Statistics and Algorithms / GARCH Models / GARCH Models (Multivariate) / MV GARCH Covariance Matrices |
The HMATRICES option provides the covariance matrices as a SERIES of SYMMETRIC arrays. The MVHSERIES option gives them as a SYMMETRIC of SERIES. The latter is more convenient if you want to graph one of the components, or use it as an input to (for instance) a regression. The former is more useful for things like forecasting the variance, where you need to be able to work with the covariance matrix as a whole.
The following uses HMATRICES and the %CVTOCORR function to generate and graph the conditional correlations. %CVTOCORR transforms a covariance matrix to a correlation matrix, then the (1,2) etc. extracts a particular element from that.
Note that the process of extracting and analyzing the covariances or correlations is the same for all types of multivariate GARCH models.
garch(p=1,q=1,mv=dcc,variances=koutmos,hmatrices=hh) / $
xjpn xfra xsui
set jpnfra = %cvtocorr(hh(t))(1,2)
set jpnsui = %cvtocorr(hh(t))(1,3)
set frasui = %cvtocorr(hh(t))(2,3)
spgraph(vfields=3,footer="Conditional Correlations")
graph(header="Japan with France",min=-1.0,max=1.0)
# jpnfra
graph(header="Japan with Switzerland",min=-1.0,max=1.0)
# jpnsui
graph(header="France with Switzerland",min=-1.0,max=1.0)
# frasui
spgraph(done)
Copyright © 2025 Thomas A. Doan