The RATS Software Forum

Hi Tom,

i have estimated a bivariate BEKK-GARCH with asymmetry and i want to do a impulse response for the estimated volatility, can i use the code available following Hafner Herwartz (2006) methodology?

thank you for your help

Fadi

No. Not with asymmetry. It looks like you already asked a similar question:

http://www.estima.com/forum/viewtopic.php?f=11&t=1432

That type of calculation can compute the unconditional covariance matrix. It can't be adapted to a VIRF because you can't get the serial conditioning that's required.