VAR-GARCH BEKK and irf

[moderator]

I think the SHOCKS option is documented fairly clearly in the "FORECAST" section of the Reference Manual.

Briefly, you can either define the vector of shocks ahead of time, or provide the vector using the literal matrix notation. For example, if you had a 3 equation model, the option would be something like

shocks=||s1, s2, s3||

replacing s1, s2, and s3 with the desired shock for each equation.

[nikosant]

I think what luxu1983 wishes to get is based on the following paper: Hafner & Herwartz 2006 - Volatility impulse responses for multivariate GARCH models: An exchange rate illustration.

[luxu1983]

I think the SHOCKS option is documented fairly clearly in the "FORECAST" section of the Reference Manual.

Briefly, you can either define the vector of shocks ahead of time, or provide the vector using the literal matrix notation. For example, if you had a 3 equation model, the option would be something like

shocks=||s1, s2, s3||

replacing s1, s2, and s3 with the desired shock for each equation.

dear
if i use a MAXIMIZE mgarch variation (bekk)
how can i get the irf

[TomDoan]

dear
if i use a MAXIMIZE mgarch variation (bekk)
how can i get the irf

I would suggest defining the VAR using SYSTEM, then put the estimated VAR coefficients into it using the %modelsetcoeffs. You just need to make sure you get the coefficient matrix set up in the right form. Once you've done that, you can just use IMPULSE the way you would if you were estimating the model using the GARCH instruction.

[vinayadhikari]

I think the above gives the impulse responses of the mean equation. Can anyone share the code of impulse responses in variances as in
& Herwartz 2006? Thank you.

[TomDoan]

That's at http://www.estima.com/forum/viewtopic.php?f=8&t=1466