Re: [Gretl-devel] the "add" and "omit" commands
On Thu, 21 Jul 2011, Riccardo (Jack) Lucchetti wrote:
On Thu, 21 Jul 2011, Allin Cottrell wrote:
> On Thu, 21 Jul 2011, I wrote, re. changes to add/omit
> in CVS:
>
>> * The --quiet flag infuences the printout only, and
>> has no effect on whether or not the current model is
>> replaced.
An improvement, no question about it. However, this has to be signalled as a
backward incompatibility (because it is, albeit minor).
>> * A new --test-only option can be used to prevent
>> replacement of the current model. This can be applied
>> for all variants of "omit", although it's already
>> implicit in the --wald option.
I like it. Actually, I like it so much that it may be worth considering to
turn the logic around (again, at the cost of backward incompatibility, but if
we have to break it, let's do it once and for all). How about having the
--full option work like in the restrict command after VECM estimation?
Yes, I'm all for that, as an explicitly backward-incompatible
change. And if we do that, I think we could solve the other
problem I mentioned (potential confusion over F-tests) by
printing the test first, and then (if --full is engaged and
--quiet is not on) printing the new model in its entirety.
We could keep the GUI relatively backward-compatible (if we
want to) by turning the --full option on unless the user
unchecks an appropriate box.
Allin