scio (version 0.9.0)

scio.cv: Sparse Column-wise Inverse Operator

Description

Cross validated estimates of a sparse inverse covariance matrix using Sparse Column-wise Inverse Operator

Usage

scio.cv(X, lambda.max=1, alpha=0.95, cv.maxit=1e2, ...)

Arguments

X

Input data of dimension samples (n) x variables (p).

lambda.max

Maximum lambda to start with in CV, which is decreased by mutliplying alpha in each iteration.

alpha

Scaling factor to decrease lambda by multiplication.

cv.maxit

Maximum number of CV iterations. Default 1e2.

...

Other option parameters in scio.

Value

A list with components:

w

Estimated inverse covariance matrix

lambda.cv

CV selected lambda

Details

This is a fast, nonparametric approach to estimate sparse inverse covariance matrices, with possibly really large dimensions. Details of this procedure are described in the reference.

This function does a simple cross validation based on likelihood.

References

Weidong Liu and Xi Luo (2012). Fast and Adaptive Sparse Precision Matrix Estimation in High Dimensions. arXiv:1203.3896.

Examples

# NOT RUN {
set.seed(100)
x<-matrix(rnorm(50*20),ncol=4)
a<-scio.cv(x)
# }