Given a covariance matrix, return a correlation matrix that has unit diagonals. We strictly impose (and check) whether the given input is a symmetric matrix of full-rank.
cov2corr(mat)Arguments
- mat
a \((p\times p)\) covariance matrix.
Value
a \((p\times p)\) correlation matrix.
Examples
# \donttest{
# generate an empirical covariance scaled
prep_mat = stats::cov(matrix(rnorm(100*10),ncol=10))
prep_vec = diag(as.vector(stats::runif(10, max=5)))
prep_cov = prep_vec%*%prep_mat%*%prep_vec
# compute correlation matrix
prep_cor = cov2corr(prep_cov)
# visualize
opar <- par(no.readonly=TRUE)
par(mfrow=c(1,2), pty="s")
image(prep_cov, axes=FALSE, main="covariance")
image(prep_cor, axes=FALSE, main="correlation")
par(opar)
# }