Hello. I want to calculate the covariance matrix for a parameter vector Q[1x24].

In this picture the formula that may be used is displayed. I need this for getting the covariance matrix of a parameter vector. In other applications in which I use AutoRegressive (AR), ARX models, I use matlab's command 'model=arx(y,24)' and to get the covariance matrix I used model.covariancematrix. However, now I use VARMA models and I only have the parameter vector. Is there a way of getting the covariance matrix out of the parameter vector (In AR case: model.parametervector)

Answer 1

[Askic V]

Is this helpful to you?

I have used the example from Matlab documentation.

load iddata9 z9
Ts = z9.Ts;
y = cumsum(z9.y);
model = ar(y, 4, 'ls', 'Ts', Ts, 'IntegrateNoise', true);
param_vector = model.Report.Parameters.ParVector

param_vector = 4×1

-0.7878 -0.4768 -0.0925 0.4653

covar = model.Report.Parameters.FreeParCovariance

covar = 4×4

0.0015 -0.0015 -0.0005 0.0006 -0.0015 0.0027 -0.0008 -0.0004 -0.0005 -0.0008 0.0028 -0.0015 0.0006 -0.0004 -0.0015 0.0014