8 p.m.
[switching to devel]
Am 03.07.2010 18:19, schrieb Allin Cottrell:
Following some recent discussion here, I have now factored testing
for fractional integration out of the "pergm" command and provided
better and more visible access via a new command, "fractint".
The fractint command uses the Local Whittle Estimator by default
but has an option to use GPH. It supports the $test and $pvalue
accessors, and has a --quiet option.
The GUI menu items have been reorganized correspondingly. This is
in CVS and snapshots. Thanks to Javier G for spotting some
breakage, and to Sven S for suggesting that we clean up the
previous muddle!
Sorry for not testing earlier, and thanks for the separation. Just one
menu layout thought: maybe the unit-root-related test menu entries in
the Variable menu could be put in a new submenu? (I mean the new
fractional integration entry, ADF, ADF-GLS, and KPSS.) Seems quite
logical and natural to me.
thanks,
sven
8:20 p.m.
New subject: [Gretl-users] pergm and fractint
On Sat, 3 Jul 2010, Sven Schreiber wrote:
Sorry for not testing earlier, and thanks for the separation. Just
one
menu layout thought: maybe the unit-root-related test menu entries in
the Variable menu could be put in a new submenu? (I mean the new
fractional integration entry, ADF, ADF-GLS, and KPSS.) Seems quite
logical and natural to me.
I agree.
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
8:37 p.m.
New subject: [Gretl-users] pergm and fractint
Em 03/07/2010, às 17:20, Riccardo escreveu:
On Sat, 3 Jul 2010, Sven Schreiber wrote:
> Sorry for not testing earlier, and thanks for the separation. Just one
> menu layout thought: maybe the unit-root-related test menu entries in
> the Variable menu could be put in a new submenu? (I mean the new
> fractional integration entry, ADF, ADF-GLS, and KPSS.) Seems quite
> logical and natural to me.
I agree.
Me too ;-)
Best,
Henrique
Enviado via iPhone
9:20 p.m.
New subject: [Gretl-users] pergm and fractint
On Sat, 3 Jul 2010, Riccardo (Jack) Lucchetti wrote:
On Sat, 3 Jul 2010, Sven Schreiber wrote:
> Sorry for not testing earlier, and thanks for the separation. Just one
> menu layout thought: maybe the unit-root-related test menu entries in
> the Variable menu could be put in a new submenu? (I mean the new
> fractional integration entry, ADF, ADF-GLS, and KPSS.) Seems quite
> logical and natural to me.
I agree.
I'm OK with that. Suggestions for a string for the parent menu
item?
Allin
9:39 p.m.
New subject: [Gretl-users] pergm and fractint
On Sat, 3 Jul 2010, Allin Cottrell wrote:
On Sat, 3 Jul 2010, Riccardo (Jack) Lucchetti wrote:
> On Sat, 3 Jul 2010, Sven Schreiber wrote:
>
>> Sorry for not testing earlier, and thanks for the separation. Just one
>> menu layout thought: maybe the unit-root-related test menu entries in
>> the Variable menu could be put in a new submenu? (I mean the new
>> fractional integration entry, ADF, ADF-GLS, and KPSS.) Seems quite
>> logical and natural to me.
>
> I agree.
I'm OK with that. Suggestions for a string for the parent menu
item?
How about "Unit root tests"?
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
11:33 p.m.
New subject: [Gretl-users] pergm and fractint
On Sat, 3 Jul 2010, Riccardo (Jack) Lucchetti wrote:
On Sat, 3 Jul 2010, Allin Cottrell wrote:
>
> On Sat, 3 Jul 2010, Riccardo (Jack) Lucchetti wrote:
>
>> On Sat, 3 Jul 2010, Sven Schreiber wrote:
>>
>>> Sorry for not testing earlier, and thanks for the separation. Just one
>>> menu layout thought: maybe the unit-root-related test menu entries in
>>> the Variable menu could be put in a new submenu? (I mean the new
>>> fractional integration entry, ADF, ADF-GLS, and KPSS.) Seems quite
>>> logical and natural to me.
>>
>> I agree.
>
> I'm OK with that. Suggestions for a string for the parent menu
> item?
How about "Unit root tests"?
OK, that's now in CVS. See what you think.
And while we're at it (GUI menus), I've renamed the item under the
Data menu that was "Select listed variables" to "Set selection
from list".
If you recall, the action is that you choose a named list (if any
are defined) and have that set the variables that are selected in
the main gretl window. Sven said (rightly) that the original label
was not very good. Is the new version an improvement? (IMO it's
very difficult to make this clear in a short string unless the
user already has some notion of what a "list" is; but hopefully
the context helps, in that the previous menu item is "Define or
edit list".)
Allin.
9:18 a.m.
New subject: [Gretl-users] pergm and fractint
On Sat, 3 Jul 2010, Allin Cottrell wrote:
> How about "Unit root tests"?
OK, that's now in CVS. See what you think.
I like it.
And while we're at it (GUI menus), I've renamed the item
under the
Data menu that was "Select listed variables" to "Set selection
from list".
If you recall, the action is that you choose a named list (if any
are defined) and have that set the variables that are selected in
the main gretl window. Sven said (rightly) that the original label
was not very good. Is the new version an improvement? (IMO it's
very difficult to make this clear in a short string unless the
user already has some notion of what a "list" is; but hopefully
the context helps, in that the previous menu item is "Define or
edit list".)
IMO, it's an improvement. And by the way, those users who are fiercely
opposed to reading the manual may learn what a list is out of curiosity.
While we're on the subject: I've been thinking for a while about a way to
use pre-defined lists in dialogs like OLS or tsls. It'd be very handy to
have a GUI way to tell gretl "remember all those variables I put in list
Z? Those are the instruments". But I couldn't picture myself anything that
wasn't too disruptive compared to the present design.
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
5286
days inactive
5287
days old
6 comments
4 participants
participants (4)
-
Allin Cottrell -
Carlos Henrique Coêlho de Andrade -
Riccardo (Jack) Lucchetti -
Sven Schreiber