasso model.

cv.adalasso

cv error for the optimal adaptive lasso model.

lambda.lasso

optimal lambda value for lasso-

lambda.adalasso

optimal lambda value for adaptive lasso.

Details

In each of the k-fold cross-validation steps, the weights for adaptive lasso are computed in terms of a lasso fit. (The optimal value of the penalty term is selected via k-fold cross-validation). Note that this implies that a lasso solution is computed k*k times!

References

H. Zou (2006) "The Adaptive Lasso and its Oracle Property", Journal of the American Statistical Association 101 (476): 1418-1429.

N. Kraemer, J. Schaefer, A.-L. Boulesteix (2009) "Regularized Estimation of Large-Scale Gene Regulatory Networks using Gaussian Graphical Models", BMC Bioinformatics, 10:384

http://www.biomedcentral.com/1471-2105/10/384/

See Also

Beta2parcor, adalasso.net

Examples

n<-100 # number of observations
p<-60 # number of variables
X<-matrix(rnorm(n*p),ncol=p) 
y<-rnorm(n)
ada.object<-adalasso(X,y,k=10)