[Gretl-users] AR-GARCH-t MLE

Riccardo (Jack) Lucchetti r.lucchetti at univpm.it
Fri Mar 23 03:49:59 EDT 2012


On Fri, 23 Mar 2012, Daniel Bencik wrote:

>
> Hello guys,
>
> on a last note. I had minor misspecifications in the ll function, so the right one for arma-garch is
>
> <hansl>
> mle ll = lngamma((dof + 1)/2) - lngamma(dof/2) - 0.5*log(dof-2) - 0.5*log(h)   - 0.5*(dof + 1)*log(1 + (dof - 2)^(-1)*h^(-1)*e^2)
> </hansl>
>
> However, Im gettin into a problem (again!). The estimation does not 
> converge. For some simpler specifications ( I mean simpler than 
> ARMA(7,5)-GARCH(1,1)-t ) I was able to help the algo to converge by 
> setting different parameter starting values. However, with this 
> complicated model, even when I set the initial values equal to "true" 
> values found by eviews, the algo still does not converge. Is there a way 
> of finding out why this happens and/or tweaking the optimization algo?

With all due respect:

1) Unless you have over one billion observations, then estimating an 
ARMA(7,5) model is asking for trouble

2) I wouldn't put my life in the hands of Eviews' optimising algorithms.


Riccardo (Jack) Lucchetti
Dipartimento di Economia

Università Politecnica delle Marche
(formerly known as Università di Ancona)

r.lucchetti at univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti


More information about the Gretl-users mailing list