Am 17.01.2019 um 07:52 schrieb Riccardo (Jack) Lucchetti:
On Wed, 16 Jan 2019, Elisabetta M. wrote:
> Hello everyone!
> I am currently working on an ARCH model. I already carried out my ARCH 1
> model and got the coefficients in my script output.
only found
> something of
> the sort on GARCH models ($h), that unfortunately doesn't work with ARCH.
> Thank you for your time and your help!
Does this help?
<hansl>
open djclose
series r = 100*ldiff(djclose)
garch 0 1 ; r
series vola = $h
</hansl>
To provide some background, the help for 'arch' says: "This command is
retained at present for backward compatibility, but you are better off
using the maximum likelihood estimator offered by the garch command; for
a plain ARCH model, set the first GARCH parameter to 0."
In the GUI you also go directly through garch and then you can save the
predicted variance from the menu.
(BTW, doing 'arch 1 r' --i.e. specifying no explicit constant-- provokes
just a generic data error. I think the error message could be more
informative. Apart from that, aren't there legitimate cases where you
are working with mean-zero series?)
cheers,
sven