Folks,
I have been working on VAR and VECM models and trying to simulate the results.
This is easy to do with a VAR model (in the "vars" package) as the
covariance matrix is easily extracted.
Unfortunately I can't figure out how to do this with a VECM fit.
I have used vec2var and created the matrices of coefficients on the lagged
levels but I am unable to get the covariance matrix of the errors.
In the VAR model I use the following matrix to simulate into the future:
unemp.l1 charges.l1 afford.l1 unemp.l2 charges.l2
afford.l2
unemp 0.75438541 -0.0196317870 1.216323 -0.438810574 0.0172960608
-4.619829202
charges 0.53544527 -0.2116682965 10.260630 0.605645192 -0.0028414949
-7.064256277
afford -0.00626801 0.0009240125 0.430278 -0.005179754 0.0003161908
-0.007898937
1.00000000 0.0000000000 0.000000 0.000000000 0.0000000000
0.000000000
0.00000000 1.0000000000 0.000000 0.000000000 0.0000000000
0.000000000
0.00000000 0.0000000000 1.000000 0.000000000 0.0000000000
0.000000000
Along with the covariance matrix that I get from summary(fit)$covres.
Is there an equivalent for VECM?
Thanks for your time,
KW
[[alternative HTML version deleted]]