constrained: Constrained Precision Matrix

In donaldRwilliams/IRCcheck: Irrepresentable Condition Check

Description

Compute the maximum likelihood estimate, given certain elements are constrained to zero (e.g., an adjacency matrix). This approach is described in \insertCitehastie2009elements;textualIRCcheck.

Usage

constrained(Sigma, adj)

Arguments

Sigma	Covariance matrix
adj	Matrix with constraints. A zero indicates that element should be constrained to zero.

Value

A list containing the inverse covariance matrix and the covariance matrix.

Note

The algorithm is written in c++.

References

\insertAllCited

Examples

# random adj 
# 90 % sparsity (roughly)
p <- 20
adj <- matrix(sample(0:1, size = p^2, replace = TRUE, 
              prob = c(0.9, 0.1) ), 
              nrow = p, ncol = p)

adj <-  symm_mat(adj)

diag(adj) <- 1

# random correlation matrix
set.seed(1)
cors <- cov2cor(
  solve(
  rWishart(1, p + 2, diag(p))[,,1])
)

# constrain to zero
net <- constrained(cors, adj = adj)

donaldRwilliams/IRCcheck documentation built on Dec. 20, 2021, 12:12 a.m.