3:52 p.m.
Not much is happening recently on the gretl lists and trackers (which is
not a bad sign).
Are there any further features in development right now? If not, would
it be an option to make another release, after leaving some time for
translators? A couple of bugs have been fixed since the last release,
including at least one crash (panel-related). Or should some of the
feature requests be targeted?
thanks,
sven
2:16 a.m.
On Sun, 27 Jan 2008, Sven Schreiber wrote:
Not much is happening recently on the gretl lists and trackers
(which is not a bad sign).
Are there any further features in development right now? If not,
would it be an option to make another release, after leaving
some time for translators? A couple of bugs have been fixed
since the last release, including at least one crash
(panel-related). Or should some of the feature requests be
targeted?
I agree, absence of a flood of bug-reports and feature requests is
not a bad thing.
The Changelog shows 9 "official" bugs fixed since the last
release, though Jack, Stefano and I have found a fixed a bunch
more than that.
I also agree that a release in the near future would be good.
However, quite a lot has been going on "behind the scenes", and
some testing of what we have at present would be good.
Some things to look at:
* The new --gls option to the ADF test (with GUI counterpart
/Variable/ADF-GLS test.
* The script editor enhancements (indentation and all that).
* Incorporation of Jack's documentation of the functions available
in "genr" (this is in CVS, but a bit rough-and-ready, and is not
yet in the Windows snapshot, though it will be soon).
Allin.
4:50 p.m.
Allin Cottrell schrieb:
I also agree that a release in the near future would be good.
However, quite a lot has been going on "behind the scenes", and
some testing of what we have at present would be good.
ok; I've used the scripting editor stuff a bit already AFAICT, and
haven't come across any problems so far.
* Incorporation of Jack's documentation of the functions
available
in "genr" (this is in CVS, but a bit rough-and-ready, and is not
yet in the Windows snapshot, though it will be soon).
I still plan to contribute to that, but in the next two weeks I won't
have the time. :-(
-sven
4:41 p.m.
On Sun, 27 Jan 2008, Sven Schreiber wrote:
Not much is happening recently on the gretl lists and trackers (which
is not
a bad sign).
Are there any further features in development right now? If not, would it be
an option to make another release, after leaving some time for translators? A
couple of bugs have been fixed since the last release, including at least one
crash (panel-related). Or should some of the feature requests be targeted?
The single thing I personally believe we're most lacking is a richer
offering of models in the GARCH family. Futurdorko is now working on a
user-level implementation, which will eventually lead to a native
implementation, but the times are a bit uncertain as of now.
Plus, after some discussion with Ignacio in Bilbao, it seems worthwile to
add a convenient way to use the Kalman filter at the user level. Several
areas could benefit from this (user-written weird arma stuff, structural
models à la Harvey, DSGE modellers...). I've got some very preliminary
material; in the (highly unlikely) event I find some time to work a little
more on a proposal, this may show up in the next version.
On the panel front, I guess sooner or later we'll have to find some time
for implementing Blundell & Bond's System-GMM estimator for dynamic linear
models, but, as usual, time is the most binding constraint.
I guess that progress will be much faster if/when the number of people
submitting code rises. Of course, ideally it should be C code, but I do
realise this is difficult. However, gretl's scripting language is IMO now
powerful enough to write non-trivial algorithms, and those would obviously
be most welcome too. I put at
http://www.econ.unian.it/lucchetti/gretl/Bilbao_slides.tgz
the presentations I did in Bilbao a couple of weeks ago. These cover
several aspects that people may find interesting. Comments are of course
welcome, especially on their suitability for a larger audience than
gretl-devel.
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
4:59 p.m.
Riccardo (Jack) Lucchetti schrieb:
Plus, after some discussion with Ignacio in Bilbao, it seems worthwile
to add a convenient way to use the Kalman filter at the user level.
Several areas could benefit from this (user-written weird arma stuff,
structural models à la Harvey, DSGE modellers...). I've got some very
preliminary material; in the (highly unlikely) event I find some time to
work a little more on a proposal, this may show up in the next version.
Ah yes, the Kalman filter would be most welcome and some sort of killer
feature actually! I would even go so far as to say that this should have
priority among the econometric features, because it is so fundamental
and versatile. Most of the other requested functionality could be
implemented via user functions and such, but the Kalman filter should be
native. And after that even more stuff can be done in user functions...
I guess that progress will be much faster if/when the number of people
submitting code rises. Of course, ideally it should be C code, but I do
realise this is difficult. However, gretl's scripting language is IMO
now powerful enough to write non-trivial algorithms, and those would
obviously be most welcome too. I put at
http://www.econ.unian.it/lucchetti/gretl/Bilbao_slides.tgz
the presentations I did in Bilbao a couple of weeks ago.
[The link doesn't work for me.] I totally agree on the gretl-scripting
front. I seem to be semi-consciously moving over from Python/Numpy to
gretl script for most things actually, because it is so nice. Eventually
I will contribute some Vecm functions as successors to the py4gretl
functions, but that still takes some time.
cheers,
sven
5:15 p.m.
Sven Schreiber schrieb:
Riccardo (Jack) Lucchetti schrieb:
>
> Plus, after some discussion with Ignacio in Bilbao, it seems worthwile
> to add a convenient way to use the Kalman filter at the user level.
> Several areas could benefit from this (user-written weird arma stuff,
> structural models à la Harvey, DSGE modellers...). I've got some very
> preliminary material; in the (highly unlikely) event I find some time
> to work a little more on a proposal, this may show up in the next
> version.
>
Ah yes, the Kalman filter would be most welcome and some sort of killer
feature actually! I would even go so far as to say that this should have
priority among the econometric features, because it is so fundamental
and versatile. Most of the other requested functionality could be
implemented via user functions and such, but the Kalman filter should be
native. And after that even more stuff can be done in user functions...
Forgot to add the following, therefore replying to self... I don't think
this type of fundamental feature should be stopping a release though.
The feature is ready when it is ready, and IMHO there's no point in
holding one's breath for it.
just my 2ct
-sven
5:32 p.m.
On Mon, 28 Jan 2008, Sven Schreiber wrote:
> Ah yes, the Kalman filter would be most welcome and some sort of
killer
> feature actually! I would even go so far as to say that this should have
> priority among the econometric features, because it is so fundamental and
> versatile. Most of the other requested functionality could be implemented
> via user functions and such, but the Kalman filter should be native. And
> after that even more stuff can be done in user functions...
>
Forgot to add the following, therefore replying to self... I don't think this
type of fundamental feature should be stopping a release though. The feature
is ready when it is ready, and IMHO there's no point in holding one's breath
for it.
Ok. If I don't fall asleep tonight, I'll post an RFC to the list.
Ps: It's strange the link doesn't work for you. I just checked it and it
seems to be there alright. Anyone with the same problem? Sven, if the
problem persists, I'll send you the tarball privately.
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
5:42 p.m.
Riccardo (Jack) Lucchetti schrieb:
On Mon, 28 Jan 2008, Sven Schreiber wrote:
>> Ah yes, the Kalman filter would be most welcome and some sort of
>> killer feature actually! I would even go so far as to say that this
>> should have priority among the econometric features, because it is so
>> fundamental and versatile. Most of the other requested functionality
>> could be implemented via user functions and such, but the Kalman
>> filter should be native. And after that even more stuff can be done
>> in user functions...
>>
>
> Forgot to add the following, therefore replying to self... I don't
> think this type of fundamental feature should be stopping a release
> though. The feature is ready when it is ready, and IMHO there's no
> point in holding one's breath for it.
Ok. If I don't fall asleep tonight, I'll post an RFC to the list.
Ps: It's strange the link doesn't work for you. I just checked it and it
seems to be there alright. Anyone with the same problem? Sven, if the
problem persists, I'll send you the tarball privately.
The server www.econ.unian.it cannot be found.
Since you're asking the list, I took the liberty to add it to the
recipients... ;-)
-sven
5:44 p.m.
Sven Schreiber schrieb:
Riccardo (Jack) Lucchetti schrieb:
>
> Ps: It's strange the link doesn't work for you. I just checked it and
> it seems to be there alright. Anyone with the same problem? Sven, if
> the problem persists, I'll send you the tarball privately.
The server www.econ.unian.it cannot be found.
fyi, the link
http://www.econ.univpm.it/lucchetti/gretl/Bilbao_slides.tgz
works.
-sven
5:48 p.m.
On Mon, 28 Jan 2008, Sven Schreiber wrote:
Sven Schreiber schrieb:
> Riccardo (Jack) Lucchetti schrieb:
>>
>> Ps: It's strange the link doesn't work for you. I just checked it and it
>> seems to be there alright. Anyone with the same problem? Sven, if the
>> problem persists, I'll send you the tarball privately.
>
> The server www.econ.unian.it cannot be found.
>
fyi, the link
http://www.econ.univpm.it/lucchetti/gretl/Bilbao_slides.tgz
DOH!!
The stupid name change of my university! (We used to be "Università di
Ancona" once --- a much nicer name, if you ask me.)
Sorry, folks!
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
7:08 p.m.
On Mon, 28 Jan 2008, Riccardo (Jack) Lucchetti wrote:
Ps: It's strange the link doesn't work for you. I just
checked
it and it seems to be there alright. Anyone with the same
problem? Sven, if the problem persists, I'll send you the
tarball privately.
FYI, from the "outside world" --
www.unian.it redirects to www.univpm.it
but
www.econ.unian.it does not redirect to www.econ.univpm.it
(or anywhere)
I guess "www.econ.unian.it" works for Jack only because he's
using the univpm nameserver.
Allin.
9:22 p.m.
Actually that link works fine for me, but the downloaded archive is damaged
and i can't open it correctly..nice!
8:28 a.m.
El Monday 28 January 2008 20:08:38 Allin Cottrell escribió:
On Mon, 28 Jan 2008, Riccardo (Jack) Lucchetti wrote:
> Ps: It's strange the link doesn't work for you. I just checked
> it and it seems to be there alright. Anyone with the same
> problem? Sven, if the problem persists, I'll send you the
> tarball privately.
FYI, from the "outside world" --
www.unian.it redirects to www.univpm.it
but
www.econ.unian.it does not redirect to www.econ.univpm.it
(or anywhere)
I guess "www.econ.unian.it" works for Jack only because he's
using the univpm nameserver.
The funny thing is that I have no problem with the link that Jack sent,
www.econ.unian.it works perfectly for me.
--
Ignacio Diaz-Emparanza
DEPARTAMENTO DE ECONOMÍA APLICADA III (ECONOMETRÍA Y ESTADÍSTICA)
UPV/EHU
Avda. Lehendakari Aguirre, 83 | 48015 BILBAO
T.: +34 946013732 | F.: +34 946013754
www.et.bs.ehu.es
2:08 p.m.
On Tue, 29 Jan 2008, Ignacio Diaz-Emparanza wrote:
> I guess "www.econ.unian.it" works for Jack only
because he's
> using the univpm nameserver.
The funny thing is that I have no problem with the link that Jack sent,
www.econ.unian.it works perfectly for me.
Hmm, strange. It works for me too, today. It wasn't found
yesterday.
Allin.
11:58 p.m.
On Mon, 28 Jan 2008, Riccardo (Jack) Lucchetti wrote:
Ok. If I don't fall asleep tonight, I'll post an RFC to the
list.
Here you are, guys. Let me know what you think.
PS: Ignacio, this is a further elaboration of the document we started
writing together -- a few things have changed.
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
6489
days inactive
6491
days old
14 comments
5 participants
participants (5)
-
Allin Cottrell -
Ignacio Diaz-Emparanza -
Riccardo (Jack) Lucchetti -
Stefano Balietti -
Sven Schreiber