This is a generalization of %LogDensityCV to complex matrices for computing the multivariate Whittle log likelihood. This computes the concentrated log likelihood. Use %CXLogDensity (http://www.estima.com/forum/viewtopic.php?f=7&t=1452) for the frequency-by-frequency log likelihood.
Because this is a function, not a procedure, you need to do a SOURCE instruction on the file before you can use it.
CVSIGMA should be a Hermitian p.d. matrix which is the theoretical value
CVSIGMAHAT is also Hermitian p.d. and represents the sample value
NOBS is the number of (effective) observations on which the sample is based.