[Gretl-users] maximum like. conditions

Allin Cottrell cottrell at wfu.edu
Sun Nov 29 11:40:43 EST 2009


On Sun, 29 Nov 2009, Davor Horvatic wrote:

> I'm working with different GARCH models and I trying to determine
> parameters.

> For test case I took example from user guide and applied
> restirction to the parameters as stated in user manual, code is
> in the end of the mail. All

> I get is:
>
> Syntax error in command line
> Error executing script: halting
>
> What am I doing wrong ?
>
> Thanks,
> Davor
>
> open djclose
> series y = 100*ldiff(djclose)
> scalar mu = 0.0
> scalar omega = 1
> scalar alpha = 0.4
> scalar beta = 0.0
> mle ll = check ? (-0.5*(log(h) + (e^2)/h)) : NA
> series e = y - mu
> series h = var(y)
> series h = omega + alpha*(e(-1))^2 + beta*h(-1)
> scalar check = (alpha>0) & (beta>0)
> params mu omega alpha beta
> end mle

The operator '&' is now strictly used for "taking the address" of
a variable; for boolean AND you must use "&&".

This has been the case since gretl 1.7.5 (see
http://gretl.sourceforge.net/Backward.html#v1-7-5 )

We didn't notice the instance in section 17.2 of the User's Guide
where '&' was used in the old way.  Thanks for spotting this; it's
now fixed in CVS.

Allin Cottrell


More information about the Gretl-users mailing list