hglassoBIC {hglasso}

BIC-type criterion for hglasso

Description

This function calculates the BIC-type criterion for tuning parameter selection for hglasso proposed in Section 3.4 in Tan et al. (2014)

Usage

hglassoBIC(x, S, c=0.2)

Arguments

x

An object of class hglasso.

S

A p by p correlation/covariance matrix. Cannot contain missing values.

c

A constant between 0 and 1. When c is small, the BIC-type criterion will favor more hub nodes. The default value is c=0.2.

Value

BIC

The calculated BIC-type criterion in Section 3.4 in Tan et al. (2014).

Author(s)

Kean Ming Tan

References

Tan et al. (2014). Learning graphical models with hubs. To appear in Journal of Machine Learning Research. arXiv.org/pdf/1402.7349.pdf.

See Also

hglasso

Examples

#library(mvtnorm)
#library(glasso)
#set.seed(1)
#n=100
#p=100

# A network with 4 hubs
#network<-HubNetwork(p,0.99,4,0.1)
#Theta <- network$Theta
#truehub <- network$hubcol
# The four hub nodes have indices 14, 42, 45, 78
#print(truehub)

# Generate data matrix x
#x <- rmvnorm(n,rep(0,p),solve(Theta))
#x <- scale(x)
#S <- cov(x)
# Run Hub Graphical Lasso with different tuning parameters
#lambdas2 <- seq(0,0.5,by=0.05)
#BICcriterion <- NULL
#for(lambda2 in lambdas2){
#res1 <- hglasso(S,0.3,lambda2,1.5)
#BICcriterion <- c(BICcriterion,hglassoBIC(res1,S)$BIC)
#}
#lambda2 <- lambdas2[which(BICcriterion==min(BICcriterion))]