Partial correlation coefficients

Share:

Description

Obtains partial correlation coefficients from a given concentration matrix.

Usage

1
qpK2ParCor(K)

Arguments

K

positive definite matrix, typically a concentration matrix.

Details

This function applies cov2cor to the given concentration matrix and then changes the sign of the off-diagonal entries in order to obtain a partial correlation matrix.

Value

A partial correlation matrix.

Author(s)

R. Castelo and A. Roverato

References

Lauritzen, S.L. Graphical models. Oxford University Press, 1996.

See Also

qpG2Sigma

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
require(graph)

n.var <- 5 # number of variables
set.seed(123)
g <- randomEGraph(as.character(1:n.var), p=0.15)

Sigma <- qpG2Sigma(g, rho=0.5)
K <- solve(Sigma)

round(qpK2ParCor(K), digits=2)

as(g, "matrix")

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.