Re: [Gretl-users] gretl: forecasting
On Wed, 22 Jun 2005, Ricardo Gonçalves Silva wrote:
I believe there is a bug in the GARCH routine of this build.
I not able even to fit a GARCH model.
The VAR's routines are OK.
Thank you for checking. I did find some bugs in the GARCH
forecasting code, which are now fixed.
As for not being able to fit a GARCH model, was the problem failure
of convergence? If so, this is something we have seen before in
some cases, but we don't have an easy fix. If you could send me a
copy of the dataset and the specification of the GARCH model that
might be helpful in debugging.
Allin Cottrell.