It computes either a non-parametric estimate of the covariance matrix or a parametric form of the covariance matrix.
sigma(obj, ...)A matrix of non-parametric estimates or a matrix of zeros and ones, say A, such that A*theta is
the halfvectorized Sigma,
where Sigma is the parametric covariance matrix.
If the first argument is of class GTree the method returns the empirical covariance matrix
of the data in the GTree object.
If the first argument is of class HRMtree the method returns
a matrix of ones and zeros which corresponds to \(hvec(\Sigma(\theta))=A\theta\)
Object of class GTree or HRMtree
additional arguments