On Sat, 22 Jul 2006, Riccardo Jack Lucchetti wrote:
> Is it, then, worth considering a bifurcation of the model
> specification dialog? A "user-friendly" option (which we have
> already) and an "expert" version (which sticks a vamped-up
> version of the current command-line model specification into a
> text-entry dialog box)?
IMO the dialog is ok the way it is now. In the long run,
there's no way any dialog box can match the power and
flexibility of the command line. Playing catch-up would only
lead to a monstrously complex dialog.
I'm not sure I disagree, but perhaps I wasn't clear enough: the
idea I was floating was not a complexification of the current
model dialog, but an option (under Tools/Preferences?) to use
the current dialog or a "type your spec" dialog.
The dialog box serves its purpose just right: a quick
interface for simple tasks. It shouldn't aim to completeness:
simply, big jobs aren't its jobs. Besides, the "expert"
version is there already, in a way: press Alt-X and there you
Yes, pretty much. A possible half-way house would be a dialog
where you get to select options by clicking (dependent on the
sort of model chosen), but you type the actual spec. I'm not
actually advocating this, just "positing" it.
A different question (which makes more sense) is: should we
just drop variables created "on the fly", after estimation is
This is maybe a good point at which to advertise a newly created
mailing list, info for which is at:
The new list is for discussion among gretl developers. The rest
of this posting really belongs there (and will go there in
future), but for the meantime here goes...
In my conception, variables created in the process of parsing a
model specification would be marked as "automatic". They would
remain in the data array (possibly subject to garbage
collection) but would not be displayed in the main window. If
and when they were invoked again, they would be checked for
up-to-dateness and either reused or regenerated as the case may