ridge.net: Partial correlations with ridge regression.

In parcor: Regularized estimation of partial correlation matrices

Description

This function computes the matrix of partial correlations via an estimation of the corresponding regression models via Ridge Regression.

Usage

ridge.net(X, lambda, plot.it = FALSE, scale = TRUE, k = 10,verbose=FALSE)

Arguments

X	matrix of observations. The rows of X contain the samples, the columns of X contain the observed variables.
lambda	Vector of penalty terms.
scale	Scale the columns of X? Default is scale=TRUE.
k	Number of splits in k-fold cross-validation. Default value is k=10.
plot.it	Plot the cross-validation error as a function of lambda? Default is FALSE.
verbose	Print information on conflicting signs etc. Default is verbose=FALSE

Value

pcor	estimated matrix of partial correlations.
lambda.opt	optimal value of lambda for each of the ncol regression models.

Author(s)

Nicole Kraemer

References

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

ridge.cv

Examples

n<-20
p<-40
X<-matrix(rnorm(n*p),ncol=p)
pc<-ridge.net(X,k=5)

Example output

Loading required package: MASS
Loading required package: glmnet
Loading required package: Matrix
Loading required package: foreach
Loaded glmnet 2.0-12

Loading required package: ppls
Loading required package: splines
Loading required package: Epi

Attaching package: 'Epi'

The following object is masked from 'package:base':

    merge.data.frame

Loading required package: GeneNet
Loading required package: corpcor
Loading required package: longitudinal
Loading required package: fdrtool
Performing local ridge regressions
Vertex no 10 ...20 ...30 ...40 ...

parcor documentation built on May 1, 2019, 9:10 p.m.