MBLasso {DWLasso}

Inferring the network using nodewise regression method

Description

This function infers the network using nodewise regression method by Meinhausen and Buhlmann.

Usage

  MBLasso(dat,lambda=0.4,w.mb)

Arguments

dat

An input matrix. The columns represent variables and the rows indicate observations.

lambda

A penalty parameter of the weighted Lasso that controls the sparsity of the inferred network.

w.mb

An unput weight vector which is computed from the degree of the inferred network.

References

Meinshausen, Nicolai, and Peter Bühlmann. "High-dimensional graphs and variable selection with the lasso." The annals of statistics (2006): 1436-1462.

Examples

library(DWLasso)
library(glmnet)
library(hglasso)


# Generate inverse covariance matrix with 3 hubs
# 20 % of the elements within a hub are zero
# 97 % of the elements that are not within hub nodes are zero
p <- 60 # Number of variables
n <- 40 # Number of samples

hub_number = 3  # Number of hubs

# Generate the adjacency matrix
Theta <- HubNetwork(p,0.97,hub_number,0.2)$Theta

# Generate a data matrix
out <- rmvnorm(n,rep(0,p),solve(Theta))

# Standardize the data
dat <- scale(out)

# Infer the network using weighted nodewise regression
w.mb <- rep(1,p)
adj.mat <- MBLasso(dat,lambda=0.4,w.mb)