DESP (version 0.2-2)

DESP_MST: Estimation of DESP using minimum spanning trees

Description

This function estimates the diagonal of the precision matrix by symmetry-enforced likelihood minimization using minimum spanning trees, when the true value of the coefficient matrix B is known or has already been estimated. The observations of the data matrix X are assumed to have zero mean.

Usage

DESP_MST(X, B, Theta = NULL)

Arguments

X

The data matrix.

B

The coefficient matrix.

Theta

The matrix orresponding to outliers.

Value

This function returns the diagonal of the precision matrix associated with X as a vector.

Details

When Theta is not NULL, we consider an additive contamination model. We assume that X = Y + E is observed, denoting the outlier-free data by Y and the matrix of errors by E. In this case, the matrix Theta should be equal to E * B.

Examples

## build the true precision matrix
p <- 12 # number of variables
Omega <- 2*diag(p)
Omega[1,1] <- p 
Omega[1,2:p] <- 2/sqrt(2)
Omega[2:p,1] <- 2/sqrt(2)
## compute the true diagonal of the precision matrix
Phi <- 1/diag(Omega)
## generate the design matrix from a zero-mean Gaussian distribution
require(MASS)
n <- 100 # sample size
X <- mvrnorm(n,rep.int(0,p),ginv(Omega))
## compute the sample mean
barX <-  apply(X,2,mean)
## subtract the mean from all the rows
X <- t(t(X)-barX)
## estimate the coefficient matrix 
B <- DESP_SRL_B(X,lambda=sqrt(2*log(p)))
## compute the squared partial correlations
SPC <- DESP_SqPartCorr(B,n)
## re-estimate the coefficient matrix by ordinary least squares
B_OLS <- DESP_OLS_B(X,SPC)
## estimate the diagonal of the precision matrix and get its inverse
hatPhiMST <- 1/DESP_MST(X,B_OLS)
## measure the performance of the estimation using l2 vector norm
sqrt(sum((Phi-hatPhiMST)^2))