Hi Toam,
The RATS program offers any example to estimate Generalized Impulse Reponse (GIRF) or VIRF in a Var-Garch-M model with bekk representation?
Which one of this could I estimate?
VAR-GARCH-M GIRF
Re: VAR-GARCH-M GIRF
The VIRF for a BEKK doesn't depend whether the model has an "M" effect or not so the calculations would be like in Hafner and Herwartz.
The GIRF is complicated by the presence of "M" terms. It's not even entirely clear what that means since the responses will depend upon the initial conditions of the variances.
The GIRF is complicated by the presence of "M" terms. It's not even entirely clear what that means since the responses will depend upon the initial conditions of the variances.