On Tue, 9 Apr 2013, Riccardo (Jack) Lucchetti wrote:
On Mon, 8 Apr 2013, Allin Cottrell wrote:
> I think it would be a good idea to make the MAIC criterion the
> default for "test-down". This would be backward incompatible, in the
> sense that the same script, using the --test-down option with the
> "adf" command, and with no additional parameter, would (in general)
> produce different results under gretl >= 1.9.13 versus gretl <=
> 1.9.12. But you could still get the old behaviour under new gretl
> using --test-down=tstat.
> So my question to you: Does this sound OK? Any objections?
If I'm not mistaken, the policy so far has been to avoid backward
incompatible changes, unless there's a good reason and as long as they're
porperly documented in the file we keep for the purpose.
The reasons you mention seem very good to me: if a statistical method has
been proven inferior, keeping it as the default is actually close to a bug.
So I'd say let's go ahead.
Thanks (and to Peter too). That's now done in CVS.