adalasso.net {parcor} | R Documentation |
This function computes the matrix of partial correlations based on an estimation of the corresponding regression models via lasso and adaptive lasso respectively.
adalasso.net(X, k = 10,use.Gram=FALSE,both=TRUE,verbose=FALSE)
X |
matrix of observations. The rows of X contain the
samples, the columns of X contain the observed variables. |
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 value 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 =FALSE. |
both |
Logical. If both=FALSE, only the lasso solution is computed. Default is both=TRUE. |
verbose |
Print information on conflicting signs etc. Default is verbose=FALSE |
For each of the columns of X
, a regression model based on
(adaptive) lasso is computed. 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! Finally, the results of the regression models are
transformed via the function Beta2parcor
.
pcor.adalasso |
estimated matrix of partial correlation coefficients for adaptive lasso. |
pcor.lasso |
estimated matrix of partial correlation coefficients for lasso. |
...
Nicole Kraemer
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", preprint
http://ml.cs.tu-berlin.de/~nkraemer/publications.html
n<-50 p<-10 X<-matrix(rnorm(n*p),ncol=p) pc<-adalasso.net(X,k=5)