From r.lucchetti@univpm.it Thu May  1 06:54:03 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] RFC: $sigma & co.
Date: Thu, 01 May 2008 12:54:00 +0200
Message-ID: <alpine.DEB.1.10.0805011246010.21301@ec-4.econ.univpm.it>
In-Reply-To: alpine.DEB.1.10.0804302307020.15440@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5860860890420787430=="

--===============5860860890420787430==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

On Wed, 30 Apr 2008, Riccardo (Jack) Lucchetti wrote:

> On Wed, 30 Apr 2008, Allin Cottrell wrote:
>
>> On Mon, 28 Apr 2008, Sven Schreiber wrote:
>>=20
>> [re, reorganizing the $sigma and $vcv accessors]
>>=20
>>> Although I will be one of the (few?) affected users I agree with
>>> Jack that $vcv should be changed, especially because it's very
>>> simple to change $vcv to $sigma to get the scripts working
>>> again. (And I like the functionality that $vcv will deliver in
>>> VAR/VECM contexts.)
>>=20
>> OK.
>
> Coding $vcv for VARs should be quite easy. I'll try to see to that tomorrow=
.=20
> Things are a bit trickier for VECMs. Should we return the v-cov matrix of t=
he=20
> coefficients in VECM or VAR form?

I just committed to CVS a few modifications to the VAR code, so that $vcv=20
returns the estimated covariance matrix for the coefficients. I don't=20
think I've broken anything, but you never know, so please test!=20
Limitations:

1) the --robust switch is ignored
2) VECMs aren't handled

Moreover, we have a slight inconsistency in the way we compute things:=20
$sigma uses the asymptotic formula (ie, E'E over T), while in the=20
displayed equations we use degrees-of-freedom corrected figures for=20
standard errors. I'm not overly bothered by this, but some may. Example=20
script:

<script>
open denmark
list Y =3D LRM LRY
scalar order =3D 2
scalar n =3D nelem(Y)*order+1
var order Y

b =3D vec($coeff)
vcv =3D $vcv
scalar correction =3D ($nobs-order)/($nobs-order-n)

#reproduce displayed stderrs
se =3D sqrt(diag(vcv)*correction)

zstat =3D b./se

eval b~se~zstat
</script>

Comments welcome.

Riccardo (Jack) Lucchetti
Dipartimento di Economia
Universit=C3=A0 Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============5860860890420787430==--


From svetosch@gmx.net Thu May  1 07:24:54 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] RFC: $sigma & co.
Date: Thu, 01 May 2008 13:24:44 +0200
Message-ID: <4819A87C.7010307@gmx.net>
In-Reply-To: alpine.DEB.1.10.0804302307020.15440@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4357665850257215501=="

--===============4357665850257215501==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Am 30.04.2008 23:16, Riccardo (Jack) Lucchetti schrieb:
> On Wed, 30 Apr 2008, Allin Cottrell wrote:
> 
>> On Mon, 28 Apr 2008, Sven Schreiber wrote:
>>
>> [re, reorganizing the $sigma and $vcv accessors]
>>
>>> Although I will be one of the (few?) affected users I agree with
>>> Jack that $vcv should be changed, especially because it's very
>>> simple to change $vcv to $sigma to get the scripts working
>>> again. (And I like the functionality that $vcv will deliver in
>>> VAR/VECM contexts.)
>>
>> OK.
> 
> Coding $vcv for VARs should be quite easy. I'll try to see to that 
> tomorrow. Things are a bit trickier for VECMs. Should we return the 
> v-cov matrix of the coefficients in VECM or VAR form?
> 

I think the VECM form is more natural for the user if she specified a 
VECM. Also the vcv for the levels-VAR form would be singular, no? (I 
guess you don't care too much about that, but still.)

-sven

--===============4357665850257215501==--


From svetosch@gmx.net Thu May  1 07:26:10 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] RFC: $sigma & co.
Date: Thu, 01 May 2008 13:26:07 +0200
Message-ID: <4819A8CF.9070804@gmx.net>
In-Reply-To: alpine.DEB.1.10.0805011246010.21301@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0250351271817204615=="

--===============0250351271817204615==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Am 01.05.2008 12:54, Riccardo (Jack) Lucchetti schrieb:

> 
> Moreover, we have a slight inconsistency in the way we compute things: 
> $sigma uses the asymptotic formula (ie, E'E over T), while in the 
> displayed equations we use degrees-of-freedom corrected figures for 
> standard errors. I'm not overly bothered by this, but some may. 

What about $sigma in other contexts, is that asymptotic or 
dof-corrected? I think that should be made consistent.

-sven


--===============0250351271817204615==--


From svetosch@gmx.net Thu May  1 07:35:38 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] RFC: $sigma & co.
Date: Thu, 01 May 2008 13:35:35 +0200
Message-ID: <4819AB07.8080901@gmx.net>
In-Reply-To: alpine.DEB.1.10.0804302307020.15440@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0771468806567440565=="

--===============0771468806567440565==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Am 30.04.2008 23:16, Riccardo (Jack) Lucchetti schrieb:
> On Wed, 30 Apr 2008, Allin Cottrell wrote:

> 
>> I think it would be easier to roll such tests into the existing
>> gretl regression suite.  I can make this available via the web.
>> Package maintainers could play a useful role, however, by
>> contributing scripts that exercise their packages.
> 
> I agree. IMHO, writing test scripts for contributed function packages is 
> be something the package writers should do. How am I supposed to write a 
> test script for econometric techniques I may know nothing about?
> 

I agree it's ideal if the package authors also provide test cases, but 
simply calling a function doesn't require that you understand the 
results. I'd say that most of the time a backwards-incompatible change 
will give a plain error instead of wrong results. (Not always, of course.)

The more widespread gretl wants to become, the more stable its 
programming interface needs to be, IMHO. For every user who openly 
complains about a problem she has with gretl (on the lists or wherever), 
there are many many more users who simply discard it silently and move on.

be user-friendly, that's all I'm trying to say :-)

-sven

--===============0771468806567440565==--


From r.lucchetti@univpm.it Thu May  1 08:03:30 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] RFC: $sigma & co.
Date: Thu, 01 May 2008 14:03:27 +0200
Message-ID: <alpine.DEB.1.10.0805011354520.22637@ec-4.econ.univpm.it>
In-Reply-To: 4819A8CF.9070804@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============7646737402284424434=="

--===============7646737402284424434==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

On Thu, 1 May 2008, Sven Schreiber wrote:

> Am 01.05.2008 12:54, Riccardo (Jack) Lucchetti schrieb:
>
>>=20
>> Moreover, we have a slight inconsistency in the way we compute things:=20
>> $sigma uses the asymptotic formula (ie, E'E over T), while in the displaye=
d=20
>> equations we use degrees-of-freedom corrected figures for standard errors.=
=20
>> I'm not overly bothered by this, but some may.=20
>
> What about $sigma in other contexts, is that asymptotic or dof-corrected? I=
=20
> think that should be made consistent.

I'm all for consistency in general, but this case is difficult. In OLS=20
estimation, for example, SSR/T and SSR/(T-k) are both consistent, but the=20
latter has the additional small advantage of being unbiased under certain=20
conditions, so that's traditionally what people use. I personally=20
think it's rather silly to worry about this when you have a decent=20
sample size; if you don't, well, I doubt very much you should be doing=20
inference _at all_.

In other contexts, Tobit models for example, you simply don't have a=20
choice. In my view, the really important thing is that you know which=20
formula is used in each case, so that if you feel like re-computing $sigma=20
to your taste, you have the tools to do it.


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Universit=C3=A0 Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============7646737402284424434==--


From svetosch@gmx.net Thu May  1 11:11:37 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] RFC: $sigma & co.
Date: Thu, 01 May 2008 17:11:31 +0200
Message-ID: <4819DDA3.5020301@gmx.net>
In-Reply-To: alpine.DEB.1.10.0805011354520.22637@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4331759307040135707=="

--===============4331759307040135707==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Am 01.05.2008 14:03, Riccardo (Jack) Lucchetti schrieb:
> On Thu, 1 May 2008, Sven Schreiber wrote:
> 
>> Am 01.05.2008 12:54, Riccardo (Jack) Lucchetti schrieb:
>>
>>>
>>> Moreover, we have a slight inconsistency in the way we compute 
>>> things: $sigma uses the asymptotic formula (ie, E'E over T), while in 
>>> the displayed equations we use degrees-of-freedom corrected figures 
>>> for standard errors. I'm not overly bothered by this, but some may. 
>>
>> What about $sigma in other contexts, is that asymptotic or 
>> dof-corrected? I think that should be made consistent.
> 
> I'm all for consistency in general, but this case is difficult. In OLS 
> estimation, for example, SSR/T and SSR/(T-k) are both consistent, but 
> the latter has the additional small advantage of being unbiased under 
> certain conditions, so that's traditionally what people use. I 
> personally think it's rather silly to worry about this when you have a 
> decent sample size; if you don't, well, I doubt very much you should be 
> doing inference _at all_.
> 
> In other contexts, Tobit models for example, you simply don't have a 
> choice. In my view, the really important thing is that you know which 
> formula is used in each case, so that if you feel like re-computing 
> $sigma to your taste, you have the tools to do it.
> 

I just meant "consistent definition" across the various contexts of 
$sigma, not consistent in the statistical sense. So if you are saying 
$sigma for Tobit models is w/o dof correction out of necessity but for 
OLS it currently is dof-corrected, then I would suggest the rule: 
"$sigma has a dof correction if at all feasible". That would mean dof 
correction for the VAR/VECM $sigma. BTW, I remember we had a discussion 
in the context of the vcv of the betas in the VECM, but I don't remember 
the result. Is there a dof correction there in the end? (I know I know 
it's probably in the manual...)

cheers,
sven

--===============4331759307040135707==--


From r.lucchetti@univpm.it Thu May  1 13:00:33 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] RFC: $sigma & co.
Date: Thu, 01 May 2008 19:00:29 +0200
Message-ID: <alpine.DEB.1.10.0805011850180.22885@ec-4.econ.univpm.it>
In-Reply-To: 4819DDA3.5020301@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5887088266033045974=="

--===============5887088266033045974==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

On Thu, 1 May 2008, Sven Schreiber wrote:

> Am 01.05.2008 14:03, Riccardo (Jack) Lucchetti schrieb:
>> On Thu, 1 May 2008, Sven Schreiber wrote:
>>=20
>>> Am 01.05.2008 12:54, Riccardo (Jack) Lucchetti schrieb:
>>>=20
>>>>=20
>>>> Moreover, we have a slight inconsistency in the way we compute things:=20
>>>> $sigma uses the asymptotic formula (ie, E'E over T), while in the=20
>>>> displayed equations we use degrees-of-freedom corrected figures for=20
>>>> standard errors. I'm not overly bothered by this, but some may.=20
>>>=20
>>> What about $sigma in other contexts, is that asymptotic or dof-corrected?=
=20
>>> I think that should be made consistent.
>>=20
>> I'm all for consistency in general, but this case is difficult. In OLS=20
>> estimation, for example, SSR/T and SSR/(T-k) are both consistent, but the =

>> latter has the additional small advantage of being unbiased under certain =

>> conditions, so that's traditionally what people use. I personally think=20
>> it's rather silly to worry about this when you have a decent sample size; =

>> if you don't, well, I doubt very much you should be doing inference _at=20
>> all_.
>>=20
>> In other contexts, Tobit models for example, you simply don't have a=20
>> choice. In my view, the really important thing is that you know which=20
>> formula is used in each case, so that if you feel like re-computing $sigma=
=20
>> to your taste, you have the tools to do it.
>>=20
>
> I just meant "consistent definition" across the various contexts of $sigma,=
=20
> not consistent in the statistical sense. So if you are saying $sigma for=20
> Tobit models is w/o dof correction out of necessity but for OLS it currentl=
y=20
> is dof-corrected, then I would suggest the rule: "$sigma has a dof correcti=
on=20
> if at all feasible". That would mean dof correction for the VAR/VECM $sigma=
.=20
> BTW, I remember we had a discussion in the context of the vcv of the betas =
in=20
> the VECM, but I don't remember the result. Is there a dof correction there =
in=20
> the end? (I know I know it's probably in the manual...)

Sorry, my answer wan't clear. The meaning I gave to the word "consistency"=20
is exactly the same as yours. My point is, in some contexts you use dof=20
correction because, well, that's what people do and in some other contexts=20
you don't because... people don't!

If this does not sound particularly scientific, too bad. Mrs McCloskey and=20
Mr Feyerabend (among others) wouldn't find it shocking, though.

Riccardo (Jack) Lucchetti
Dipartimento di Economia
Universit=C3=A0 Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============5887088266033045974==--


From svetosch@gmx.net Fri May  2 07:40:10 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] new tracker for incompatible changes
Date: Fri, 02 May 2008 13:40:03 +0200
Message-ID: <481AFD93.70900@gmx.net>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============9013092594502537075=="

--===============9013092594502537075==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Dear all,

as we discussed recently I have created a new sourceforge tracker named 
"Backwards-incompatible changes", and have added a first item relating 
to the changed use of the $vcv accessors. Please check it out, correct 
it and/or expand it. (Allin and Jack, could you just drop some keywords 
what other incompatible changes will be new in 1.7.5, if any. thanks)

I have designated the "group" field of the tracker to refer to different 
gretl versions, so that later it will be possible to search for all 
incompatible changes in a specific version. Currently there is only a 
group "new in 1.7.5", but as time progresses there will of course be 
more groups. (And there is a bogus group "changed in gretl version" that 
I erroneously created and now can't get rid of anymore.)

This tracker is already visible on the sourceforge tracker page. I'm not 
sure if regular users are able to submit new items (which they normally 
shouldn't do), but at least you need to be logged in before doing that, 
and submitted items can always be deleted later.

Before I announce this on the users list I would like to wait a bit for 
feedback, however.

cheers,
sven

--===============9013092594502537075==--


From cottrell@wfu.edu Fri May  2 10:19:22 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] new tracker for incompatible changes
Date: Fri, 02 May 2008 10:19:13 -0400
Message-ID: <alpine.LRH.1.10.0805021009060.6977@ricardo.ecn.wfu.edu>
In-Reply-To: 481AFD93.70900@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4228288176932691253=="

--===============4228288176932691253==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 2 May 2008, Sven Schreiber wrote:

> as we discussed recently I have created a new sourceforge 
> tracker named "Backwards-incompatible changes", and have added a 
> first item relating to the changed use of the $vcv accessors. 
> Please check it out, correct it and/or expand it...

Thanks very much, Sven.  But now I see the new tracker in place I 
have to say, I wonder if this is the right way to go.

It's certainly a good thing to have a searchable database of such 
changes, but IMO this should be (a) static and (b) editable only 
by developers (including yourself, of course).  The regular SF 
tracker apparatus (Assigned to/Status/Priority, and so on) doesn't 
seem to apply, and might be confusing.

However, that's just my first reaction and I'm willing to be 
persuaded otherwise.

Allin.





--===============4228288176932691253==--


From svetosch@gmx.net Fri May  2 10:38:59 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] new tracker for incompatible changes
Date: Fri, 02 May 2008 16:38:59 +0200
Message-ID: <481B2783.7080002@gmx.net>
In-Reply-To: alpine.LRH.1.10.0805021009060.6977@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2101749068397148025=="

--===============2101749068397148025==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Am 02.05.2008 16:19, Allin Cottrell schrieb:
> On Fri, 2 May 2008, Sven Schreiber wrote:
> 
>> as we discussed recently I have created a new sourceforge 
>> tracker named "Backwards-incompatible changes", and have added a 
>> first item relating to the changed use of the $vcv accessors. 
>> Please check it out, correct it and/or expand it...
> 
> Thanks very much, Sven.  But now I see the new tracker in place I 
> have to say, I wonder if this is the right way to go.
> 
> It's certainly a good thing to have a searchable database of such 
> changes, but IMO this should be (a) static and (b) editable only 
> by developers (including yourself, of course).  The regular SF 
> tracker apparatus (Assigned to/Status/Priority, and so on) doesn't 
> seem to apply, and might be confusing.

quickly @b: I'm not sure right who may edit it. I have set some 
permissions, where the six gretl developers I could identify in the 
sourceforge list are made admins of that new tracker. I'm not sure what 
that means for regular (logged-in) users, it would be nice if someone 
who's not in the sourceforge gretl developers list could test.

-sven


--===============2101749068397148025==--


From pyz@brama.com Fri May  2 19:04:04 2008
From: Max Pyziur <pyz@brama.com>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] Problems building and installing on Fedora 8
Date: Fri, 02 May 2008 19:01:25 -0400
Message-ID: <Pine.LNX.4.60.0805021837260.30928@brama.com>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============7578722933411227746=="

--===============7578722933411227746==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


Greetings,

I've been trying to get gretl running on my Fedora 8 Core 2 machine.

The pre-compile rpm fails with:
error: Failed dependencies:
         libreadline.so.4 is needed by gretl-1.7.4-1gtk2.i586

even though readline is installed:
#  rpm -qa | grep readline
readline-devel-5.2-10.fc8
readline-5.2-10.fc8
readline-5.2-10.fc8
readline-devel-5.2-10.fc8

I've created a symlink to libreadline.so.4
  libreadline.so.4 -> libreadline.so.5.2
but that doesn't move the process forward.

I've tried compiling an rpm from source by taking 
gretl-1.7.4/redhat/gretl.spec.in and modifying it and placing it in 
/usr/src/redhat/SPECS, and the binary into /usr/src/redhat/SOURCES and 
issueng the following command:
rpmbuild -bb --clean gretl.spec (when in the /usr/src/redhat/SPECS 
directory)

I get the following two sets of errors:
1)
checking for LAPACK... no
*** Could not run LAPACK test program, checking why...
*** The test program failed to compile or link. See config.log for the
*** exact error that occured. This may mean LAPACK was incorrectly 
installed
*** or that you have moved LAPACK since it was installed.


LAPACK is installed:
# rpm -qa | grep lapack
lapack-devel-3.1.1-2.fc8
lapack-devel-3.1.1-2.fc8
lapack-3.1.1-2.fc8
lapack-3.1.1-2.fc8

2) The rpmbuild fails with the following:
mkdir .libs
gcc -o .libs/gretlcli gretlcli.o complete.o  ../lib/.libs/libgretl-1.0.so 
-ldl -L/usr/local/lib -lz -lxml2 -lglib-2.0 -lgmp -lfftw3 -lm -lreadline 
-ltermcap
../lib/.libs/libgretl-1.0.so: undefined reference to `dtrcon_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dgels_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dspsv_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dpptri_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dpptrf_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dsycon_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dpocon_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dsyev_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dgeev_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dtrtri_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dgetrf_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dgelss_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dorgqr_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dsytri_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dpotrf_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dsytrf_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dpotri_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dgecon_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dgeqrf_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dpotrs_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dgesvd_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dgetrs_'
../lib/.libs/libgretl-1.0.so: undefined reference to `dgetri_'
collect2: ld returned 1 exit status
make[1]: *** [gretlcli] Error 1
make[1]: Leaving directory `/usr/src/redhat/BUILD/gretl-1.7.4/cli'
make: *** [cli] Error 2
error: Bad exit status from /var/tmp/rpm-tmp.19815 (%build)


RPM build errors:
     Bad exit status from /var/tmp/rpm-tmp.19815 (%build)


It seems that these errors have been encountered before 
(http://aur.archlinux.org/packages.php?ID=13752), but there is no solution 
offered.


The contents of my gretl.spec file is appended below. Any help is 
appreciated. Apologies if this has been covered elsewhere; if there are 
links showing information as to how to get through this, that would be 
great.

Max Pyziur
pyz(a)brama.com

######################
# cat gretl.spec

%define name    gretl
%define ver 1.7.4
%define rel     mp
%define prefix /usr

Summary: econometrics package
Name: %{name}
Version: %{ver}
Release: %{rel}
License: GPL
Group: Applications/Math
Source: ftp://ricardo.ecn.wfu.edu/pub/gretl/gretl-%{ver}.tar.bz2
Buildroot: %{_tmppath}/%{name}-%{ver}-%{rel}-root
Packager: Allin Cottrell <cottrell(a)wfu.edu>
URL: http://gretl.sourceforge.net/
Vendor: Allin Cottrell <cottrell(a)wfu.edu>
Docdir: %{prefix}/share/doc
Requires: gtk+ >= 1.2.3

%description

gretl is a free econometrics package. It comprises a shared library,
a command-line client and a gui client that uses GTK.  gretl offers
several least-squares based estimators.  Besides reading data files
in its own format it also reads RATS 4 databases.  It has a built-in
spreadsheet for editing data, and uses gnuplot for graphing.  It can
output regression results in LaTeX format.

%prep
%setup -q

%build
%configure
make

%install
rm -fr %{buildroot}
%makeinstall

%post
/sbin/ldconfig

%clean
rm -fr %{buildroot}

%files
%doc COPYING README INSTALL
%{prefix}/lib/lib*
%{prefix}/lib/gretl*/*
%{prefix}/bin/gretlcli
%{prefix}/bin/gretl_x11
%{prefix}/bin/gretl
%{prefix}/share/gretl/*
%{prefix}/share/locale/*
%{prefix}/share/man/*


--===============7578722933411227746==--


From cottrell@wfu.edu Fri May  2 19:14:45 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Problems building and installing on Fedora 8
Date: Fri, 02 May 2008 19:14:36 -0400
Message-ID: <alpine.LRH.1.10.0805021908260.7821@ricardo.ecn.wfu.edu>
In-Reply-To: Pine.LNX.4.60.0805021837260.30928@brama.com
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============8675082569507280180=="

--===============8675082569507280180==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 2 May 2008, Max Pyziur wrote:

> I've been trying to get gretl running on my Fedora 8 Core 2 machine.
> 
> The pre-compile rpm fails with:
> error: Failed dependencies:
>         libreadline.so.4 is needed by gretl-1.7.4-1gtk2.i586

The only RHEL system I have available to me, for building the 
gretl rpm, is a fairly conservative one that uses readline 4.0.  
On Fedora 8, my recommendation would be to ignore the rpm and 
build from the current CVS source.  

http://sourceforge.net/cvs/?group_id=36234

Allin Cottrell.

--===============8675082569507280180==--


From pyz@brama.com Sat May  3 14:32:04 2008
From: Max Pyziur <pyz@brama.com>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Problems building and installing on Fedora 8
Date: Sat, 03 May 2008 14:29:40 -0400
Message-ID: <Pine.LNX.4.60.0805031410580.30839@brama.com>
In-Reply-To: alpine.LRH.1.10.0805021908260.7821@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============1822517895795258154=="

--===============1822517895795258154==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 2 May 2008, Allin Cottrell wrote:

> On Fri, 2 May 2008, Max Pyziur wrote:
>
>> I've been trying to get gretl running on my Fedora 8 Core 2 machine.
>>
>> The pre-compile rpm fails with:
>> error: Failed dependencies:
>>         libreadline.so.4 is needed by gretl-1.7.4-1gtk2.i586
>
> The only RHEL system I have available to me, for building the
> gretl rpm, is a fairly conservative one that uses readline 4.0.
> On Fedora 8, my recommendation would be to ignore the rpm and
> build from the current CVS source.
>
> http://sourceforge.net/cvs/?group_id=36234
>
> Allin Cottrell.

Much thanks for your reply.

I have an Fedora Core 2 machine which I use as a home server. Your latest 
rpm build of gretl does install on that machine. However, I still can't 
build it on that machine.

I have installed gretl-1.7.4-1gtk2.i586.rpm on my Core 2 using the 
--nodeps flag. As yet, I haven't tried using it, other than bringing up 
the initial interface.

I have tried setting the PKG_CONFIG_PATH variable both on the FC2 as 
well as the FC8 machine in the hopes of getting gretl to build. No luck.

Any further guidance would be appreciated.

Thanks!

Max Pyziur
pyz(a)brama.com

--===============1822517895795258154==--


From cottrell@wfu.edu Sat May  3 15:25:36 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Problems building and installing on Fedora 8
Date: Sat, 03 May 2008 15:25:25 -0400
Message-ID: <alpine.LRH.1.10.0805031509560.26145@ricardo.ecn.wfu.edu>
In-Reply-To: Pine.LNX.4.60.0805031410580.30839@brama.com
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5874224501980241026=="

--===============5874224501980241026==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sat, 3 May 2008, Max Pyziur wrote:

> I have an Fedora Core 2 machine which I use as a home server. 
> Your latest rpm build of gretl does install on that machine. 
> However, I still can't build it on that machine.
> 
> I have installed gretl-1.7.4-1gtk2.i586.rpm on my Core 2 using 
> the --nodeps flag. As yet, I haven't tried using it, other than 
> bringing up the initial interface.
> 
> I have tried setting the PKG_CONFIG_PATH variable both on the 
> FC2 as well as the FC8 machine in the hopes of getting gretl to 
> build. No luck.

I'll go back to your previous message...

"I get the following two sets of errors:
1) checking for LAPACK... no 
*** Could not run LAPACK test program, checking why... 
*** The test program failed to compile or link. See config.log for the 
*** exact error that occured. This may mean LAPACK was incorrectly installed 
*** or that you have moved LAPACK since it was installed."

Well, like it says, take a look at config.log.  Lapack 3.1.1 is 
installed, according to rpm -qa, but maybe libblas is missing, or 
maybe none of libf2c/libg2c/libgfortran can be found? liblapack is 
not usable without libblas, and the lapack+blas combination in 
turn is useless without a basic fortran library.  On recent Fedora 
I'd expect that to be libgfortran.so; on older systems it was 
libg2c.so, and on even older systems, libf2c.so.

"2) The rpmbuild fails with the following:
mkdir .libs
gcc -o .libs/gretlcli gretlcli.o complete.o  
../lib/.libs/libgretl-1.0.so 
-ldl -L/usr/local/lib -lz -lxml2 -lglib-2.0 -lgmp -lfftw3 -lm 
-lreadline -ltermcap
../lib/.libs/libgretl-1.0.so: undefined reference to `dtrcon_'"

That's just the same error: dtrcon_ is defined by liblapack.

On my system:

waverley:~$ /bin/ls -1 /usr/lib/liblapack*
/usr/lib/liblapack.so
/usr/lib/liblapack.so.3
/usr/lib/liblapack.so.3.1
/usr/lib/liblapack.so.3.1.1
waverley:~$ /bin/ls -1 /usr/lib/libblas*
/usr/lib/libblas.so
/usr/lib/libblas.so.3
/usr/lib/libblas.so.3.1
/usr/lib/libblas.so.3.1.1
waverley:~$ locate libgfortran.so
/opt/gcc4/lib/libgfortran.so
/opt/gcc4/lib/libgfortran.so.3
/opt/gcc4/lib/libgfortran.so.3.0.0

Allin Cottrell





--===============5874224501980241026==--


From pyz@brama.com Sat May  3 16:46:37 2008
From: Max Pyziur <pyz@brama.com>
To: gretl-devel@gretlml.univpm.it
Subject:
 Progress!,  was [Gretl-devel] Problems building and installing on Fedora 8
Date: Sat, 03 May 2008 16:43:50 -0400
Message-ID: <Pine.LNX.4.60.0805031638341.17017@brama.com>
In-Reply-To: alpine.LRH.1.10.0805031509560.26145@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============7422508592919858861=="

--===============7422508592919858861==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fedora 8, libblas is installed separately. However, reviewing your 
notes below I also installed blas-devel. This caused the 
rpmbuild/compilation to go farther. (It might be worthwhile to this 
dependency in the spec file).

The following message registered after 
the configuration (which didn't before blas-devel):
Configuration:

   Installation path:                      /usr
   Use readline library:                   yes
   Use gnuplot for graphs:                 yes
   Use PNG for gnuplot graphs:             yes
   Use LaTeX for typesetting output:       yes
   Gnu Multiple Precision support:         yes
   MPFR support:                           yes
   LAPACK support:                         yes
   FFTW3 support:                          yes
   Build with GTK version:                 2.0
   Script syntax highlighting:             yes
   Use installed gtksourceview:            no
   Use GTK printing apparatus:             yes
   Build with gnome support:               2.0
   Build gretl documentation:              no
   Use Lucida fonts:                       no
   Build message catalogs:                 yes
   Gnome installation prefix:              /usr
   X-12-ARIMA support:                     yes
   TRAMO/SEATS support:                    yes
   Experimental audio support:             no


Now type 'make' to build gretl.

+ make

...

but then it failed with the following:

gcc -o .libs/gretl_x11 about.o boxplots.o calculator.o callbacks.o 
clipboard.o cmdstack.o console.o database.o datafiles.o dialogs.o 
dlgutils.o filelists.o fileselect.o filters.o fncall.o fnsave.o 
graph_page.o gpt_control.o gpt_dialog.o gretl.o guiprint.o gui_recode.o 
gui_utils.o helpfiles.o library.o menustate.o model_table.o objectsave.o 
obsbutton.o selector.o series_view.o session.o settings.o ssheet.o 
textbuf.o textutil.o toolbar.o treeutils.o update.o gtkfontselhack.o 
-pthread -lgnomeui-2 -lSM -lICE -lbonoboui-2 -lgnomevfs-2 -lgnomecanvas-2 
-lgnome-2 -lpopt -lbonobo-2 -lbonobo-activation -lart_lgpl_2 -lgtk-x11-2.0 
-lgdk-x11-2.0 -latk-1.0 -lpangocairo-1.0 -lpango-1.0 -lcairo -lgconf-2 
-lORBit-2 -lgthread-2.0 -lrt 
-L/usr/src/redhat/BUILD/gretl-1.7.4/gui2/gtksourceview 
-lgtksourceview-lite 
-L/usr/src/redhat/BUILD/gretl-1.7.4/gui2/gtkextra-lite -lgtkextra-lite 
../lib/.libs/libgretl-1.0.so -L/usr/lib -llapack -lblas -lgfortran 
-L/usr/local/lib -lz -lxml2 -lgmp -lfftw3 -lm -lgdk_pixbuf-2.0 
-lgobject-2.0 -lgmodule-2.0 -ldl -lglib-2.0
/usr/bin/ld: skipping incompatible /usr/lib/libgnomeui-2.so when searching 
for -lgnomeui-2
/usr/bin/ld: skipping incompatible /usr/lib/libSM.so when searching for 
-lSM
/usr/bin/ld: skipping incompatible /usr/lib/libbonoboui-2.so when 
searching for -lbonoboui-2
/usr/bin/ld: skipping incompatible /usr/lib/libgnomevfs-2.so when 
searching for -lgnomevfs-2
/usr/bin/ld: skipping incompatible /usr/lib/libgnomecanvas-2.so when 
searching for -lgnomecanvas-2
/usr/bin/ld: skipping incompatible /usr/lib/libgnome-2.so when searching 
for -lgnome-2
/usr/bin/ld: skipping incompatible /usr/lib/libpopt.so when searching for 
-lpopt
/usr/bin/ld: skipping incompatible /usr/lib/libpopt.so when searching for 
-lpopt
/usr/bin/ld: cannot find -lpopt
collect2: ld returned 1 exit status
make[1]: *** [gretl_x11] Error 1
make[1]: Leaving directory `/usr/src/redhat/BUILD/gretl-1.7.4/gui2'
make: *** [gui2] Error 2
error: Bad exit status from /var/tmp/rpm-tmp.91719 (%build)


RPM build errors:
     Bad exit status from /var/tmp/rpm-tmp.91719 (%build)
...

How should this be pursued?

Much thanks for all of your help.

Max Pyziur
pyz(a)brama.com


On Sat, 3 May 2008, Allin Cottrell wrote:

> On Sat, 3 May 2008, Max Pyziur wrote:
>
>> I have an Fedora Core 2 machine which I use as a home server.
>> Your latest rpm build of gretl does install on that machine.
>> However, I still can't build it on that machine.
>>
>> I have installed gretl-1.7.4-1gtk2.i586.rpm on my Core 2 using
>> the --nodeps flag. As yet, I haven't tried using it, other than
>> bringing up the initial interface.
>>
>> I have tried setting the PKG_CONFIG_PATH variable both on the
>> FC2 as well as the FC8 machine in the hopes of getting gretl to
>> build. No luck.
>
> I'll go back to your previous message...
>
> "I get the following two sets of errors:
> 1) checking for LAPACK... no
> *** Could not run LAPACK test program, checking why...
> *** The test program failed to compile or link. See config.log for the
> *** exact error that occured. This may mean LAPACK was incorrectly installed
> *** or that you have moved LAPACK since it was installed."
>
> Well, like it says, take a look at config.log.  Lapack 3.1.1 is
> installed, according to rpm -qa, but maybe libblas is missing, or
> maybe none of libf2c/libg2c/libgfortran can be found? liblapack is
> not usable without libblas, and the lapack+blas combination in
> turn is useless without a basic fortran library.  On recent Fedora
> I'd expect that to be libgfortran.so; on older systems it was
> libg2c.so, and on even older systems, libf2c.so.
>
> "2) The rpmbuild fails with the following:
> mkdir .libs
> gcc -o .libs/gretlcli gretlcli.o complete.o
> ../lib/.libs/libgretl-1.0.so
> -ldl -L/usr/local/lib -lz -lxml2 -lglib-2.0 -lgmp -lfftw3 -lm
> -lreadline -ltermcap
> ../lib/.libs/libgretl-1.0.so: undefined reference to `dtrcon_'"
>
> That's just the same error: dtrcon_ is defined by liblapack.
>
> On my system:
>
> waverley:~$ /bin/ls -1 /usr/lib/liblapack*
> /usr/lib/liblapack.so
> /usr/lib/liblapack.so.3
> /usr/lib/liblapack.so.3.1
> /usr/lib/liblapack.so.3.1.1
> waverley:~$ /bin/ls -1 /usr/lib/libblas*
> /usr/lib/libblas.so
> /usr/lib/libblas.so.3
> /usr/lib/libblas.so.3.1
> /usr/lib/libblas.so.3.1.1
> waverley:~$ locate libgfortran.so
> /opt/gcc4/lib/libgfortran.so
> /opt/gcc4/lib/libgfortran.so.3
> /opt/gcc4/lib/libgfortran.so.3.0.0
>
> Allin Cottrell
>
>
>
>
> _______________________________________________
> Gretl-devel mailing list
> Gretl-devel(a)lists.wfu.edu
> http://lists.wfu.edu/mailman/listinfo/gretl-devel
>

--===============7422508592919858861==--


From cottrell@wfu.edu Sat May  3 17:25:41 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: Progress!,  was [Gretl-devel] Problems building and installing on
 Fedora 8
Date: Sat, 03 May 2008 17:25:31 -0400
Message-ID: <alpine.LRH.1.10.0805031712060.21625@ricardo.ecn.wfu.edu>
In-Reply-To: Pine.LNX.4.60.0805031638341.17017@brama.com
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5830728665126961134=="

--===============5830728665126961134==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sat, 3 May 2008, Max Pyziur wrote:

> On Fedora 8, libblas is installed separately. However, reviewing 
> your notes below I also installed blas-devel. This caused the 
> rpmbuild/compilation to go farther. (It might be worthwhile to 
> this dependency in the spec file).

The difficulty here is this: liblapack is useless without _some_ 
form of libblas (BLAS = Basic Linear Algebra System), but there 
are various blas's.  There's the libblas from the same (netlib) 
source as lapack itself, and there are versions that are tuned for 
best performance on a specific processor.  I'd say that it's 
really up to the OS/distro to ensure that if you choose to install 
lapack, some suitable form of blas is pulled in as a dependency.

Or in other words, this seems like a Fedora packaging bug to me.

> The following message registered after the configuration (which 
> didn't before blas-devel):

> Configuration...
[snip]
>   Build with gnome support:               2.0

So far as the configure script is concerned, then, gnome is OK.

> but then it failed with the following:
> 
> gcc -o .libs/gretl_x11 [object modules]
> -pthread -lgnomeui-2 -lSM -lICE -lbonoboui-2
> -lgnomevfs-2 -lgnomecanvas-2 -lgnome-2 -lpopt -lbonobo-2 -lbonobo-activation
> -lart_lgpl_2 -lgtk-x11-2.0 -lgdk-x11-2.0 -latk-1.0 -lpangocairo-1.0
> -lpango-1.0 -lcairo -lgconf-2 -lORBit-2 -lgthread-2.0 -lrt
> -L/usr/src/redhat/BUILD/gretl-1.7.4/gui2/gtksourceview -lgtksourceview-lite
> -L/usr/src/redhat/BUILD/gretl-1.7.4/gui2/gtkextra-lite -lgtkextra-lite
> ../lib/.libs/libgretl-1.0.so -L/usr/lib -llapack -lblas -lgfortran
> -L/usr/local/lib -lz -lxml2 -lgmp -lfftw3 -lm -lgdk_pixbuf-2.0 -lgobject-2.0
> -lgmodule-2.0 -ldl -lglib-2.0
> /usr/bin/ld: skipping incompatible /usr/lib/libgnomeui-2.so when searching for
> -lgnomeui-2

[ plus more "skipping incompatible" messages for gnome-related 
libraries.]

But so far as the linker is concerned, gnome is a mess!  Very 
weird.  What's wrong with your gnome libraries that the linker 
finds them "incompatible"?  Is this a 64-bit system?

You could try passing the option "--without-gnome" to configure 
(or put it in the .spec file), since the linker is not barfing on 
the other libraries that gretl wants.  And in fact, you don't lose 
much by dropping the gnome linkage.

Allin Cottrell


--===============5830728665126961134==--


From pyz@brama.com Sun May  4 14:42:18 2008
From: Max Pyziur <pyz@brama.com>
To: gretl-devel@gretlml.univpm.it
Subject: Re: Progress!,  was [Gretl-devel] Problems building and installing on
 Fedora 8
Date: Sun, 04 May 2008 14:27:27 -0400
Message-ID: <Pine.LNX.4.60.0805041407480.21325@brama.com>
In-Reply-To: Pine.LNX.4.60.0805041256230.6532@brama.com
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============3380095534547272678=="

--===============3380095534547272678==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


Trying to unravel my own thread, I appended the following lines to 
gretl.spec on the 64-bit machine:
%{prefix}/share/gretl/*
%{prefix}/share/locale/*
%{prefix}/share/man/*
%{prefix}/include/*
%{prefix}/lib64/*

and have reduced errors to the following two lines on the 64-bit machine:
RPM build errors:
     File not found by glob: /var/tmp/gretl-1.7.4-mp-root/usr/lib/lib*
     File not found by glob: /var/tmp/gretl-1.7.4-mp-root/usr/lib/gretl*/*


On the 32-bit laptop, I appended the following lines to the gretl.spec 
file:
%{prefix}/include/*
%{prefix}/lib/pkgconfig/*

and the compilation is successful. The 32-bit laptop compilation did not 
use the "--without-gnome" directive in the "configure" section. The rpm 
installs successfully both on the 32-bit machine and the 64-bit machine.

On the 64-bit machine, executing:
"gretl" brings up the gretl window and has the following messages in the 
terminal window
Gtk-Message: Failed to load module "gnomebreakpad": libgnomebreakpad.so: 
cannot open shared object file: No such file or directory
setlocale(LC_NUMERIC, "en_US.UTF-8") returned en_US.UTF-8

"gretl_x11" produces the same results as "gretl"

"gretlcli" produces a command-line interface w/ a "?" prompt.

On the 32-bit machine, executing both "gretl" and "gretl_x11" brings up 
the gretl window and has the following message in the terminal window
setlocale(LC_NUMERIC, "en_US.UTF-8") returned en_US.UTF-8

"gretlcli" produces a command-line interface on a second try. Initially, 
it produced the following:
pyz(a)mercury ~> gretlcli
gretl version 1.7.4
Copyright Ramu Ramanathan, Allin Cottrell and Riccardo "Jack" Lucchetti
This is free software with ABSOLUTELY NO WARRANTY
Current session: 2008/05/04 14:17
/home/pyz/.gretl/x12arima: No such file or directory
/home/pyz/.gretl/tramo: No such file or directory
No such file or directory
No such file or directory
No such file or directory
No such file or directory


Thanks! And what else is necessary to get the compilation to work on the 
64-bit machine?

Max Pyziur
pyz(a)brama.com




On Sun, 4 May 2008, Max Pyziur wrote:

> Thanks again for your help.
>
> I'm now trying to compile/build rpms on two different Fedora 8 machines:
> a 64-bit system Intel Core 2 Duo box, and
> a Dell 600M Inspiron laptop w/ a Pentium M processor
>
> I've tried your suggestion of including "--without-gnome" in the "configure" 
> section of the spec file on the 64-bit machine.
>
> The rpm compilation/build proceeds further and ends with the output appended 
> below.
>
> On the laptop, no "--without-gnome" directive was used. However, on the rpm 
> build the results were similar to those for the 64-bit machine (see second 
> section below).
>
> Thanks in advance.
>
> Max Pyziur
> pyz(a)brama.com
>
>
> [...]
>
>
>> But so far as the linker is concerned, gnome is a mess!  Very
>> weird.  What's wrong with your gnome libraries that the linker
>> finds them "incompatible"?  Is this a 64-bit system?
>> 
>> You could try passing the option "--without-gnome" to configure
>> (or put it in the .spec file), since the linker is not barfing on
>> the other libraries that gretl wants.  And in fact, you don't lose
>> much by dropping the gnome linkage.
>> 
>> Allin Cottrell
>> 
>
>
> ################ RPM build error on 64-bit machine ###################
> RPM build errors:
>    File not found by glob: /var/tmp/gretl-1.7.4-mp-root/usr/lib/lib*
>    File not found by glob: /var/tmp/gretl-1.7.4-mp-root/usr/lib/gretl*/*
>    Installed (but unpackaged) file(s) found:
>   /usr/include/adf_kpss.h
>   /usr/include/bhhh_max.h
>   /usr/include/bootstrap.h
>   /usr/include/calendar.h
>   /usr/include/compare.h
>   /usr/include/compat.h
>   /usr/include/csvdata.h
>   /usr/include/dataio.h
>   /usr/include/dataset.h
>   /usr/include/dbread.h
>   /usr/include/describe.h
>   /usr/include/discrete.h
>   /usr/include/estimate.h
>   /usr/include/forecast.h
>   /usr/include/genfuncs.h
>   /usr/include/genmain.h
>   /usr/include/graphing.h
>   /usr/include/gretl_commands.h
>   /usr/include/gretl_errors.h
>   /usr/include/gretl_fft.h
>   /usr/include/gretl_func.h
>   /usr/include/gretl_intl.h
>   /usr/include/gretl_list.h
>   /usr/include/gretl_matrix.h
>   /usr/include/gretl_model.h
>   /usr/include/gretl_panel.h
>   /usr/include/gretl_paths.h
>   /usr/include/gretl_prn.h
>   /usr/include/gretl_restrict.h
>   /usr/include/gretl_string_table.h
>   /usr/include/gretl_utils.h
>   /usr/include/gretl_win32.h
>   /usr/include/gretl_www.h
>   /usr/include/gretl_xml.h
>   /usr/include/interact.h
>   /usr/include/johansen.h
>   /usr/include/kalman.h
>   /usr/include/libgretl.h
>   /usr/include/libset.h
>   /usr/include/matrix_extra.h
>   /usr/include/missing.h
>   /usr/include/modelprint.h
>   /usr/include/modelspec.h
>   /usr/include/monte_carlo.h
>   /usr/include/nls.h
>   /usr/include/nonparam.h
>   /usr/include/objstack.h
>   /usr/include/options.h
>   /usr/include/plotspec.h
>   /usr/include/plugins.h
>   /usr/include/printout.h
>   /usr/include/printscan.h
>   /usr/include/pvalues.h
>   /usr/include/qr_estimate.h
>   /usr/include/random.h
>   /usr/include/strutils.h
>   /usr/include/subsample.h
>   /usr/include/system.h
>   /usr/include/texprint.h
>   /usr/include/transforms.h
>   /usr/include/tsls.h
>   /usr/include/usermat.h
>   /usr/include/var.h
>   /usr/include/varprint.h
>   /usr/include/vartest.h
>   /usr/lib64/libgretl-1.0.la
>   /usr/lib64/libgretl-1.0.so
>   /usr/lib64/libgretl-1.0.so.0
>   /usr/lib64/libgretl-1.0.so.0.0.44
>   /usr/lib64/pkgconfig/gretl.pc
>
>
> ################## RPM Build error on 32-bit machine #################
> RPM build errors:
>    Installed (but unpackaged) file(s) found:
>   /usr/include/adf_kpss.h
>   /usr/include/bhhh_max.h
>   /usr/include/bootstrap.h
>   /usr/include/calendar.h
>   /usr/include/compare.h
>   /usr/include/compat.h
>   /usr/include/csvdata.h
>   /usr/include/dataio.h
>   /usr/include/dataset.h
>   /usr/include/dbread.h
>   /usr/include/describe.h
>   /usr/include/discrete.h
>   /usr/include/estimate.h
>   /usr/include/forecast.h
>   /usr/include/genfuncs.h
>   /usr/include/genmain.h
>   /usr/include/graphing.h
>   /usr/include/gretl_commands.h
>   /usr/include/gretl_errors.h
>   /usr/include/gretl_fft.h
>   /usr/include/gretl_func.h
>   /usr/include/gretl_intl.h
>   /usr/include/gretl_list.h
>   /usr/include/gretl_matrix.h
>   /usr/include/gretl_model.h
>   /usr/include/gretl_panel.h
>   /usr/include/gretl_paths.h
>   /usr/include/gretl_prn.h
>   /usr/include/gretl_restrict.h
>   /usr/include/gretl_string_table.h
>   /usr/include/gretl_utils.h
>   /usr/include/gretl_win32.h
>   /usr/include/gretl_www.h
>   /usr/include/gretl_xml.h
>   /usr/include/interact.h
>   /usr/include/johansen.h
>   /usr/include/kalman.h
>   /usr/include/libgretl.h
>   /usr/include/libset.h
>   /usr/include/matrix_extra.h
>   /usr/include/missing.h
>   /usr/include/modelprint.h
>   /usr/include/modelspec.h
>   /usr/include/monte_carlo.h
>   /usr/include/nls.h
>   /usr/include/nonparam.h
>   /usr/include/objstack.h
>   /usr/include/options.h
>   /usr/include/plotspec.h
>   /usr/include/plugins.h
>   /usr/include/printout.h
>   /usr/include/printscan.h
>   /usr/include/pvalues.h
>   /usr/include/qr_estimate.h
>   /usr/include/random.h
>   /usr/include/strutils.h
>   /usr/include/subsample.h
>   /usr/include/system.h
>   /usr/include/texprint.h
>   /usr/include/transforms.h
>   /usr/include/tsls.h
>   /usr/include/usermat.h
>   /usr/include/var.h
>   /usr/include/varprint.h
>   /usr/include/vartest.h
>   /usr/lib/pkgconfig/gretl.pc
>
>

--===============3380095534547272678==--


From pyz@brama.com Sun May  4 18:46:06 2008
From: Max Pyziur <pyz@brama.com>
To: gretl-devel@gretlml.univpm.it
Subject: Re: Progress!,  was [Gretl-devel] Problems building and installing on
 Fedora 8
Date: Sun, 04 May 2008 13:10:07 -0400
Message-ID: <Pine.LNX.4.60.0805041256230.6532@brama.com>
In-Reply-To: alpine.LRH.1.10.0805031712060.21625@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0781297457480706409=="

--===============0781297457480706409==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Thanks again for your help.

I'm now trying to compile/build rpms on two different Fedora 8 machines:
a 64-bit system Intel Core 2 Duo box, and
a Dell 600M Inspiron laptop w/ a Pentium M processor

I've tried your suggestion of including "--without-gnome" in the 
"configure" section of the spec file on the 64-bit machine.

The rpm compilation/build proceeds further and ends with the output 
appended below.

On the laptop, no "--without-gnome" directive was used. However, on the 
rpm build the results were similar to those for the 64-bit machine (see 
second section below).

Thanks in advance.

Max Pyziur
pyz(a)brama.com


[...]


> But so far as the linker is concerned, gnome is a mess!  Very
> weird.  What's wrong with your gnome libraries that the linker
> finds them "incompatible"?  Is this a 64-bit system?
>
> You could try passing the option "--without-gnome" to configure
> (or put it in the .spec file), since the linker is not barfing on
> the other libraries that gretl wants.  And in fact, you don't lose
> much by dropping the gnome linkage.
>
> Allin Cottrell
>


################ RPM build error on 64-bit machine ###################
RPM build errors:
     File not found by glob: /var/tmp/gretl-1.7.4-mp-root/usr/lib/lib*
     File not found by glob: /var/tmp/gretl-1.7.4-mp-root/usr/lib/gretl*/*
     Installed (but unpackaged) file(s) found:
    /usr/include/adf_kpss.h
    /usr/include/bhhh_max.h
    /usr/include/bootstrap.h
    /usr/include/calendar.h
    /usr/include/compare.h
    /usr/include/compat.h
    /usr/include/csvdata.h
    /usr/include/dataio.h
    /usr/include/dataset.h
    /usr/include/dbread.h
    /usr/include/describe.h
    /usr/include/discrete.h
    /usr/include/estimate.h
    /usr/include/forecast.h
    /usr/include/genfuncs.h
    /usr/include/genmain.h
    /usr/include/graphing.h
    /usr/include/gretl_commands.h
    /usr/include/gretl_errors.h
    /usr/include/gretl_fft.h
    /usr/include/gretl_func.h
    /usr/include/gretl_intl.h
    /usr/include/gretl_list.h
    /usr/include/gretl_matrix.h
    /usr/include/gretl_model.h
    /usr/include/gretl_panel.h
    /usr/include/gretl_paths.h
    /usr/include/gretl_prn.h
    /usr/include/gretl_restrict.h
    /usr/include/gretl_string_table.h
    /usr/include/gretl_utils.h
    /usr/include/gretl_win32.h
    /usr/include/gretl_www.h
    /usr/include/gretl_xml.h
    /usr/include/interact.h
    /usr/include/johansen.h
    /usr/include/kalman.h
    /usr/include/libgretl.h
    /usr/include/libset.h
    /usr/include/matrix_extra.h
    /usr/include/missing.h
    /usr/include/modelprint.h
    /usr/include/modelspec.h
    /usr/include/monte_carlo.h
    /usr/include/nls.h
    /usr/include/nonparam.h
    /usr/include/objstack.h
    /usr/include/options.h
    /usr/include/plotspec.h
    /usr/include/plugins.h
    /usr/include/printout.h
    /usr/include/printscan.h
    /usr/include/pvalues.h
    /usr/include/qr_estimate.h
    /usr/include/random.h
    /usr/include/strutils.h
    /usr/include/subsample.h
    /usr/include/system.h
    /usr/include/texprint.h
    /usr/include/transforms.h
    /usr/include/tsls.h
    /usr/include/usermat.h
    /usr/include/var.h
    /usr/include/varprint.h
    /usr/include/vartest.h
    /usr/lib64/libgretl-1.0.la
    /usr/lib64/libgretl-1.0.so
    /usr/lib64/libgretl-1.0.so.0
    /usr/lib64/libgretl-1.0.so.0.0.44
    /usr/lib64/pkgconfig/gretl.pc


################## RPM Build error on 32-bit machine #################
RPM build errors:
     Installed (but unpackaged) file(s) found:
    /usr/include/adf_kpss.h
    /usr/include/bhhh_max.h
    /usr/include/bootstrap.h
    /usr/include/calendar.h
    /usr/include/compare.h
    /usr/include/compat.h
    /usr/include/csvdata.h
    /usr/include/dataio.h
    /usr/include/dataset.h
    /usr/include/dbread.h
    /usr/include/describe.h
    /usr/include/discrete.h
    /usr/include/estimate.h
    /usr/include/forecast.h
    /usr/include/genfuncs.h
    /usr/include/genmain.h
    /usr/include/graphing.h
    /usr/include/gretl_commands.h
    /usr/include/gretl_errors.h
    /usr/include/gretl_fft.h
    /usr/include/gretl_func.h
    /usr/include/gretl_intl.h
    /usr/include/gretl_list.h
    /usr/include/gretl_matrix.h
    /usr/include/gretl_model.h
    /usr/include/gretl_panel.h
    /usr/include/gretl_paths.h
    /usr/include/gretl_prn.h
    /usr/include/gretl_restrict.h
    /usr/include/gretl_string_table.h
    /usr/include/gretl_utils.h
    /usr/include/gretl_win32.h
    /usr/include/gretl_www.h
    /usr/include/gretl_xml.h
    /usr/include/interact.h
    /usr/include/johansen.h
    /usr/include/kalman.h
    /usr/include/libgretl.h
    /usr/include/libset.h
    /usr/include/matrix_extra.h
    /usr/include/missing.h
    /usr/include/modelprint.h
    /usr/include/modelspec.h
    /usr/include/monte_carlo.h
    /usr/include/nls.h
    /usr/include/nonparam.h
    /usr/include/objstack.h
    /usr/include/options.h
    /usr/include/plotspec.h
    /usr/include/plugins.h
    /usr/include/printout.h
    /usr/include/printscan.h
    /usr/include/pvalues.h
    /usr/include/qr_estimate.h
    /usr/include/random.h
    /usr/include/strutils.h
    /usr/include/subsample.h
    /usr/include/system.h
    /usr/include/texprint.h
    /usr/include/transforms.h
    /usr/include/tsls.h
    /usr/include/usermat.h
    /usr/include/var.h
    /usr/include/varprint.h
    /usr/include/vartest.h
    /usr/lib/pkgconfig/gretl.pc


--===============0781297457480706409==--


From cottrell@wfu.edu Sun May  4 20:29:08 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: Progress!,  was [Gretl-devel] Problems building and installing on
 Fedora 8
Date: Sun, 04 May 2008 20:29:04 -0400
Message-ID: <alpine.LRH.1.10.0805042020580.3717@ricardo.ecn.wfu.edu>
In-Reply-To: Pine.LNX.4.60.0805041256230.6532@brama.com
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============8088780440240911760=="

--===============8088780440240911760==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sun, 4 May 2008, Max Pyziur wrote:

> Thanks again for your help.
> 
> I'm now trying to compile/build rpms on two different Fedora 8 machines:
> a 64-bit system Intel Core 2 Duo box, and
> a Dell 600M Inspiron laptop w/ a Pentium M processor
> 
> I've tried your suggestion of including "--without-gnome" in the 
> "configure" section of the spec file on the 64-bit machine.
> 
> The rpm compilation/build proceeds further and ends with the 
> output appended below.
> 
> On the laptop, no "--without-gnome" directive was used. However, 
> on the rpm build the results were similar to those for the 
> 64-bit machine (see second section below).
> 
> ################ RPM build error on 64-bit machine ###################
> RPM build errors:
>     File not found by glob: /var/tmp/gretl-1.7.4-mp-root/usr/lib/lib*
>     File not found by glob: /var/tmp/gretl-1.7.4-mp-root/usr/lib/gretl*/*
>     Installed (but unpackaged) file(s) found:
>    /usr/include/adf_kpss.h
>    /usr/include/bhhh_max.h...

I'm not an rpm expert, but it appears your .spec file must be 
wrong.  RPM is apparently complaining about the fact that the 
libgretl headers (e.g. /usr/include/adf_kpss.h) are being 
installed but not packaged.  That's to be expected, in a sense: 
the headers are installed by a normal "make install", but one 
would not expect them to be included in a binary .rpm, only in the 
associated "devel" .rpm.  

I suggest either (a) taking a closer look at the RPM 
documentation, or (b) ignoring RPM and just installing gretl 
independently of that mechanism.  If you're looking to create an 
RPM that you can redistribute then (a) will be required, 
otherwise (b) will probably save you grief.

Allin Cottrell

--===============8088780440240911760==--


From cottrell@wfu.edu Sun May  4 20:49:54 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: Progress!,  was [Gretl-devel] Problems building and installing on
 Fedora 8
Date: Sun, 04 May 2008 20:49:50 -0400
Message-ID: <alpine.LRH.1.10.0805042042250.3717@ricardo.ecn.wfu.edu>
In-Reply-To: Pine.LNX.4.60.0805041407480.21325@brama.com
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0523759101422133688=="

--===============0523759101422133688==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

This is in reverse order, but...

On Sun, 4 May 2008, Max Pyziur wrote:

> Trying to unravel my own thread, I appended the following lines 
> to gretl.spec on the 64-bit machine:

> %{prefix}/share/gretl/*
> %{prefix}/share/locale/*
> %{prefix}/share/man/*
> %{prefix}/include/*

That last line is wrong, isn't it?  You don't want the include 
directory in the spec if you're building a binary rpm.

> On the 64-bit machine, executing:
> "gretl" brings up the gretl window and has the following messages in the
> terminal window
> Gtk-Message: Failed to load module "gnomebreakpad": libgnomebreakpad.so:
> cannot open shared object file: No such file or directory

I have no idea about libgnomebreakpad.so; it's not referenced 
anywhere in the gretl build files.  I suppose it's being dragged 
in by pkg-config, in which case the pkg-config setup is broken.

Allin Cottrell.


--===============0523759101422133688==--


From ignacio.diaz-emparanza@ehu.es Tue May  6 06:26:26 2008
From: Ignacio Diaz-Emparanza <ignacio.diaz-emparanza@ehu.es>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] ARMAX and diffs
Date: Tue, 06 May 2008 12:26:22 +0200
Message-ID: <200805061226.22766.ignacio.diaz-emparanza@ehu.es>
In-Reply-To: alpine.LRH.1.10.0805042042250.3717@ricardo.ecn.wfu.edu
Content-Type: multipart/mixed; boundary="===============4622637573314744714=="

--===============4622637573314744714==
Content-Type: application/x-gzip
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="import-NE.gdt"
MIME-Version: 1.0
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--===============4622637573314744714==--


From cottrell@wfu.edu Tue May  6 09:12:54 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] ARMAX and diffs
Date: Tue, 06 May 2008 09:12:46 -0400
Message-ID: <alpine.LRH.1.10.0805060905410.18484@ricardo.ecn.wfu.edu>
In-Reply-To: 200805061226.22766.ignacio.diaz-emparanza@ehu.es
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============3532640808853742385=="

--===============3532640808853742385==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Tue, 6 May 2008, Ignacio Diaz-Emparanza wrote:

> I was trying to estimate an ARIMA model with a intervention (a 
> level shift)... and found what I think is an inconsistency in 
> the treatment of the explanatory variable.
> 
> Could you, Allin or Jack, please correct this behaviour?

Thanks, Ignacio.  I think I'm going to wait for Jack's reaction 
on this point rather than jumping in myself.  But just to check my 
understanding: your point is that in an ARIMA model any exogenous 
variables should be subject to the same differencing that is 
applied to y(t)?

Allin.

--===============3532640808853742385==--


From r.lucchetti@univpm.it Tue May  6 09:24:19 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] ARMAX and diffs
Date: Tue, 06 May 2008 15:24:16 +0200
Message-ID: <alpine.DEB.1.10.0805061443180.14342@ec-4.econ.univpm.it>
In-Reply-To: 200805061226.22766.ignacio.diaz-emparanza@ehu.es
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5005175726708367178=="

--===============5005175726708367178==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Tue, 6 May 2008, Ignacio Diaz-Emparanza wrote:

> I was trying to estimate an ARIMA model with a intervention (a level shift).
...
> Could you, Allin or Jack, please correct this behaviour?

Ignacio,

after some pondering, I believe you're right. What I would consider a 
proper fix for this issue, however, takes us right in the middle of the 
parallel discussion about backward-incompatible changes. Let me explain.

To recap briefly what's in the manual already, what we currently have 
under the "arima" heading is in fact a pair of quite different things. On 
one hand, we have that ML estimation (carried out via the Kalman filter) 
estimates a model like

(1) A(L) (y_t - x_t'\beta) = C(L) e_t

If we believe that the differencing operator is part and parcel of the 
A(L) polynomial, we ought to difference x_t as well, not just y_t 
(contrary to what we do now). This would be, in itself, a 
backward-incompatible change already.

Now consider the class of models that we handle via conditional ML: here 
we have

(2) A(L) y_t = x_t'\beta + C(L) e_t.

As the manual says, whenever x_t includes anything more than the constant, 
(1) and (2) are fundamentally different models, since the term x_t'\beta 
can be considered as the conditional mean of y_t with respect to two 
different information set. In the context of (2), the A(L) polynomial does 
_not_ apply to x_t, so there's no reason to difference x_t too.

The important thing here is that (1) is the formulation generally adopted 
in the time-series literature, while (2) resembles more the models 
econometricians use, (ARDL models, GARCH models, and so on). If we 
implement the change proposed by Ignacio (which, I repeat, makes perfect 
sense for (1)), the two models would diverge in their behaviour and 
interpretation even more, to the point that I wouldn't find it 
inconceivable to have two separate commands and two different GUI menu 
entries for (1) and (2) (although, it must be said, it'd be relatively 
easy to estimate (1) via conditional ML if we wanted to).

Opinions?


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============5005175726708367178==--


From cottrell@wfu.edu Tue May  6 12:14:52 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] new tracker for incompatible changes
Date: Tue, 06 May 2008 12:14:46 -0400
Message-ID: <alpine.LRH.1.10.0805061209230.18484@ricardo.ecn.wfu.edu>
In-Reply-To: 481B2783.7080002@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5930380672659681799=="

--===============5930380672659681799==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 2 May 2008, Sven Schreiber wrote:

> quickly @b: I'm not sure right who may edit it. I have set some 
> permissions, where the six gretl developers I could identify in 
> the sourceforge list are made admins of that new tracker. I'm 
> not sure what that means for regular (logged-in) users, it would 
> be nice if someone who's not in the sourceforge gretl developers 
> list could test.

I wonder if we might be better with a cvs-controlled log of 
incompatible changes, much like the current ChangeLog.

I've done a mock-up of this; it's not linked to from the main 
page, but you can find it at:

http://gretl.sourceforge.net/Backward.html

Any thoughts on the relative merits of these approaches?

Allin.

--===============5930380672659681799==--


From svetosch@gmx.net Tue May  6 12:48:58 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] new tracker for incompatible changes
Date: Tue, 06 May 2008 18:50:05 +0200
Message-ID: <48208C3D.2050800@gmx.net>
In-Reply-To: alpine.LRH.1.10.0805061209230.18484@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2360413906034140860=="

--===============2360413906034140860==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Allin Cottrell schrieb:
> On Fri, 2 May 2008, Sven Schreiber wrote:
> 
>> quickly @b: I'm not sure right who may edit it. I have set some 
>> permissions, where the six gretl developers I could identify in 
>> the sourceforge list are made admins of that new tracker. I'm 
>> not sure what that means for regular (logged-in) users, it would 
>> be nice if someone who's not in the sourceforge gretl developers 
>> list could test.
> 
> I wonder if we might be better with a cvs-controlled log of 
> incompatible changes, much like the current ChangeLog.
> 
> I've done a mock-up of this; it's not linked to from the main 
> page, but you can find it at:
> 
> http://gretl.sourceforge.net/Backward.html
> 
> Any thoughts on the relative merits of these approaches?

your page looks good to me. I have no particular vested interested in 
the tracker solution, I just want a transparent documentation.

cheers,
sven


--===============2360413906034140860==--


From r.lucchetti@univpm.it Tue May  6 12:52:41 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] new tracker for incompatible changes
Date: Tue, 06 May 2008 18:52:37 +0200
Message-ID: <alpine.DEB.1.10.0805061832210.16965@ec-4.econ.univpm.it>
In-Reply-To: alpine.LRH.1.10.0805061209230.18484@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============7563538674890185062=="

--===============7563538674890185062==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Tue, 6 May 2008, Allin Cottrell wrote:

> On Fri, 2 May 2008, Sven Schreiber wrote:
>
>> quickly @b: I'm not sure right who may edit it. I have set some
>> permissions, where the six gretl developers I could identify in
>> the sourceforge list are made admins of that new tracker. I'm
>> not sure what that means for regular (logged-in) users, it would
>> be nice if someone who's not in the sourceforge gretl developers
>> list could test.
>
> I wonder if we might be better with a cvs-controlled log of
> incompatible changes, much like the current ChangeLog.
>
> I've done a mock-up of this; it's not linked to from the main
> page, but you can find it at:
>
> http://gretl.sourceforge.net/Backward.html
>
> Any thoughts on the relative merits of these approaches?

I like the Changelog-style better. Pros:

a) no spam involved

b) it's rather easy to look for stuff: you open the web page, and 
everything is there. If in the future the list greows, all you have to do 
is press Ctrl-F and you're pretty much done.

c) requires very little maintenance, but more or less the same holds for 
the SF tracker

d) optionally, it can be turned into an Emacs-style "forward-incompatible 
unfeatures" (if you have no idea of what I'm talking about, never mind, 
it's an in-joke)

Cons: none I can see right now.

While we're at it: the Changelog is nice, but not exactly newbie-friendly. 
For the next release, it'd be nice if someone volunteered to write a 
"Release Notes" page, more or less in the style of Abiword's 
(http://www.abiword.org/release-notes/2.6.0.phtml) or Gnumeric's
(http://www.gnome.org/projects/gnumeric/announcements/1.8/gnumeric-1.8.shtml).

There are several people on the devel list that I think would do an 
excellent job (Sven, Andreas, Talha, to name but a few).


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============7563538674890185062==--


From andreas.rosenblad@ltv.se Wed May  7 05:11:27 2008
From: andreas.rosenblad@ltv.se
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] kalman filter status?
Date: Wed, 07 May 2008 11:11:22 +0200
Message-ID: <OF030925FD.CDABE285-ONC1257442.00325D63-C1257442.00327B26@ltv.se>
In-Reply-To: OFF582DD1D.F1829D45-ONC1257439.003BC4BF@LocalDomain
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============8531800082263469920=="

--===============8531800082263469920==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


svetosch(a)gmx.net @ INTERNET skrev 2008-04-28 12:52:49 :

> Hi,
>
> I think I've asked about it before, but the "big" feature I'm
> missing in gretl is the Kalman filter for state-space models.

The "big" feature I am missing in gretl is quantile regression.

Andreas


--===============8531800082263469920==--


From r.lucchetti@univpm.it Wed May  7 05:39:15 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] kalman filter status?
Date: Wed, 07 May 2008 11:39:10 +0200
Message-ID: <alpine.DEB.1.10.0805071114480.24982@ec-4.econ.univpm.it>
In-Reply-To: OF030925FD.CDABE285-ONC1257442.00325D63-C1257442.00327B26@ltv.se
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0560889647314510063=="

--===============0560889647314510063==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Wed, 7 May 2008, andreas.rosenblad(a)ltv.se wrote:

>
> svetosch(a)gmx.net @ INTERNET skrev 2008-04-28 12:52:49 :
>
>> Hi,
>>
>> I think I've asked about it before, but the "big" feature I'm
>> missing in gretl is the Kalman filter for state-space models.
>
> The "big" feature I am missing in gretl is quantile regression.

Both features are, at the moment, in the planning stage. My guess is that 
we'll release 1.7.5 quite soon to incorporate in the official version the 
bug fixes we've accumulated so far, plus a few extra features. If you look 
at the Changelog at http://gretl.sourceforge.net/ChangeLog.html, you'll 
see there's a hefty amount of stuff already.

What I'd like to have in before releasing 1.7.5 are

1) we're working on an "ODBC connector", which enables gretl to fetch data 
from DBMSes like Oracle, MySQL, Postgres etcetera. The basic 
infrastructure is already in CVS: the missing bits are (a) documentation 
(b) thorough testing (c) a reasonable GUI interface. IMO, chances are this 
can be done within the end of May and the ODBC facility (marked as 
"experimental") can be included in 1.7.5;

2) I haven't had time to look at geno's anova patch (sorry, geno!), but it 
seems likely that it shouldn't take long to incorporate it into CVS.

IMO, once this is done we may release 1.7.5 and concentrate on 
kalman + quantile regression (and more).


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============0560889647314510063==--


From cottrell@wfu.edu Wed May  7 07:58:39 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] kalman filter status?
Date: Wed, 07 May 2008 07:58:32 -0400
Message-ID: <alpine.LRH.1.10.0805070757470.10090@ricardo.ecn.wfu.edu>
In-Reply-To: OF030925FD.CDABE285-ONC1257442.00325D63-C1257442.00327B26@ltv.se
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============1163940697328421687=="

--===============1163940697328421687==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Wed, 7 May 2008, andreas.rosenblad(a)ltv.se wrote:

> The "big" feature I am missing in gretl is quantile regression.

I'm actually working on that at present, and may have something to 
report before too long.

Allin.

--===============1163940697328421687==--


From svetosch@gmx.net Thu May  8 05:31:42 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] new tracker for incompatible changes
Date: Thu, 08 May 2008 11:32:53 +0200
Message-ID: <4822C8C5.8040309@gmx.net>
In-Reply-To: alpine.DEB.1.10.0805061832210.16965@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============8973951766223541974=="

--===============8973951766223541974==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Riccardo (Jack) Lucchetti schrieb:

> 
> While we're at it: the Changelog is nice, but not exactly 
> newbie-friendly. For the next release, it'd be nice if someone 
> volunteered to write a "Release Notes" page, more or less in the style 
> of Abiword's (http://www.abiword.org/release-notes/2.6.0.phtml) or 
> Gnumeric's
> (http://www.gnome.org/projects/gnumeric/announcements/1.8/gnumeric-1.8.shtml). 
> 
> 
> There are several people on the devel list that I think would do an 
> excellent job (Sven, Andreas, Talha, to name but a few).
> 

If writing the release notes is not strictly tied to the release date 
(but for example could happen a week after release), then I guess I 
would be willing to do that. OTOH, I seem to remember that Allin said it 
would be easier if he or you do it.

Just let me know.

-sven

--===============8973951766223541974==--


From svetosch@gmx.net Thu May  8 05:34:25 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] kalman filter status?
Date: Thu, 08 May 2008 11:35:37 +0200
Message-ID: <4822C969.1010605@gmx.net>
In-Reply-To: alpine.DEB.1.10.0805071114480.24982@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============1419377711270454020=="

--===============1419377711270454020==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Riccardo (Jack) Lucchetti schrieb:

> 
> 1) we're working on an "ODBC connector", which enables gretl to fetch 
> data from DBMSes like Oracle, MySQL, Postgres etcetera. The basic 
> infrastructure is already in CVS: the missing bits are (a) documentation 
> (b) thorough testing (c) a reasonable GUI interface. IMO, chances are 
> this can be done within the end of May and the ODBC facility (marked as 
> "experimental") can be included in 1.7.5;

in principle this sounds cool, but what actually is the pressing need 
for it? Or to put it differently: should 1.7.5 really be delayed for 
that reason?


> 
> IMO, once this is done we may release 1.7.5 and concentrate on kalman + 
> quantile regression (and more).
> 

let me repeat my other question: do you want/need suggestions for the 
Kalman "API"/syntax/interface?

thanks,
sven

--===============1419377711270454020==--


From ignacio.diaz-emparanza@ehu.es Thu May  8 05:42:14 2008
From: Ignacio Diaz-Emparanza <ignacio.diaz-emparanza@ehu.es>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] ARMAX and diffs
Date: Thu, 08 May 2008 11:42:10 +0200
Message-ID: <200805081142.10971.ignacio.diaz-emparanza@ehu.es>
In-Reply-To: alpine.DEB.1.10.0805061443180.14342@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5555462491031084314=="

--===============5555462491031084314==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

El Tuesday 06 May 2008 15:24:16 Riccardo (Jack) Lucchetti escribi=C3=B3:
>
> Ignacio,
>
> after some pondering, I believe you're right. What I would consider a
> proper fix for this issue, however, takes us right in the middle of the
> parallel discussion about backward-incompatible changes. Let me explain.
>
> To recap briefly what's in the manual already, what we currently have
> under the "arima" heading is in fact a pair of quite different things. On
> one hand, we have that ML estimation (carried out via the Kalman filter)
> estimates a model like
>
> (1) A(L) (y_t - x_t'\beta) =3D C(L) e_t
>
> If we believe that the differencing operator is part and parcel of the
> A(L) polynomial, we ought to difference x_t as well, not just y_t
> (contrary to what we do now). This would be, in itself, a
> backward-incompatible change already.

I think by default we should not do any "backward incomptible" changes unless
1) We are correcting an error, or
2) We are making a great improvement of some command or feature

With respect to the gretl native command "arima" I think we are not in any of=
=20
these cases, so I propose we should re-think how to solve this in another=20
way.


>
> Now consider the class of models that we handle via conditional ML: here
> we have
>
> (2) A(L) y_t =3D x_t'\beta + C(L) e_t.
>
> As the manual says, whenever x_t includes anything more than the constant,
> (1) and (2) are fundamentally different models, since the term x_t'\beta
> can be considered as the conditional mean of y_t with respect to two
> different information set. In the context of (2), the A(L) polynomial does
> _not_ apply to x_t, so there's no reason to difference x_t too.

The question here is the interpretation about where the unit root is. It can =

go as a multiplicative factor in A(L), as it seems yo are thinking,=20
A(L)=3DA(L)'*(1-L) having A(L)' the roots outside the unit circle, or it can =
go=20
as a denominator of C(L)=3DC'(L)/(1-L) havin C'(L) its roots outside the unit=
=20
circle.

In equation (1) being (1-L) in A(L) or in C(L) has the same effect, but not i=
n=20
equation (2).

In the context of transfer function models is very common to write the=20
equation as

(3) y_t =3D v(L)*x_t + Nt

where the noise Nt is in general Nt=3DC(L)/A(L)*(1-L)^d. I am working with th=
is=20
type of models so I thought was natural that the xt variable I was puting in =

the equation was no differenced.

>
> The important thing here is that (1) is the formulation generally adopted
> in the time-series literature, while (2) resembles more the models
> econometricians use, (ARDL models, GARCH models, and so on). If we
> implement the change proposed by Ignacio (which, I repeat, makes perfect
> sense for (1)), the two models would diverge in their behaviour=20

No, please.

> and=20
> interpretation even more, to the point that I wouldn't find it
> inconceivable to have two separate commands and two different GUI menu
> entries for (1) and (2) (although, it must be said, it'd be relatively
> easy to estimate (1) via conditional ML if we wanted to).

I think we shoud leave the things as simple as possible. Your interpretation =

about the model is reasonable, so we only need that the user can know this=20
interpretation. If the gretl native command is left how it is now, we only=20
need that the user know what gretl is doing with the xt variable. So the=20
problem is solved if we only add a paragraph in the manual explaining how the=
=20
differences are applied to the model when there are independent variables.

But when you use the option --x12-arima the model that gretl estimates is=20
different that the one being estimated without this option. And I think this =

is an inconsistency that should be corrected, in spite of it implies=20
a "backward incompatible" change.




--=20
Ignacio Diaz-Emparanza =20
DEPARTAMENTO DE ECONOM=C3=8DA APLICADA III (ECONOMETR=C3=8DA Y ESTAD=C3=8DSTI=
CA)                                       =20
UPV/EHU
Avda. Lehendakari Aguirre, 83 | 48015 BILBAO
T.: +34 946013732 | F.: +34 946013754
www.et.bs.ehu.es=20


--===============5555462491031084314==--


From svetosch@gmx.net Fri May  9 17:41:34 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] new tracker for incompatible changes
Date: Fri, 09 May 2008 23:41:26 +0200
Message-ID: <4824C506.9000109@gmx.net>
In-Reply-To: alpine.LRH.1.10.0805061209230.18484@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2606571556605802594=="

--===============2606571556605802594==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Am 06.05.2008 18:14, Allin Cottrell schrieb:
> On Fri, 2 May 2008, Sven Schreiber wrote:
> 
>> quickly @b: I'm not sure right who may edit it. I have set some 
>> permissions, where the six gretl developers I could identify in 
>> the sourceforge list are made admins of that new tracker. I'm 
>> not sure what that means for regular (logged-in) users, it would 
>> be nice if someone who's not in the sourceforge gretl developers 
>> list could test.
> 
> I wonder if we might be better with a cvs-controlled log of 
> incompatible changes, much like the current ChangeLog.
> 
> I've done a mock-up of this; it's not linked to from the main 
> page, but you can find it at:
> 
> http://gretl.sourceforge.net/Backward.html
> 

It seems everybody agrees to have that cvs-based thing on the webpage. 
So I guess it should be linked to from the main page. And the new 
tracker must be removed, but it looks as if only Allin has the 
permissions to do it, I just tried but couldn't find any relevant button.

-sven




--===============2606571556605802594==--


From cottrell@wfu.edu Fri May  9 23:28:00 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] kalman filter status?
Date: Fri, 09 May 2008 23:27:40 -0400
Message-ID: <alpine.LRH.1.10.0805092320271.30820@ricardo.ecn.wfu.edu>
In-Reply-To: 4822C969.1010605@gmx.net
Content-Type: multipart/mixed; boundary="===============3274299171833104426=="

--===============3274299171833104426==
Content-Type: text/plain
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="Kalman.tex"
MIME-Version: 1.0

XGRvY3VtZW50Y2xhc3NbYTRwYXBlciwxMXB0XXthcnRpY2xlfQ0KDQpcdXNlcGFja2FnZXtibX0N
Clx1c2VwYWNrYWdle3ZlcmJhdGltfQ0KDQpcbmV3Y29tbWFuZHtcb2JzdmVjfXtcZW5zdXJlbWF0
aFxtYXRoYmZ7eX19DQpcbmV3Y29tbWFuZHtcb2JzbWF0fXtcZW5zdXJlbWF0aFxtYXRoYmZ7SH19
DQpcbmV3Y29tbWFuZHtcb2JzZGlzdH17XGVuc3VyZW1hdGhcbWF0aGJme3d9fQ0KXG5ld2NvbW1h
bmR7XG9ic3Zhcn17XGVuc3VyZW1hdGhcbWF0aGJme1J9fQ0KDQpcbmV3Y29tbWFuZHtcc3RhdGV2
ZWN9e1xlbnN1cmVtYXRoe1xibVx4aX19DQpcbmV3Y29tbWFuZHtcc3RybWF0fXtcZW5zdXJlbWF0
aFxtYXRoYmZ7Rn19DQpcbmV3Y29tbWFuZHtcc3RyZGlzdH17XGVuc3VyZW1hdGhcbWF0aGJme3Z9
fQ0KXG5ld2NvbW1hbmR7XHN0cnZhcn17XGVuc3VyZW1hdGhcbWF0aGJme1F9fQ0KXG5ld2NvbW1h
bmR7XGhhdHN0YXRldmFyfXtcZW5zdXJlbWF0aFxtYXRoYmZ7UH19DQoNClxuZXdjb21tYW5ke1xw
cmVkZXJyfXtcZW5zdXJlbWF0aFxtYXRoYmZ7ZX19DQpcbmV3Y29tbWFuZHtccHJlZHZhcn17XGVu
c3VyZW1hdGh7XGJtXFNpZ21hfX0NCg0KXG5ld2NvbW1hbmR7XGV4b2d9e1xlbnN1cmVtYXRoe1xt
YXRoYmZ7en19fQ0KXG5ld2NvbW1hbmR7XHBhcmFtfXtcZW5zdXJlbWF0aHtcYm1cdGhldGF9fQ0K
DQpcdGl0bGV7UkZDOiBJbXBsZW1lbnRhdGlvbiBpbiBcdGV4dHR0e2dyZXRsfSBvZiB0aGUgS2Fs
bWFuIGZpbHRlcg0KICBhdCB0aGUgdXNlciBsZXZlbH0NClxhdXRob3J7SmFja30NCg0KXGJlZ2lu
e2RvY3VtZW50fQ0KDQpcbWFrZXRpdGxlDQoNClRoZSBLYWxtYW4gZmlsdGVyIGlzIG9uZSBvZiB0
aG9zZSBmdW5jdGlvbmFsaXRpZXMgdGhhdCBtYXkgY29tZSBpbg0KaGFuZHkgaW4gYSBudW1iZXIg
b2YgY2FzZXM6IG9mIGNvdXJzZSB3ZSBhbHJlYWR5IHVzZSBpbnRlcm5hbGx5IGl0IGZvcg0KQVJN
QSBlc3RpbWF0aW9uLCBidXQgZm9yIGV4YW1wbGUsIGl0J2QgYmUgbmljZSB0byBoYXZlIGEgdG9v
bCBsaWtlDQp0aGF0IGZvciBwZW9wbGUgd2hvIGRvIERTR0UgbW9kZWxzIG9yIG1heWJlIGZvciBw
ZW9wbGUgd2hvIHdvdWxkIGxpa2UNCnRvIGltcGxlbWVudCBBUk1BIGVzdGltYXRpb24gd2l0aCBz
ZXJpZXMgd2l0aCBtaXNzaW5nIHZhbHVlcyBhbmQgc28NCm9uLg0KDQpcc2VjdGlvbip7Tm90YXRp
b259DQoNCk5vdGF0aW9uIFxlbXBoe2lzfSBhIHByb2JsZW0uIEl0IHNlZW1zIHRoYXQgaW4gZWNv
bm9tZXRyaWNzIGV2ZXJ5b25lDQppcyBoYXBweSB3aXRoICR5ID0gWCBcYmV0YSArIHUkLCBidXQg
d2UgY2FuJ3QsIGFzIGEgY29tbXVuaXR5LCBtYWtlIHVwDQpvdXIgbWluZHMgb24gYSBzdGFuZGFy
ZCBmb3Igc3RhdGUtc3BhY2UgbW9kZWxzLiBIYXJ2ZXkgKDE5OTEpLA0KSGFtaWx0b24gKDE5OTQp
LCBIYXJ2ZXkgXCYgUHJvaWV0dGkgKDIwMDUpIGFuZCBQb2xsb2NrICgxOTk5KSBhbGwgdXNlDQpk
aWZmZXJlbnQgY29udmVudGlvbnMuIFRoZSBub3RhdGlvbiBoZXJlIHdpbGwgYmUgYmFzZWQgb24g
SGFtaWx0b24ncywNCndpdGggc2xpZ2h0IHZhcmlhdGlvbnMuDQoNCkEgc3RhdGUtc3BhY2UgbW9k
ZWwgY2FuIGJlIHdyaXR0ZW4gZG93biBhcw0KXGJlZ2lue2VxbmFycmF5fQ0KICBcbGFiZWx7ZXE6
b2JzZXF9DQogIFxvYnN2ZWNfdCAmID0gJiBcb2JzbWF0X3QgXHN0YXRldmVjX3QgKyBcbWF0aGJm
e2R9X3QgKyBcb2JzZGlzdF90IFxcDQogIFxsYWJlbHtlcTpzdHJhbmVxfQ0KICBcc3RhdGV2ZWNf
e3R9ICYgPSAmIFxzdHJtYXRfe3R9IFxzdGF0ZXZlY197dC0xfSArIFxtYXRoYmZ7Y31fdCArIFxz
dHJkaXN0X3QNClxlbmR7ZXFuYXJyYXl9DQpUaGUgc3ltYm9sICRUJCAobm90IGluIGJvbGQpIGlu
ZGljYXRlcyB0aGUgc2FtcGxlIHNpemUuIFRoZQ0Kb2JzZXJ2YWJsZXMgJFxvYnN2ZWNfdCQgYXJl
IGFuICRuJC12ZWN0b3IgKGluIG1hbnkgY2FzZXMsICRuPTEkLA0KYnV0IG5vdCBuZWNlc3Nhcmls
eSksIHdoaWxlIHRoZSB1bm9ic2VydmFibGUgdmVjdG9yIG9mIHN0YXRlcw0KJFxzdGF0ZXZlY190
JCBoYXMgJG0kIGVsZW1lbnRzLiU7IHVzdWFsbHksIGlmIG5vdCBhbHdheXMsICRuIDwgbSQuDQpU
aGUgY292YXJpYW5jZSBtYXRyaWNlcyBmb3IgJFxvYnNkaXN0X3QkIGFuZCAkXHN0cmRpc3RfdCQg
YXJlICRcb2JzdmFyX3QkIGFuZA0KJFxzdHJ2YXJfdCQsIHJlc3BlY3RpdmVseS4NCg0KSW4gdGhl
IHNwZWNpYWwgY2FzZSB3aGVuICRcc3RybWF0X3QgPSBcc3RybWF0JCwgJFxvYnNtYXRfdCA9DQpc
b2JzbWF0JCBhbmQgc28gb24sIHRoZSBtb2RlbCB3aWxsIGJlIGNhbGxlZCBcZW1waHt0aW1lLWlu
dmFyaWFudH0uDQoNCldlIGhhdmUgdHdvIG9wdGlvbnM6IHdlIG1heSBnbyBmb3J3YXJkIGluIHRp
bWUgYW5kIGRvIHRoZQ0KXGVtcGh7ZmlsdGVyaW5nfSBwcm9wZXIsIG9yIGFmdGVyIHRoYXQgd2Ug
Z28gYmFjayBhbmQgZG8gdGhlDQpcZW1waHtzbW9vdGhpbmd9LiANCg0KVG8gcXVvdGUgZnJvbSBQ
b2xsb2NrICgxOTk5KSwgcC4gMjQxICh3aXRoIGRpZmZlcmVudCBub3RhdGlvbiksIA0KXGJlZ2lu
e3F1b3RlfQ0KW1xsZG90c10gaXQgaXMgYXNzdW1lZCB0aGF0IHByaW9yIGluZm9ybWF0aW9uIG9u
IHRoZSBwcmV2aW91cyBzdGF0ZSB2ZWN0b3INCiRcc3RhdGV2ZWNfMCQgaXMgYXZhaWxhYmxlIGlu
IHRoZSBmb3JtIG9mIGFuIHVuYmlhc2VkIGVzdGltYXRlICRcaGF0e1xzdGF0ZXZlY31fMCQgd2hp
Y2ggaGFzIGJlZW4gZHJhd24gZnJvbQ0KYSBkaXN0cmlidXRpb24gd2l0aCBhIG1lYW4gb2YgJFxz
dGF0ZXZlY18wJCBhbmQgYSBkaXNwZXJzaW9uIG1hdHJpeCBvZiAkXGhhdHN0YXRldmFyXzAkLg0K
XGVuZHtxdW90ZX0NCg0KSXQgbWF5IGhhcHBlbiB0aGF0ICRcaGF0c3RhdGV2YXJfezB9IFx0byBc
aW5mdHkkIGZvciBkaWZmdXNlIHByaW9ycy4gSW4NCnRoaXMgY2FzZSwgd2UgbmVlZCB0byBjaGVj
ayBLb29wbWFuICgxOTk3KTsgaW4gdGhlIHN0YW5kYXJkIGNhc2UsDQpob3dldmVyLCB0aGUgcHJl
ZGljdGlvbiBlcXVhdGlvbnMgYXJlOg0KXGJlZ2lue2VxbmFycmF5fQ0KICBcbGFiZWx7ZXE6Zmls
dHN0YXRlc30NCiAgRV97dC0xfShcc3RhdGV2ZWNfdCkgPSBcaGF0e1xzdGF0ZXZlY31fe3R8dC0x
fSAmID0gJiANCiAgXHN0cm1hdF90IFxoYXR7XHN0YXRldmVjfV97dC0xfSArIFxtYXRoYmZ7Y31f
dCAgXGxhYmVse2VxOnN0YXRlcHJlZH1cXA0KICBWX3t0LTF9KFxzdGF0ZXZlY190KSA9IFxoYXRz
dGF0ZXZhcl97dHx0LTF9ICYgPSAmDQogIFxzdHJtYXRfdCBcaGF0c3RhdGV2YXJfe3QtMX0gXHN0
cm1hdCdfdCArIFxzdHJ2YXJfdCBcbGFiZWx7ZXE6UHByZWR9XFwNCiAgRV97dC0xfShcb2JzdmVj
X3QpID0gXGhhdHtcb2JzdmVjfV97dHx0LTF9ICYgPSAmIFxvYnNtYXRfdCBcaGF0e1xzdGF0ZXZl
Y31fe3R8dC0xfSArIFxtYXRoYmZ7ZH1fdCAgXFwNCiAgXHByZWRlcnJfdCAmID0gJiBcb2JzdmVj
X3QgLSBcaGF0e1xvYnN2ZWN9X3t0fHQtMX0gXFwNCiAgVl97dC0xfShccHJlZGVycl90KSA9IFxw
cmVkdmFyX3QgJiA9ICYgXG9ic21hdF90IFxoYXRzdGF0ZXZhcl97dHx0LTF9IFxvYnNtYXRfdCcg
KyBcb2JzdmFyX3QNCiBcZW5ke2VxbmFycmF5fQ0Kc28gdGhlIHVwZGF0aW5nIGVxdWF0aW9ucyBh
cmU6DQpcYmVnaW57ZXFuYXJyYXl9DQogIFxoYXR7XHN0YXRldmVjfV90ICYgPSAmIA0KICBcaGF0
e1xzdGF0ZXZlY31fe3R8dC0xfSArIA0KICBcbWF0aGJme0t9X3t0fSBccHJlZGVycl90IFxsYWJl
bHtlcTpzdGF0ZXVwZH1cXA0KICBcaGF0c3RhdGV2YXJfe3R9ICYgID0gJiANCiAgXGhhdHN0YXRl
dmFyX3t0fHQtMX0gLSANCiAgXGhhdHN0YXRldmFyX3t0fHQtMX0gXG9ic21hdF90JyBcbWF0aGJm
e0Z9XnstMX1fdCBcb2JzbWF0X3QgXGhhdHN0YXRldmFyX3t0fHQtMX0gXGxhYmVse2VxOlB1cGR9
DQpcZW5ke2VxbmFycmF5fQ0Kd2hlcmUgJFxtYXRoYmZ7S31fe3R9ID0gXGhhdHN0YXRldmFyX3t0
fHQtMX0gXG9ic21hdF90Jw0KXG9ic3Zhcl90XnstMX0kIGlzIHRoZSBLYWxtYW4gZ2Fpbi4gIEVx
dWF0aW9ucw0KKFxyZWZ7ZXE6c3RhdGVwcmVkfSktLShccmVme2VxOnN0YXRldXBkfSkgYW5kDQoo
XHJlZntlcTpQcHJlZH0pLS0oXHJlZntlcTpQdXBkfSkgY2FuIGJlIHdyaXR0ZW4gYXMgYSBzaW5n
bGUgc2V0IG9mDQpyZWN1cnNpb25zIGdvaW5nIGRpcmVjdGx5IGZyb20gJFxoYXR7XHN0YXRldmVj
fV97dC0xfSQgdG8gJFxoYXR7XHN0YXRldmVjfV90JA0KYW5kIGZyb20gJFxoYXRzdGF0ZXZhcl97
dC0xfSQgdG8gJFxoYXRzdGF0ZXZhcl97dH0kLg0KDQpOb3RlIHRoYXQgaW4gcHJhY3RpY2Ugd2Ug
bWF5DQp3YW50IHRvIGltcGxlbWVudCBzb21ldGhpbmcgY29tcHV0YXRpb25hbGx5IHNtYXJ0ZXIs
IGxpa2UgdGhlDQppbmZvcm1hdGlvbiBmaWx0ZXIgKHNvbWUgYmlibGlvZ3JhcGhpYyBzZWFyY2gg
aXMgbmVlZGVkIGhlcmUsIEkgY2FuJ3QNCnJlbWVtYmVyIGV4YWN0bHkgd2hhdCBpdCBsb29rcyBs
aWtlKS4NCg0KSSBjYW4ndCBiZSBib3RoZXJlZCByaWdodCBub3cgdG8gd3JpdGUgdGhlIHNtb290
aGluZyByZWN1cnNpb25zLCBidXQNCml0J3Mgbm90IHRlcnJpYmx5IGltcG9ydGFudC4NCg0KXHNl
Y3Rpb24qe1VzZXIgc3ludGF4fQ0KDQpUaGUgaWRlYSBpcyB0byBkZWZpbmUgYSBjb21tYW5kIGJs
b2NrIGxpa2Ugd2UgZG8gd2l0aCBcdGV4dHR0e21sZX0gb3INClx0ZXh0dHR7Z21tfToNClxiZWdp
bnt2ZXJiYXRpbX0NCmxpc3QgeSA9IHgxIHgyDQojIG9yIHBvc3NpYmx5DQptYXRyaXggWSA9IHsg
eDEgeDIgfQ0KDQprYWxtYW4gWyAtLXNtb290aCBdDQogIG9ic3kgeQ0KICBvYnNtYXQgSA0KICBv
YnNleCBkDQogIG9ic3ZhciBPbWVnYQ0KICBzdHJtYXQgUGhpDQogIHN0cmV4IGMNCiAgc3RydmFy
IFBzaQ0KICBpbmlzdGF0ZSB4MA0KICBpbmlwcmVjIFAwDQplbmQga2FsbWFuDQpcZW5ke3ZlcmJh
dGltfQ0Kd2hlcmUgXHRleHR0dHtvYnN5fSBzdGFuZHMgZm9yIGBgb2JzZXJ2YWJsZSAkXG9ic3Zl
YyQnJywNClx0ZXh0dHR7b2JzbWF0fSBzdGFuZHMgZm9yIGBgbWF0cml4IGluIHRoZSBvYnNlcnZh
dGlvbiBlcXVhdGlvbicnDQooXHJlZntlcTpvYnNlcX0pLCBcdGV4dHR0e29ic2V4fSBmb3IgYGBl
eG9nZW5vdXMgdGVybSBpbiB0aGUNCm9ic2VydmF0aW9uIGVxdWF0aW9uJycgYW5kIFx0ZXh0dHR7
b2JzdmFyfSBmb3IgYGB2YXJpYW5jZSBvZiB0aGUNCm9ic2VydmF0aW9uIGVxdWF0aW9uJycuIEEg
c2ltaWxhciBjb252ZW50aW9ucyB3b3VsZCBob2xkIGZvciB0aGUgc3RhdGUNCnRyYW5zaXRpb24g
ZXF1YXRpb24gKFxyZWZ7ZXE6c3RyYW5lcX0pLCB3aXRoIFx0ZXh0dHR7c3RyfSByZXBsYWNpbmcN
Clx0ZXh0dHR7b2JzfS4gVGhlIHN0cmluZ3MgXHRleHR0dHtpbmlzdGF0ZX0gYW5kIFx0ZXh0dHR7
aW5pcHJlY30NCnNob3VsZCBiZSBzZWxmLWV4cGxhbmF0b3J5Lg0KDQpUaGUgZWxlbWVudHMgXHRl
eHR0dHtvYnNleH0sIFx0ZXh0dHR7c3RyZXh9IGFuZCBcdGV4dHR0e2luaXN0YXRlfSBuZWVkDQpu
b3QgYmUgY29tcHVsc29yeS4gSWYgYWJzZW50LCB0aGV5IGFyZSB1bmRlcnN0b29kIHRvIGJlIDAu
IE1vcmUgb3INCmxlc3MgdGhlIHNhbWUgbWF5IGdvIGZvciBcdGV4dHR0e2luaXByZWN9LCBpZiB0
aGUgaW5pdGlhbGlzYXRpb24gaXMNCmRvbmUgdmlhIGEgZGlmZnVzZSBwcmlvciAodGhlIHNpbXBs
ZXN0IHRoaW5nIHdvdWxkIGJlIHRvIHVzZSBzb21laGluZw0KbGlrZSAkUF8wID0gMTBeezMwfVxj
ZG90IEkkLCBidXQgc21hcnRlciBtZXRob2RzIGFyZSBhdmFpbGFibGUgLS0NCmFnYWluLCBLb29w
bWFuICgxOTk3KSBzaG91bGQgY29tZSB0byB0aGUgcmVzY3VlKS4NCg0KRm9yIHRpbWUtaW52YXJp
YW50IG1vZGVscywgdGhlIGFyZ3VtZW50cyB0byBcdGV4dHR0e29ic21hdCwgc3RybWF0fQ0KZXRj
ZXRlcmEgbXVzdCBiZSBtYXRyaWNlcyBvZiB0aGUgYXBwcm9wcmlhdGUgc2l6ZS4gSW4gdGhlIGNh
c2Ugb2YNCnRpbWUtdmFyeWluZyBtb2RlbHMsIHRoZSBiZXN0IGlkZWEgSSd2ZSBiZWVuIGFibGUg
dG8gY29tZSB1cCB3aXRoIHNvDQpmYXJcZm9vdG5vdGV7QW5vdGhlciBzb2x1dGlvbiBpcyB0byBw
YXNzIG1hdHJpY2VzIHdpdGggJFQkIHJvd3MgYW5kIGFuDQogIGFwcHJvcHJpYXRlIG51bWJlciBv
ZiBjb2x1bW5zLCB3aGVyZSBlYWNoIHJvdyBpcyB0aGUgdmVjdG9yaXphdGlvbg0KICBvZiB0aGUg
Y29ycmVzcG9uZGluZyBtYXRyaXggYXQgdGltZSAkdCQ7IGZvciBleGFtcGxlLCBpZiAkXHN0cm1h
dF90JA0KICBpcyB0aW1lLXZhcnlpbmcsIHRoZW4gdGhlIGFyZ3VtZW50IHRvIFx0ZXh0dHR7c3Ry
bWF0fSB3aWxsIGhhdmUgdG8NCiAgYmUgYSAkVCBcdGltZXMgbV4yJCBtYXRyaXguIEl0J2xsIGJl
IGdyZXRsJ3Mgam9iIHRvIHJlc2hhcGUgdGhlDQogIHBlcnRpbmVudCByb3cgb2Ygc3VjaCBhIG1h
dHJpeCwgYnV0IHRoZSBtb3JlIEkgdGhpbmsgdGhlIG1vcmUgSSBmaW5kDQogIGl0IHVnbHkufSBp
cyB0byBwYXNzIG1hdHJpY2VzIHdoaWNoIHJlc3VsdCBmcm9tIHVzZXItZGVmaW5lZA0KZnVuY3Rp
b25zLiBUaGlzIGlzIHZlcnkgZ2VuZXJhbCBhbmQgZWxlZ2FudDsgc2V2ZXJhbCBpbnRlcmVzdGlu
Zw0KcG9zc2liaWxpdGllcyBhcmUgb3BlbmVkIGhlcmUsIHN1Y2ggYXMgYWxsb3dpbmcsIHNheSwg
dGhlIGVsZW1lbnRzIG9mDQokXG9ic21hdF90JCB0byBkZXBlbmQgb24gJFxwcmVkZXJyX3t0LTF9
LCBcb2JzdmVjX3t0LTF9JCwgc3RydWN0dXJhbA0KcGFyYW1ldGVycyAoYXMgY2FuIGJlIHRoZSBj
YXNlIHdpdGggRFNHRSBtb2RlbHMpIGFuZCBzbyBvbi4gDQoNCkhlbmNlLCB5b3Ugd291bGQgaGF2
ZSBzb21ldGhpbmcgbGlrZQ0KXFsNCiAgXG9ic21hdF90ID0gXG9ic21hdChcZXhvZ190LCBccGFy
YW0pLA0KXF0NCndoZXJlICRcb2JzbWF0JCBjb3JyZXNwb25kcyB0byBhIHVzZXItd3JpdHRlbiBm
dW5jdGlvbiBvZiB0aGUga2luZA0KXGJlZ2lue3ZlcmJhdGltfQ0KICAgIGZ1bmN0aW9uIE15T2Jz
TWF0KGxpc3QgWCwgbWF0cml4IHRoZXRhKQ0KXGVuZHt2ZXJiYXRpbX0NCndob3NlIGNvbnRlbnQg
ZGVwZW5kcyBvbiB0aGUgY29udGV4dCAocmVjdXJzaXZlIE9MUywgQVJNQXMsIHlvdSBuYW1lDQpp
dCksIHRvIGJlIGNhbGxlZCBhcw0KXGJlZ2lue3ZlcmJhdGltfQ0KICAgIGxpc3QgWiA9IHgxIHgy
IHgzDQogICAgbWF0cml4IHBhcm0gPSB7IDAuNSwgLTAuOCB9DQogICAga2FsbWFuDQogICAgICBv
YnNtYXQgTXlPYnNNYXQoWiwgcGFybSkNCiAgICAgIC4uLg0KICAgIGVuZCBrYWxtYW4NClxlbmR7
dmVyYmF0aW19DQoNCk9mIGNvdXJzZSwgb25lIGFsc28gbmVlZHMgdG8gcmV0cmlldmUgdGhlIGlu
dGVyZXN0aW5nIHF1YW50aXRpZXM6IHRoaXMNCmNvdWxkIGJlIGRvbmUgdmlhIHZpYSB0aGUgZXhp
c3RpbmcgYWNjZXNzb3JzIFx0ZXh0dHR7XCR1aGF0fSBmb3INCiRccHJlZGVycl90JCBhbmQgXHRl
eHR0dHtcJHNpZ21hfSAob3RoZXIgY2FuZGlkYXRlcyBtYXkgYmUNClx0ZXh0dHR7XCR2Y3Z9IGFu
ZCBcdGV4dHR0e1wkaH0gLS0tIGJ1dCBJIHBlcnNvbmFsbHkgbGlrZQ0KXHRleHR0dHtcJHNpZ21h
fSBiZXR0ZXIpIGZvciAkXHByZWR2YXJfdCQuIFdoZW4gJG4gPSAxJCwgdGhlc2Ugd291bGQgYmUN
CnNlcmllczsgb3RoZXJ3aXNlIFx0ZXh0dHR7XCR1aGF0fSB3b3VsZCBiZSBhICRUIFx0aW1lcyBu
JCBtYXRyaXggKGENCmJpdCBsaWtlIHdlIGRvIGZvciBzaW5nbGUgZXF1YXRpb24vbXVsdGlwbGUg
ZXF1YXRpb25zIG1vZGVscyksIHdoaWxlDQpcdGV4dHR0e1wkc2lnbWF9IHNob3VsZCBiZSBhICRU
IFx0aW1lcyBcZnJhY3tuIChuKzEpfXsyfSQgbWF0cml4LCB3aXRoDQp0aGUgJHQkLXRoIHJvdyBo
b2xkaW5nICRcbWF0aHJte3ZlY2h9KFxwcmVkdmFyX3QpJyQuIE5ldyBhY2Nlc3NvcnMNCnNob3Vs
ZCBiZSBkZXZvdGVkIHRvIHJldHVybmluZyB0aGUgc3RhdGVzIGFuZCB0aGVpciB2YXJpYW5jZXM6
DQpzdWl0YWJsZSBjYW5kaWRhdGVzIGNvdWxkIGJlIGFuZCBcdGV4dHR0e1wkc3RhdGVzfSBmb3Ig
JFxzdGF0ZXZlY190JCBhbmQNClx0ZXh0dHR7XCRzdHZhcnN9IGZvciAkXG1hdGhybXt2ZWNofShc
aGF0c3RhdGV2YXJfdCknJC4NCg0KTW9yZW92ZXIsIGl0IGlzIHByb2JhYmx5IHVzZWZ1bCB0byBh
bHNvIHJldHVybiB0aGUgc2VyaWVzDQpcWw0KICBcZWxsX3QgPSAtMC41IFxsZWZ0WyANCiAgICBu
IFxsbigyIFxwaSkgKyBcbGVmdHxccHJlZHZhcl90XHJpZ2h0fCArIA0KICAgIFxwcmVkZXJyX3Qn
XHByZWR2YXJfdF57LTF9IFxwcmVkZXJyX3QNCiAgXHJpZ2h0XSAsDQpcXQ0KcG9zc2libHkgaW4g
dGhlIGFjY2Vzc29yIFx0ZXh0dHR7XCRsbmx9LCB3aGljaCB3b3VsZCB5aWVsZCB0aGUNCmxpa2Vs
aWhvb2QuIEl0IHRoZW4gYmVjb21lcyB0cml2aWFsIHRvIGluc2VydCB0aGUgd2hvbGUgdGhpbmcg
aW4gYW4NClx0ZXh0dHR7bWxlfSBibG9jayB0byBlc3RpbWF0ZSBhbGwgc29ydHMgb2YgbW9kZWwu
IE5vdGUsIGhvd2V2ZXIsIHRoYXQNCnRoZSBcdGV4dHR0e2thbG1hbn0gYmxvY2sgaXRzZWxmIHdv
dWxkIHBlcmZvcm0gbm8gb3B0aW1pc2F0aW9uDQp3aGF0c29ldmVyOiBpdHMgam9iIHdvdWxkIG9u
bHkgYmUgdGhlIGZpbHRlcmluZyAob3IgdGhlIHNtb290aGluZywNCndpdGggdGhlIFx2ZXJifC0t
c21vb3RofCBvcHRpb24pLg0KDQpcc2VjdGlvbip7UmVmZXJlbmNlc30NCg0KSGFtaWx0b24sIEph
bWVzIEQuICgxOTk0KSwgXGVtcGh7VGltZSBTZXJpZXMgQW5hbHlzaXN9LCBQcmluY2V0b24NClVu
aXZlcnNpdHkgUHJlc3MuDQpcc21hbGxza2lwDQoNClxub2luZGVudA0KSGFydmV5LCBBLkMuICgx
OTkxKSwgXGVtcGh7Rm9yZWNhc3RpbmcsIFN0cnVjdHVyYWwgVGltZSBTZXJpZXMgTW9kZWxzDQog
IGFuZCB0aGUgS2FsbWFuIEZpbHRlcn0sIENhbWJyaWRnZSBVbml2ZXJzaXR5IFByZXNzLg0KXHNt
YWxsc2tpcA0KDQpcbm9pbmRlbnQNCkhhcnZleSwgQS5DLiBhbmQgVC4gUHJvaWV0dGkgKDIwMDUp
LCBcZW1waHtSZWFkaW5ncyBpbiBVbm9ic2VydmVkDQogIENvbXBvbmVudCBNb2RlbHN9LCBPeGZv
cmQgVW5pdmVyc2l0eSBQcmVzcy4NClxzbWFsbHNraXANCg0KXG5vaW5kZW50DQpLb29wbWFuLCBT
LiBKLiAoMTk5NyksIGBgRXhhY3QgSW5pdGlhbCBLYWxtYW4gRmlsdGVyaW5nIGFuZCBTbW9vdGhp
bmcNCmZvciBOb25zdGF0aW9uYXJ5IFRpbWUgU2VyaWVzIE1vZGVscycnLCBKb3VybmFsIG9mIHRo
ZSBBbWVyaWNhbg0KU3RhdGlzdGljYWwgQXNzb2NpYXRpb24sIFZvbC4gOTIsIE5vLiA0NDAsIHBw
LiAxNjMwLS0xNjM4IFxzbWFsbHNraXANCg0KXG5vaW5kZW50DQpQb2xsb2NrLCBELlMuRy4gKDE5
OTkpLCBcZW1waHtBIEhhbmRib29rIG9mIFRpbWUtU2VyaWVzIEFuYWx5c2lzLA0KICBTaWduYWwg
UHJvY2Vzc2luZyBhbmQgRHluYW1pY3N9LCBBY2FkZW1pYyBQcmVzcy4NCg0KDQpcZW5ke2RvY3Vt
ZW50fQ0KDQolJSUgTG9jYWwgVmFyaWFibGVzOiANCiUlJSBtb2RlOiBsYXRleA0KJSUlIFRlWC1t
YXN0ZXI6IHQNCiUlJSBFbmQ6IA0K

--===============3274299171833104426==--


From r.lucchetti@univpm.it Sat May 10 05:26:00 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] kalman filter status?
Date: Sat, 10 May 2008 11:25:55 +0200
Message-ID: <alpine.DEB.1.10.0805101120470.27173@ec-4.econ.univpm.it>
In-Reply-To: alpine.LRH.1.10.0805092320271.30820@ricardo.ecn.wfu.edu
Content-Type: multipart/mixed; boundary="===============5866647211392915187=="

--===============5866647211392915187==
Content-Type: text/x-tex
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="Kalman.tex"
MIME-Version: 1.0
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--===============5866647211392915187==--


From cottrell@wfu.edu Wed May 14 15:23:49 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] Quantile regression in gretl
Date: Wed, 14 May 2008 15:23:33 -0400
Message-ID: <alpine.LRH.1.10.0805141520150.27842@ricardo.ecn.wfu.edu>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4677744741793467920=="

--===============4677744741793467920==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

I'm not finished with this yet, but I'm far enough along that it 
would be good to get some comments.

There's a document describing the current state (in CVS and 
Windows snapshot) at

http://www.ecn.wfu.edu/~cottrell/gretl/QReg.pdf

Allin.

--===============4677744741793467920==--


From andreas.rosenblad@ltv.se Thu May 15 03:34:49 2008
From: andreas.rosenblad@ltv.se
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] Bug in the new quantile regression function
Date: Thu, 15 May 2008 09:34:43 +0200
Message-ID: <OFF7621856.546E3E51-ONC125744A.0028B298-C125744A.0029A1C3@ltv.se>
In-Reply-To: OF68E931FE.2D4EC052-ONC1257449.006AAF49@LocalDomain
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2556904729305818787=="

--===============2556904729305818787==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


First, thanks for the new quantile regression function in gretl. It looks
great.

However, I found a bug (using the Windows snapshot from 2008-05-14):

1. Using the GUI, open the Ramanathan data file "data 4-1".
2. Choose Model > Robust estimation > Quantile regression...
3. Set price as dependent and srft, bedrms, baths as independent
4. Set 0.25 0.50 0.75 as desired quantiles
5. Press OK: The program freezes

Best regards
Andreas


--===============2556904729305818787==--


From andreas.rosenblad@ltv.se Thu May 15 05:06:15 2008
From: andreas.rosenblad@ltv.se
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] Another quantile regression bug
Date: Thu, 15 May 2008 11:06:10 +0200
Message-ID: <OFE2D55D31.0F8B28D9-ONC125744A.00309AC2-C125744A.00320133@ltv.se>
In-Reply-To: OF68E931FE.2D4EC052-ONC1257449.006AAF49@LocalDomain
Content-Type: multipart/mixed; boundary="===============0598698127302432434=="

--===============0598698127302432434==
Content-Type: application/pdf
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="graph.pdf"
MIME-Version: 1.0
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--===============0598698127302432434==--


From cottrell@wfu.edu Thu May 15 10:16:46 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Bug in the new quantile regression function
Date: Thu, 15 May 2008 10:16:34 -0400
Message-ID: <alpine.LRH.1.10.0805151001360.13445@ricardo.ecn.wfu.edu>
In-Reply-To: OFF7621856.546E3E51-ONC125744A.0028B298-C125744A.0029A1C3@ltv.se
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4666651943539846276=="

--===============4666651943539846276==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Thu, 15 May 2008, andreas.rosenblad(a)ltv.se wrote:

> First, thanks for the new quantile regression function in gretl. 
> It looks great.

Glad you like it!

> However, I found a bug (using the Windows snapshot from 2008-05-14):
> 
> 1. Using the GUI, open the Ramanathan data file "data 4-1".
> 2. Choose Model > Robust estimation > Quantile regression...
> 3. Set price as dependent and srft, bedrms, baths as independent
> 4. Set 0.25 0.50 0.75 as desired quantiles
> 5. Press OK: The program freezes

Thanks for the report.  The freeze is happening inside the Fortran 
routine rqbr.  I'll try to figure out what's going on.

Allin.

--===============4666651943539846276==--


From cottrell@wfu.edu Thu May 15 10:39:11 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Bug in the new quantile regression function
Date: Thu, 15 May 2008 10:39:04 -0400
Message-ID: <alpine.LRH.1.10.0805151037190.13445@ricardo.ecn.wfu.edu>
In-Reply-To: OFF7621856.546E3E51-ONC125744A.0028B298-C125744A.0029A1C3@ltv.se
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5460173643533256441=="

--===============5460173643533256441==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Thu, 15 May 2008, andreas.rosenblad(a)ltv.se wrote:

> 1. Using the GUI, open the Ramanathan data file "data 4-1".
> 2. Choose Model > Robust estimation > Quantile regression...
> 3. Set price as dependent and srft, bedrms, baths as independent
> 4. Set 0.25 0.50 0.75 as desired quantiles
> 5. Press OK: The program freezes

Well, we're replicating R OK -- it goes into an infinite loop on 
this problem too!

Allin.

--===============5460173643533256441==--


From cottrell@wfu.edu Thu May 15 12:01:12 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Another quantile regression bug
Date: Thu, 15 May 2008 12:00:54 -0400
Message-ID: <alpine.LRH.1.10.0805151147240.13445@ricardo.ecn.wfu.edu>
In-Reply-To: OFE2D55D31.0F8B28D9-ONC125744A.00309AC2-C125744A.00320133@ltv.se
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============6982062710717879852=="

--===============6982062710717879852==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Thu, 15 May 2008, andreas.rosenblad(a)ltv.se wrote:

> I have found another bug for the quantile regression function 
> (again using the Windows snapshot from 2008-05-14):
> 
> 1. Using the GUI, open the Ramanathan data file "data 4-1".
> 2. Choose Model > Robust estimation > Quantile regression...
> 3. Set price as dependent and srft, bedrms, baths as independent
> 4. .1 .2 .3 .4 .5 .6 .7 .8 .9 as desired quantiles
> 
> Output...
> 
>       VARIABLE      TAU    COEFFICIENT      LOWER        UPPER
> 
>   const              0.1      1.13915     -1.#IND0     -1.#IND0
>                      0.2    -0.856990     -1.#IND0      296.952

Hmm, for tau = .10, R reports the confidence intervals as 
(-1.797693e+308, 1.797693e+308) for all coefficients.  There is a 
certain lack of error checking in the underlying Fortran code.

I think this is now fixed in CVS and a new snapshot.  Thanks for
testing!

Allin.


--===============6982062710717879852==--


From svetosch@gmx.net Tue May 20 10:56:53 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] translation time?
Date: Tue, 20 May 2008 16:56:50 +0200
Message-ID: <20080520145650.149270@gmx.net>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============8829808345721762650=="

--===============8829808345721762650==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

I know I can update the (German) translation anytime, but I was still wondering whether any kind of deadline is approaching. I would like to catch up before the release of 1.7.5, but of course I would like to do it rather later than sooner (standard time discounting ;-)

cheers,
sven

--===============8829808345721762650==--


From cottrell@wfu.edu Tue May 20 11:02:11 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] translation time?
Date: Tue, 20 May 2008 11:02:02 -0400
Message-ID: <alpine.LRH.1.10.0805201059490.30408@ricardo.ecn.wfu.edu>
In-Reply-To: 20080520145650.149270@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2243513158913628364=="

--===============2243513158913628364==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Tue, 20 May 2008, Sven Schreiber wrote:

> I know I can update the (German) translation anytime, but I was 
> still wondering whether any kind of deadline is approaching. I 
> would like to catch up before the release of 1.7.5, but of 
> course I would like to do it rather later than sooner (standard 
> time discounting ;-)

Good point.  I'm thinking that we've now accumulated enough bug 
fixes that we really should release soon.

There are a couple of small things that need attention in relation 
to the translatable strings.  I'll try to fix those today, and 
will commit a new gretl.pot before long.

Allin.

--===============2243513158913628364==--


From svetosch@gmx.net Thu May 22 04:55:20 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] omit bug
Date: Thu, 22 May 2008 10:55:16 +0200
Message-ID: <483534F4.2080600@gmx.net>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============8429036882311736335=="

--===============8429036882311736335==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Hi,
there seems to be a new bug introduced after 1.7.4; as always, I can 
provide more details later and/or file a bug report, but maybe this is 
already enough, and before I forget about it, here goes:

In a script of ours, we have an "omit" test statement after a standard 
(OLS) estimation, but in between there are some other calculations 
("scalar" and "matrix", so effectively "genr" I guess).

This worked fine up to 1.7.4, but current snapshots complain at the 
"omit" line, saying that no estimate has been performed so far.

cheers,
sven

--===============8429036882311736335==--


From svetosch@gmx.net Thu May 22 05:01:09 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] kalman filter status?
Date: Thu, 22 May 2008 11:01:06 +0200
Message-ID: <48353652.1060705@gmx.net>
In-Reply-To: alpine.DEB.1.10.0805101120470.27173@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============6379550867776550881=="

--===============6379550867776550881==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Am 10.05.2008 11:25, Riccardo (Jack) Lucchetti schrieb:
> On Fri, 9 May 2008, Allin Cottrell wrote:
> 

>>
>> Jack produced some useful working notes on this back in February.
>> I'm taking the liberty of attaching his TeX source; comments are
>> welcome.
> 
> I'm attaching a revised version, which contains a quite cool idea (If I 
> may say so myself) I had some time ago, that is using the KF as a 
> simulation tool too. Comments very welcome.
> 

With some delay I have now read it, and I have nothing much to add. I 
like the idea about the user-defined function to specify the 
time-varying parameters.

It could be that there is some room for minor syntactical improvements, 
but I guess that is better discussed after the prototype implementations 
are done.

With respect to simulation, if it's not difficult to implement, then it 
certainly would be nice. Whether it's terribly important or useful I 
can't say.

cheers,
sven

--===============6379550867776550881==--


From cottrell@wfu.edu Thu May 22 09:07:04 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] omit bug
Date: Thu, 22 May 2008 09:06:57 -0400
Message-ID: <alpine.LRH.1.10.0805220903550.435@ricardo.ecn.wfu.edu>
In-Reply-To: 483534F4.2080600@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0075113891165241170=="

--===============0075113891165241170==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Thu, 22 May 2008, Sven Schreiber wrote:

> there seems to be a new bug introduced after 1.7.4; as always, I 
> can provide more details later and/or file a bug report, but 
> maybe this is already enough, and before I forget about it, here 
> goes:
> 
> In a script of ours, we have an "omit" test statement after a 
> standard (OLS) estimation, but in between there are some other 
> calculations ("scalar" and "matrix", so effectively "genr" I 
> guess).

Please send a test case, I can't replicate this.  For example, the 
following works fine:

<script>
open data9-7
list xlist = 0 2 3 4
# see 'omit' at end
ols 1 xlist
matrix X = {xlist}
matrix XTXinv = inv(X'X)
scalar s2 = $sigma * $sigma
matrix V = s2*XTXinv
matrix se = sqrt(diag(V))
printf "\n%.12E\n", se
# 'omit' matches ols above
omit 3 4
</script>

Allin.

--===============0075113891165241170==--


From r.lucchetti@univpm.it Fri May 23 08:36:53 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] R and odbc
Date: Fri, 23 May 2008 14:36:50 +0200
Message-ID: <alpine.DEB.1.10.0805231430350.15136@ec-4.econ.univpm.it>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============9143488602511845030=="

--===============9143488602511845030==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

Folks,

Allin and I have been working recently on documenting a bit more properly 
a few things that have been in CVS for a few weeks without being too much 
publicised: ODBC and gretl <-> R interaction.

If you compile the version of the manual you find in CVS, you'll find two 
new chapters on these topics. Comments are appreciated. If you could try 
the new features out, that would be even more appreciated (obviously, 
you'll need to build from CVS or use the win snapshot). I realise odbc is 
not everyone's cup of tea, but I expect that quite a few of us have R at 
least installed.


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============9143488602511845030==--


From svetosch@gmx.net Fri May 23 08:39:12 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] gtksourceview dev
Date: Fri, 23 May 2008 14:39:08 +0200
Message-ID: <4836BAEC.3000306@gmx.net>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2975438714408918352=="

--===============2975438714408918352==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

I am just building gretl cvs for the first time on Ubuntu 8.04 and have 
noticed that the configure stage went fine although apparently 
libgtksourceview-dev or somesuch was missing (got make error later).

After installing that package, everything is ok.

-sven

--===============2975438714408918352==--


From svetosch@gmx.net Fri May 23 08:57:49 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] omit bug
Date: Fri, 23 May 2008 14:57:46 +0200
Message-ID: <4836BF4A.5050903@gmx.net>
In-Reply-To: alpine.LRH.1.10.0805220903550.435@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4364043497293895667=="

--===============4364043497293895667==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Am 22.05.2008 15:06, Allin Cottrell schrieb:
> On Thu, 22 May 2008, Sven Schreiber wrote:
> 
>> there seems to be a new bug introduced after 1.7.4; as always, I 
>> can provide more details later and/or file a bug report, but 
>> maybe this is already enough, and before I forget about it, here 
>> goes:
>>
>> In a script of ours, we have an "omit" test statement after a 
>> standard (OLS) estimation, but in between there are some other 
>> calculations ("scalar" and "matrix", so effectively "genr" I 
>> guess).
> 
> Please send a test case, I can't replicate this.  For example, the 
> following works fine:

Ok, it seems to be related to calling a user-defined function after 
estimation and before the restriction test; here is a demo script:

<script>
open denmark

function trivial(void)
result = 1
return scalar result
end function

ols LRM 0 LRY
scalar check = trivial()
omit LRY
</script>

This gives the mentioned error on today's CVS. I can't guarantee right 
now that this demo script works with 1.7.4, but our real-world script 
(where we noticed the bug) did.

In this case there is obviously an easy workaround (move the function 
call after the omit), but I can imagine cases where you want to use 
'restrict' with some value that you need to calculate in a function 
beforehand. So I'm inclined to think it's a bug.

thanks,
sven

--===============4364043497293895667==--


From ignacio.diaz-emparanza@ehu.es Fri May 23 10:20:53 2008
From: Ignacio Diaz-Emparanza <ignacio.diaz-emparanza@ehu.es>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] R and odbc
Date: Fri, 23 May 2008 16:20:49 +0200
Message-ID: <200805231620.50109.ignacio.diaz-emparanza@ehu.es>
In-Reply-To: alpine.DEB.1.10.0805231430350.15136@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2751514036873010240=="

--===============2751514036873010240==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

El Friday 23 May 2008 14:36:50 Riccardo (Jack) Lucchetti escribi=C3=B3:
> Folks,
>
> Allin and I have been working recently on documenting a bit more properly
> a few things that have been in CVS for a few weeks without being too much
> publicised: ODBC and gretl <-> R interaction.
>
> If you compile the version of the manual you find in CVS, you'll find two
> new chapters on these topics. Comments are appreciated. If you could try
> the new features out, that would be even more appreciated (obviously,
> you'll need to build from CVS or use the win snapshot). I realise odbc is
> not everyone's cup of tea, but I expect that quite a few of us have R at
> least installed.

The improvements about R are quite interesting. Do you plan to allow these
at the command line as well? I mean, I think we need a command block=20

R_script
... (R commands)
end R

such that when gretl processes the R_script, the dataset is transfered in the=
=20
same way as we do it through the GUI (tools/start GNU R or through the new R =

script capabilities), and the commands inside the block are passed to R.



--=20
Ignacio Diaz-Emparanza =20
DEPARTAMENTO DE ECONOM=C3=8DA APLICADA III (ECONOMETR=C3=8DA Y ESTAD=C3=8DSTI=
CA)                                       =20
UPV/EHU
Avda. Lehendakari Aguirre, 83 | 48015 BILBAO
T.: +34 946013732 | F.: +34 946013754
www.et.bs.ehu.es=20
--
Para leer y escribir documentos en formato "open document"
(.ods, .odt, .odb, odp) utilice OpenOffice.org o tambi=C3=A9n=20
MS Office, instalando el "plugin":
http://www.sun.com/software/star/odf_plugin


--===============2751514036873010240==--


From cottrell@wfu.edu Fri May 23 12:45:26 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] gtksourceview dev
Date: Fri, 23 May 2008 12:45:18 -0400
Message-ID: <alpine.LRH.1.10.0805231241190.2802@ricardo.ecn.wfu.edu>
In-Reply-To: 4836BAEC.3000306@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============9108221076471124149=="

--===============9108221076471124149==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 23 May 2008, Sven Schreiber wrote:

> I am just building gretl cvs for the first time on Ubuntu 8.04 
> and have noticed that the configure stage went fine although 
> apparently libgtksourceview-dev or somesuch was missing (got 
> make error later).
> 
> After installing that package, everything is ok.

We're not using an installed copy of gtkourceview any more, but 
strictly the slimmed-down version that comes bundled with the 
gretl sources.

However, there was one residual issue with the gtksourceview 
build, depending on the version of the glibc regex.h header; I 
think this is now fixed in CVS.

Allin.

--===============9108221076471124149==--


From cottrell@wfu.edu Fri May 23 14:50:10 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] omit bug
Date: Fri, 23 May 2008 14:50:01 -0400
Message-ID: <alpine.LRH.1.10.0805231421450.2802@ricardo.ecn.wfu.edu>
In-Reply-To: 4836BF4A.5050903@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5702107795933779660=="

--===============5702107795933779660==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 23 May 2008, Sven Schreiber wrote:

> Am 22.05.2008 15:06, Allin Cottrell schrieb:
> > On Thu, 22 May 2008, Sven Schreiber wrote:
> > 
> > > In a script of ours, we have an "omit" test statement after 
> > > a standard (OLS) estimation, but in between there are some 
> > > other calculations ("scalar" and "matrix", so effectively 
> > > "genr" I guess).
>
> Ok, it seems to be related to calling a user-defined function 
> after estimation and before the restriction test; here is a demo 
> script:
> 
> <script>
> open denmark
> 
> function trivial(void)
> result = 1
> return scalar result
> end function
> 
> ols LRM 0 LRY
> scalar check = trivial()
> omit LRY
> </script>

Thanks, I now see (and remember!) what's happening. As you know, 
"omit" works on the "last model".  Sometime after the 1.7.4 
release I noticed that we had a problem with functions: if a 
function estimated a model, this would remain the "last model" on 
exit from the function, which is not what we want.

I then "fixed" this by simply NULLing out the last model on exit 
from a function, with the results you found.

I've now implemented the right fix, namely, when we enter a 
function we save a pointer to the current "last model" and on exit 
from a function we restore that pointer.

Should be OK in CVS and Windows snapshot.

Allin.

--===============5702107795933779660==--


From svetosch@gmx.net Fri May 23 16:27:37 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] omit bug
Date: Fri, 23 May 2008 22:27:33 +0200
Message-ID: <483728B5.9010201@gmx.net>
In-Reply-To: alpine.LRH.1.10.0805231421450.2802@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============1686508718255524980=="

--===============1686508718255524980==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Am 23.05.2008 20:50, Allin Cottrell schrieb:

> 
> I've now implemented the right fix, namely, when we enter a 
> function we save a pointer to the current "last model" and on exit 
> from a function we restore that pointer.
> 
> Should be OK in CVS and Windows snapshot.
> 

Yes it looks like it, thanks.

However, another new problem just appeared, the context is as follows:

First we do IV estimation:

tsls rpivstand endorhs d_predrhs  ; d_predrhs d1_wox --robust

Then we do various stuff:

 >accessing $coeff, $vcv>
<some 'printf'-ing, 'scalar', 'matrix' operations>
<a function call -- see the previous bug>
<some 'omit' tests>
<some 'restrict' tests>

and then we access the hausman test result and calculate a p-value by 
assuming one d.o.f. less than usual (don't ask why right now):

matrix hausmanresult = $hausman
pvalue X hausmanresult[1,2]-1 hausmanresult[1,1]

-- and this now gives an "invalid argument" error (it didn't in 1.7.4).

Note that the following reformulation of the last line provides a 
workaround, so it seems it has to do with "on-the-fly calculations" (if 
that's the right term):

temp=hausmanresult[1,2]-1
pvalue X temp hausmanresult[1,1]

BTW, on various occasions I noticed that arithmetic expressions cannot 
always be used in place of simple variables. But sometimes it does seem 
to work. Is there some rule when it should work, or is it sheer luck 
when it does?

Thanks,
Sven





--===============1686508718255524980==--


From cottrell@wfu.edu Fri May 23 16:44:31 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] omit bug
Date: Fri, 23 May 2008 16:44:17 -0400
Message-ID: <alpine.LRH.1.10.0805231633170.2802@ricardo.ecn.wfu.edu>
In-Reply-To: 483728B5.9010201@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============3181131404379649293=="

--===============3181131404379649293==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 23 May 2008, Sven Schreiber wrote:

> Am 23.05.2008 20:50, Allin Cottrell schrieb:
> > Should be OK in CVS and Windows snapshot.
> > 
> 
> Yes it looks like it, thanks.

Good.

> However, another new problem just appeared...
> 
> matrix hausmanresult = $hausman
> pvalue X hausmanresult[1,2]-1 hausmanresult[1,1]
> 
> -- and this now gives an "invalid argument" error (it didn't in 
> 1.7.4)...

Oof, this is to do with a change I introduced at Stefano's 
behest: not "casting" 1x1 matrix results to scalars automatically.  
I'll have to think about that some more.

> BTW, on various occasions I noticed that arithmetic expressions 
> cannot always be used in place of simple variables. But 
> sometimes it does seem to work. Is there some rule when it 
> should work, or is it sheer luck when it does?

In general, arbitrary arithmetic should work for arguments to 
built-in functions (or user-defined functions for that matter),
while it's not very likely to work for arguments to straight
"commands".  However, in this particular case using the pvalue 
function won't help, right now, since it's a matrix/scalar thing.

I'll try to fix this properly for 1.7.5.

Allin.

--===============3181131404379649293==--


From r.lucchetti@univpm.it Fri May 23 17:15:50 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] R and odbc
Date: Fri, 23 May 2008 23:15:47 +0200
Message-ID: <alpine.DEB.1.10.0805232314170.29575@ec-4.econ.univpm.it>
In-Reply-To: 200805231620.50109.ignacio.diaz-emparanza@ehu.es
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============1078383265516549754=="

--===============1078383265516549754==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

On Fri, 23 May 2008, Ignacio Diaz-Emparanza wrote:

> The improvements about R are quite interesting. Do you plan to allow these
> at the command line as well? I mean, I think we need a command block
>
> R_script
> ... (R commands)
> end R
>
> such that when gretl processes the R_script, the dataset is transfered in t=
he
> same way as we do it through the GUI (tools/start GNU R or through the new R
> script capabilities), and the commands inside the block are passed to R.

Very interesting idea. I'll think about it.


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Universit=C3=A0 Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============1078383265516549754==--


From r.lucchetti@univpm.it Fri May 23 17:32:18 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] gtksourceview dev
Date: Fri, 23 May 2008 23:32:16 +0200
Message-ID: <alpine.DEB.1.10.0805232325080.29575@ec-4.econ.univpm.it>
In-Reply-To: alpine.LRH.1.10.0805231241190.2802@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4980141610460599729=="

--===============4980141610460599729==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Fri, 23 May 2008, Allin Cottrell wrote:

> We're not using an installed copy of gtkourceview any more, but
> strictly the slimmed-down version that comes bundled with the
> gretl sources.
>
> However, there was one residual issue with the gtksourceview
> build, depending on the version of the glibc regex.h header; I
> think this is now fixed in CVS.

Allin,

due to my very limited familiarity with gtk I know little about that 
region of our source code (and even that little I'm not sure I 
understand).

I'm sure there are very good reasons to use an in-house version of 
gtksourceview, but generally speaking, there are many advantages from 
linking to an external library; could you please explain what the 
disadvantages are in this case?


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============4980141610460599729==--


From cottrell@wfu.edu Fri May 23 17:34:35 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] omit bug
Date: Fri, 23 May 2008 17:34:25 -0400
Message-ID: <alpine.LRH.1.10.0805231732280.15153@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.LRH.1.10.0805231633170.2802@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4862777875704302182=="

--===============4862777875704302182==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 23 May 2008, Allin Cottrell wrote:

> On Fri, 23 May 2008, Sven Schreiber wrote:
> > matrix hausmanresult = $hausman
> > pvalue X hausmanresult[1,2]-1 hausmanresult[1,1]
> > 
> > -- and this now gives an "invalid argument" error (it didn't in 
> > 1.7.4)...
> 
> Oof, this is to do with a change I introduced at Stefano's 
> behest: not "casting" 1x1 matrix results to scalars 
> automatically.  I'll have to think about that some more.

In CVS and the snapshot there's a fix which takes care of all the 
probability-related functions (including pvalue).  I'll search 
some more later for other functions that might be affected.

Allin.

--===============4862777875704302182==--


From cottrell@wfu.edu Fri May 23 17:47:59 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] gtksourceview dev
Date: Fri, 23 May 2008 17:47:45 -0400
Message-ID: <alpine.LRH.1.10.0805231735250.15153@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.DEB.1.10.0805232325080.29575@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============6323063126263412348=="

--===============6323063126263412348==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 23 May 2008, Riccardo (Jack) Lucchetti wrote:

> On Fri, 23 May 2008, Allin Cottrell wrote:
> 
> > We're not using an installed copy of gtkourceview any more, but
> > strictly the slimmed-down version that comes bundled with the
> > gretl sources.
> 
> I'm sure there are very good reasons to use an in-house version 
> of gtksourceview, but generally speaking, there are many 
> advantages from linking to an external library; could you please 
> explain what the disadvantages are in this case?

Good question.  

One of the advantages of using an external library is that you get 
to benefit, for free, from bug fixes as new versions of the 
library become available, and from new features (if you want to 
make use of them).

In the case of gtksourceview, however, the main line of 
development has now shifted to gtksourceview-2.0.  This has more 
dependencies, an incompatible API, and (if I understand 
correctly) an incompatible format for the ".lang" language 
definition files.  

Now, gtksourceview-1.0 is fine for our purposes; and I don't want 
to have to support both major versions (remember, this has to work 
on Windows and OSX too).  

In addition, we're fine with a subset of gtksourceview's 
functionality.  For example, sourceview will do automatic 
indentation and smart tabs, up to a point, but that's of little 
use to us because it can't know the gretl syntax as well as we do, 
and we're better with home-rolled versions of those functions.

If there are are more bug-fix releases for gtksourceview 1.0 I'll 
incorporate the updates.  And maybe some day we'll find a reason 
to shift to version 2.0 on all platforms.

Allin.

--===============6323063126263412348==--


From r.lucchetti@univpm.it Fri May 23 17:53:20 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] omit bug
Date: Fri, 23 May 2008 23:53:18 +0200
Message-ID: <alpine.DEB.1.10.0805232344470.29575@ec-4.econ.univpm.it>
In-Reply-To: alpine.LRH.1.10.0805231633170.2802@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============6501469394723584393=="

--===============6501469394723584393==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Fri, 23 May 2008, Allin Cottrell wrote:

> Oof, this is to do with a change I introduced at Stefano's
> behest: not "casting" 1x1 matrix results to scalars automatically.
> I'll have to think about that some more.

I can see Stefano's point, but there are also advantages in auto-casting a 
1x1 matrix to a scalar. By the way (Stefano, are you there?) if you really 
need to ensure that the result of an operation is a matrix, you can always 
force it via the "matrix" alias, as in

matrix x = foo(z)

So, in my view, there are pros and cons to either choice. I think we ought 
to reach a collective decision here. So my proposal is: everyone 
interested, please have your say now. (Myself, I'd say "auto-cast".) 
Allin, just wait for a few days and then whatever you decide is the law.


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============6501469394723584393==--


From futur.dorko@gmail.com Sat May 24 06:32:04 2008
From: Stefano Balietti <futur.dorko@gmail.com>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] null list as parameter
Date: Sat, 24 May 2008 12:32:00 +0200
Message-ID: <1c7d4a100805240332m47191b03l1d11873e9f41b4b7@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2577563680288524566=="

--===============2577563680288524566==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

I had already mentioned this bug, but I can't find where. Anyway, it
is still here. With a function declaration like this:

function foo ( series a, scalar b[0::1], scalar c[0:2:1], const list
d[null], scalar e[0:2:0])

I'm supposed to be able to invoke the function just with a simple:

foo(a)

where a is a a series obviously.

Instead, gretl wants at least:

foo(a,b,c,d)

because does not accept the absence of the list, whereas it should be
allowed by the [null] tag.

Cheers.

Stefano

--===============2577563680288524566==--


From svetosch@gmx.net Sat May 24 11:00:45 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] omit bug
Date: Sat, 24 May 2008 17:00:43 +0200
Message-ID: <48382D9B.6010307@gmx.net>
In-Reply-To: alpine.LRH.1.10.0805231633170.2802@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0480683326762508205=="

--===============0480683326762508205==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Am 23.05.2008 22:44, Allin Cottrell schrieb:
> On Fri, 23 May 2008, Sven Schreiber wrote:
> 

>> However, another new problem just appeared...
>>
>> matrix hausmanresult = $hausman
>> pvalue X hausmanresult[1,2]-1 hausmanresult[1,1]
>>
>> -- and this now gives an "invalid argument" error (it didn't in 
>> 1.7.4)...
> 
> Oof, this is to do with a change I introduced at Stefano's 
> behest: not "casting" 1x1 matrix results to scalars automatically.  
> I'll have to think about that some more.

first, one thing I don't understand here: the second term 
(hausmanresult[1,1]) doesn't seem problematic (see the workaround I 
mentioned, where it is unchanged). But that would mean it is treated as 
a scalar, no? Then why does the first term (hausmanresult[1,2]) give 
something different?

secondly, I think there are two different issues involved here:
1) The first one is what data type is returned by a function (built-in 
or user-defined).
2) The second issue is what happens if you pick elements from a matrix, 
as in the above example.

 From my experience with earlier discussions about those issues on the 
numpy (numerical python) mailing list, these issues can and should be 
decided separately.

With respect to the first issue, I have some sympathy with Stefano's 
view that the returned datatype should not depend on the specific matrix 
dimension in particular cases.

As for the second issue, I would say that explicitly indexing a single 
scalar element from a matrix --as in the above example-- should return a 
scalar. However, that would *not* mean that each indexing range which 
refers to a 1x1 submatrix should be returned as a scalar. For example, 
let's say somebody picks the first row of a matrix mymat by using 
mymat[1,] in an expression. In general the first row would not be just a 
single element, and as such mymat[1,] should always be a matrix, even if 
it turns out to be just a 1x1 matrix.


[and now continuing a different topic:]

> 
>> BTW, on various occasions I noticed that arithmetic expressions 
>> cannot always be used in place of simple variables. But 
>> sometimes it does seem to work. Is there some rule when it 
>> should work, or is it sheer luck when it does?
> 
> In general, arbitrary arithmetic should work for arguments to 
> built-in functions (or user-defined functions for that matter),
> while it's not very likely to work for arguments to straight
> "commands".  

In principle I think I understand, but I'm not sure the demarcation 
between commands and functions is very clear in gretl. What would that 
'pvalue' thing be for example?

-sven

--===============0480683326762508205==--


From cottrell@wfu.edu Sat May 24 13:33:13 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] gtksourceview dev
Date: Sat, 24 May 2008 13:33:01 -0400
Message-ID: <alpine.LRH.1.10.0805241322560.2831@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.LRH.1.10.0805231735250.15153@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============3016226298932272813=="

--===============3016226298932272813==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 23 May 2008, Allin Cottrell wrote:

> On Fri, 23 May 2008, Riccardo (Jack) Lucchetti wrote:
> 
> > I'm sure there are very good reasons to use an in-house version 
> > of gtksourceview, but generally speaking, there are many 
> > advantages from linking to an external library; could you please 
> > explain what the disadvantages are in this case?
> 
> Good question...

On second thoughts, my arguments for always using the bundled 
version of gtksourceview (which we use on Windows and OSX) were a 
bit lame.

The changes in API from 1.0 to 2.0 of this library were not all 
that substantial.  So I've figured out the required changes and 
reinstated the possibility of using an installed libgtksourceview, 
with support for version 2.0.  What should now happen on Linux at 
configure time is:

* If the option --without-gtksourceview is given, we don't look
  for an installed version and use the bundled one.  (Can be 
  useful if you're trying to avoid dragging extra dependencies
  into a build of gretl -- since gtksourceview 1.0 requires
  libgnomeprint and v 2.0 requires GTK 2.12.)

* Otherwise we look for version 2.0 and use it if found.

* Failing that we look for version 1.0 and use it if found.

* And failing that we fall back on the bundled version.

Please let me know of any build problems.

Allin.

--===============3016226298932272813==--


From futur.dorko@gmail.com Sat May 24 13:34:23 2008
From: Stefano Balietti <futur.dorko@gmail.com>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] Gretl and R integration
Date: Sat, 24 May 2008 19:34:20 +0200
Message-ID: <1c7d4a100805241034p26b2775dv77aa20c6862a4dc0@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============3569658774397040642=="

--===============3569658774397040642==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Some considerations about this new feature after a little using.

- Sometimes, when executing (non-interactively) an R script which
contains syntax errors Gretl crashes after informing the user about
the error.

Here an example is where I'm trying to estimate an APARCH with fGarch.

<i>
library(fGarch)

rr = gretldata[,"rr"]

a = garchFit(~aparch(1,1),data=rr,cond.dist="dged",trace=FALSE)
</i>

- Besides, Gretl does not keep trace of the opened R script, like it
does for normal *.inp files in the recent opened files list. I think
they are supposed to be listed there as well, maybe underlined with a
different color or font.

- Strangely, Gretl does not support the opening of an R script by
simply dragging and dropping the file within the program window. I
have to open it through the graphical menu.

- Last, when I have an R script window opened and I open somehow a new
dataset, the R script window is automatically closed. That is pretty
unfair.

That's all!

Bye

--===============3569658774397040642==--


From cottrell@wfu.edu Sat May 24 13:49:37 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] null list as parameter
Date: Sat, 24 May 2008 13:49:17 -0400
Message-ID: <alpine.LRH.1.10.0805241343560.2831@ricardo.ecn.wfu.edu>
In-Reply-To: 1c7d4a100805240332m47191b03l1d11873e9f41b4b7@mail.gmail.com
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5324679301860876177=="

--===============5324679301860876177==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sat, 24 May 2008, Stefano Balietti wrote:

> I had already mentioned this bug, but I can't find where. Anyway, it
> is still here. With a function declaration like this:
> 
> function foo ( series a, scalar b[0::1], scalar c[0:2:1], const list
> d[null], scalar e[0:2:0])
> 
> I'm supposed to be able to invoke the function just with a simple:
> 
> foo(a)
> 
> where a is a a series obviously.
> 
> Instead, gretl wants at least:
> 
> foo(a,b,c,d)

Could you post a complete test case, please?  I think there must 
be some special complication at work, because the following script 
works fine here:

<script>
function myfoo (series a, scalar b[0::1], scalar c[0:2:1], 
                const list d[null], scalar e[0:2:0])
  print a -o
  print d
  print b c e
end function

open data4-1
myfoo(sqft)
</script>

It prints:

<output>
? myfoo(sqft)

     Obs             a

       1          1065
       2          1254
       3          1300
       4          1577
       5          1600
       6          1750
       7          1800
       8          1870
       9          1935
      10          1948
      11          2254
      12          2600
      13          2800
      14          3000


              b =  1.00000
              c =  1.00000
              e =  0.00000
</output>

Allin.

--===============5324679301860876177==--


From cottrell@wfu.edu Sat May 24 14:55:54 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Sat, 24 May 2008 14:55:44 -0400
Message-ID: <alpine.LRH.1.10.0805241412440.2831@ricardo.ecn.wfu.edu>
In-Reply-To: 1c7d4a100805241034p26b2775dv77aa20c6862a4dc0@mail.gmail.com
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============3285403644297898480=="

--===============3285403644297898480==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sat, 24 May 2008, Stefano Balietti wrote:

> Some considerations about this new feature after a little using.
> 
> - Sometimes, when executing (non-interactively) an R script which
> contains syntax errors Gretl crashes after informing the user about
> the error...

Oops, the verbose message from R was overflowing our error-message 
buffer. Fixed in CVS.

> - Besides, Gretl does not keep trace of the opened R script, like it
> does for normal *.inp files in the recent opened files list. I think
> they are supposed to be listed there as well, maybe underlined with a
> different color or font.

That could perhaps be added.  It would require a fair number of 
changes.

> - Strangely, Gretl does not support the opening of an R script by
> simply dragging and dropping the file within the program window. I
> have to open it through the graphical menu.

Well, there's nothing very "strange" about that, it doesn't happen 
by magic ;-)  However, that's now coded in.

> - Last, when I have an R script window opened and I open somehow a new
> dataset, the R script window is automatically closed. That is pretty
> unfair.

OK, fixed in CVS.

Allin.

--===============3285403644297898480==--


From cottrell@wfu.edu Sat May 24 15:15:50 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] omit bug
Date: Sat, 24 May 2008 15:15:39 -0400
Message-ID: <alpine.LRH.1.10.0805241503340.2831@ricardo.ecn.wfu.edu>
In-Reply-To: 48382D9B.6010307@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============3074446892293343110=="

--===============3074446892293343110==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sat, 24 May 2008, Sven Schreiber wrote:

[ first, about cating 1x1 matrices to scalars, to which I'll reply 
later, then... ]

> > > BTW, on various occasions I noticed that arithmetic 
> > > expressions cannot always be used in place of simple 
> > > variables. But sometimes it does seem to work. Is there some 
> > > rule when it should work, or is it sheer luck when it does?
> > 
> > In general, arbitrary arithmetic should work for arguments to 
> > built-in functions (or user-defined functions for that 
> > matter), while it's not very likely to work for arguments to 
> > straight "commands".
> 
> In principle I think I understand, but I'm not sure the 
> demarcation between commands and functions is very clear in 
> gretl. What would that 'pvalue' thing be for example?

For the most part the distinction is pretty clear: a function is 
something where all the arguments are wrapped in parentheses and 
separated by commas, and a command is everything else.  But there 
are a few hybrid cases, and "pvalue" is one such.  Since we have a 
perfectly good pvalue function with the same semantics, we 
implement the "command" form by composing and executing a call to 
"genr"; this gives more flexibility on the form of the arguments.

Allin.

--===============3074446892293343110==--


From r.lucchetti@univpm.it Sat May 24 17:49:16 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Sat, 24 May 2008 23:49:13 +0200
Message-ID: <alpine.DEB.1.10.0805242335320.18194@ec-4.econ.univpm.it>
In-Reply-To: alpine.LRH.1.10.0805241412440.2831@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2141130878964932733=="

--===============2141130878964932733==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit


While we're on this: I just committed a few changes, mostly cosmetic, to 
the "export from R" machine:

1) the function gretl_export and the string gretl_dotdir were renamed to 
gretl.export and gretl.dotdir. I'm no R export, but it looks like more 
polite to use the dot rather than the underscore in R-land.

2) we had a naming problem when exporting a time-seried object with just 
one variable; now fixed (a bit clumsily, but as I said I'm no R expert).

We have an open problem: if some sort of subsampling is in effect on the 
gretl side, mostl likely the append mechanism won't work. This is due to 
the fact that I haven't been able to figure out how to stuff into the csv 
file row labels in a format that that gretl can digest, so I removed row 
labels altogether. An alternative possibility would be to intervene on the 
gretl side: in fact, the script below (which has nothing to do with R 
interaction _per se_) won't work

<script>
 	nulldata 10
 	x = normal()
 	smpl +1 -1
 	store @dotdir/foo.csv
 	delete x
 	append @dotdir/foo.csv
</script>

unless, of course, you get rid of the "smpl" command on line 3. I don't 
know if this is actually a problem. It certainly makes things more 
difficult when it comes to R interaction.


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============2141130878964932733==--


From futur.dorko@gmail.com Sat May 24 18:11:18 2008
From: Stefano Balietti <futur.dorko@gmail.com>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] null list as parameter
Date: Sun, 25 May 2008 00:11:15 +0200
Message-ID: <1c7d4a100805241511u16862897jfda0289a332665c9@mail.gmail.com>
In-Reply-To: alpine.LRH.1.10.0805241343560.2831@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4150579208840394641=="

--===============4150579208840394641==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Well, you are right that wasn't the problem. The problem was that the
null list was passed to the ols command and it caused the error.
Now the point is that the error message may mislead, and perhaps it
should be changed.

In fact, this way the  ols produces the error msg:

"Command has insufficient arguments"

If that happens to a function (the case I thought)  we have:

"*function_name*: not enough arguments
Command has insufficient arguments"

The best would be having a similar msg for the ols as well:

"*command_name*: not enough arguments
Command has insufficient arguments"

But that is far from being an urgent matter.

Cheers.

Stefano

--===============4150579208840394641==--


From cottrell@wfu.edu Sat May 24 20:57:03 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] null list as parameter
Date: Sat, 24 May 2008 20:56:52 -0400
Message-ID: <alpine.LRH.1.10.0805242051460.9028@ricardo.ecn.wfu.edu>
In-Reply-To: 1c7d4a100805241511u16862897jfda0289a332665c9@mail.gmail.com
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============6527030967062209701=="

--===============6527030967062209701==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sun, 25 May 2008, Stefano Balietti wrote:

> Well, you are right that wasn't the problem. The problem was 
> that the null list was passed to the ols command and it caused 
> the error. Now the point is that the error message may mislead, 
> and perhaps it should be changed...

Ah, there's a difference that should be resolved, between the 
gretlcli output and the GUI script output.

<script>
function myfoo (series a, const list x[null])
  ols a x
end function

open data4-1
myfoo(price)
</script>

Running the above with gretlcli produces:

<output>
Command has insufficient arguments
error on line 1 of function myfoo
> ols a x

gretlcli: error executing script: halting
> myfoo(price)
</output>

which seems to me fine.  But it's true, in the GUI the error 
message is not so informative.

Allin.

--===============6527030967062209701==--


From cottrell@wfu.edu Sat May 24 21:17:23 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Sat, 24 May 2008 21:17:13 -0400
Message-ID: <alpine.LRH.1.10.0805242109580.9028@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.DEB.1.10.0805242335320.18194@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============1107064845328119532=="

--===============1107064845328119532==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sat, 24 May 2008, Riccardo (Jack) Lucchetti wrote:

> While we're on this: I just committed a few changes, mostly 
> cosmetic, to the "export from R" machine:
> 
> 1) the function gretl_export and the string gretl_dotdir were 
> renamed to gretl.export and gretl.dotdir. I'm no R export, but 
> it looks like more polite to use the dot rather than the 
> underscore in R-land.

It seems to me a weird tic in the R language that '.' has no 
special syntactical charge in the middle of an identifier -- but 
let it not be said that we are lacking in politesse!

BTW, please watch out for some changes in CVS: I'm moving the 
R-file writing apparatus from gui_utils.c to a new file, 
lib/src/foreign.c, so as to support command-line invocation of 
chunks of external script, as per Ignacio's suggestion.

> 2) we had a naming problem when exporting a time-seried object 
> with just one variable; now fixed (a bit clumsily, but as I said 
> I'm no R expert).

If it works, then hey!

> We have an open problem: if some sort of subsampling is in 
> effect on the gretl side, mostl likely the append mechanism 
> won't work...

Append has some "issues".  That's been on my agenda for a while, 
but I'll see if I can do anything that might be useful in this 
context.

Allin.

--===============1107064845328119532==--


From cottrell@wfu.edu Sat May 24 21:40:17 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Sat, 24 May 2008 21:40:01 -0400
Message-ID: <alpine.LRH.1.10.0805242133350.9028@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.LRH.1.10.0805242109580.9028@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4003433837059527313=="

--===============4003433837059527313==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sat, 24 May 2008, Allin Cottrell wrote:

> BTW, please watch out for some changes in CVS: I'm moving the 
> R-file writing apparatus from gui_utils.c to a new file, 
> lib/src/foreign.c, so as to support command-line invocation of 
> chunks of external script, as per Ignacio's suggestion.

E.g., based on the example in the manual:

<script>
open bjg.gdt
foreign language=R
  # extract the log series 
  y <- gretldata[, "lg"]
  # estimate the model
  strmod <- StructTS(y)
  # save the fitted components
  compon <- as.ts(strmod$fitted)
  # save the estimated variances
  vars <- as.matrix(strmod$coef)
  # export into gretl's temp dir
  gretl.export(compon)
  gretl.export(vars)
end foreign
append @dotdir/compon.csv
vars = mread("@dotdir/vars.mat")
print vars
</script>

This works in CVS.  I'll do some proper testing tomorrow. 
("foreign language=ox" should be a possibility too, but it's not 
ready yet.)

Allin.

--===============4003433837059527313==--


From cottrell@wfu.edu Sat May 24 22:22:57 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Sat, 24 May 2008 22:22:41 -0400
Message-ID: <alpine.LRH.1.10.0805242219160.9028@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.DEB.1.10.0805242335320.18194@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============3620878565260773286=="

--===============3620878565260773286==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sat, 24 May 2008, Riccardo (Jack) Lucchetti wrote:

> [T]he script below (which has nothing to do with R interaction 
> _per se_) won't work
> 
> <script>
> 	nulldata 10
> 	x = normal()
> 	smpl +1 -1
> 	store @dotdir/foo.csv
> 	delete x
> 	append @dotdir/foo.csv
> </script>

I think this problem is specific to unstructured data.  If you 
stick in, for example, "setobs 1 1950" after the "nulldata" 
command, the script works OK.  I'll take a further look at this.

Allin.



--===============3620878565260773286==--


From r.lucchetti@univpm.it Sun May 25 02:48:34 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Sun, 25 May 2008 08:48:31 +0200
Message-ID: <alpine.DEB.1.10.0805250844330.21109@ec-4.econ.univpm.it>
In-Reply-To: alpine.LRH.1.10.0805242133350.9028@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0913475780270860562=="

--===============0913475780270860562==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Sat, 24 May 2008, Allin Cottrell wrote:

> This works in CVS.  I'll do some proper testing tomorrow.
> ("foreign language=ox" should be a possibility too, but it's not
> ready yet.)

Well, we do have a "special relationship" with R, so generalising this 
(which may include octave, gnuplot... whatever) requires a bit of 
planning. For example, the way we launch R is hardwired. In the case of, 
say, ox, the user should have a way to tell gretl where the executable is, 
and so on. I'd say, let's focus on R right now and postpone the 
generalisation for 1.7.5++.

Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============0913475780270860562==--


From r.lucchetti@univpm.it Sun May 25 02:53:18 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Sun, 25 May 2008 08:53:10 +0200
Message-ID: <alpine.DEB.1.10.0805250850500.21109@ec-4.econ.univpm.it>
In-Reply-To: alpine.LRH.1.10.0805242219160.9028@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4673314891339647982=="

--===============4673314891339647982==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Sat, 24 May 2008, Allin Cottrell wrote:

> I think this problem is specific to unstructured data.  If you
> stick in, for example, "setobs 1 1950" after the "nulldata"
> command, the script works OK.  I'll take a further look at this.

Hmm, not quite. For example:

<script>
       nulldata 10
       setobs 4 1990:1
       x = normal()
       smpl +1 -1
       store @dotdir/foo.csv --omit-obs
       delete x
       append @dotdir/foo.csv
</script>

My point is that we ought to find a way to do without the row labels if 
the number of observations in the current sample matches the number of 
rows in the csv file.

Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============4673314891339647982==--


From cottrell@wfu.edu Sun May 25 10:49:52 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Sun, 25 May 2008 10:49:48 -0400
Message-ID: <alpine.LRH.1.10.0805251049050.26272@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.DEB.1.10.0805250844330.21109@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2492080796740252990=="

--===============2492080796740252990==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sun, 25 May 2008, Riccardo (Jack) Lucchetti wrote:

> On Sat, 24 May 2008, Allin Cottrell wrote:
> 
> > This works in CVS.  I'll do some proper testing tomorrow.
> > ("foreign language=ox" should be a possibility too, but it's not
> > ready yet.)
> 
> Well, we do have a "special relationship" with R, so 
> generalising this (which may include octave, gnuplot... 
> whatever) requires a bit of planning. For example, the way we 
> launch R is hardwired. In the case of, say, ox, the user should 
> have a way to tell gretl where the executable is, and so on. I'd 
> say, let's focus on R right now and postpone the generalisation 
> for 1.7.5++.

Yes, agreed.

Allin.


--===============2492080796740252990==--


From cottrell@wfu.edu Sun May 25 11:07:48 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Sun, 25 May 2008 11:07:40 -0400
Message-ID: <alpine.LRH.1.10.0805251105100.26272@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.LRH.1.10.0805242133350.9028@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2785892284853441575=="

--===============2785892284853441575==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sat, 24 May 2008, Allin Cottrell wrote:

> <script>
> open bjg.gdt
> foreign language=R
...
> end foreign

"foreign" now has two options: --send-data, which does what you 
think, and --quiet, which suppresses non-error printed output from
the foreign program (i.e. R).

Allin.

--===============2785892284853441575==--


From r.lucchetti@univpm.it Mon May 26 08:58:53 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Mon, 26 May 2008 14:58:50 +0200
Message-ID: <alpine.DEB.1.10.0805261457380.14689@ec-4.econ.univpm.it>
In-Reply-To: alpine.LRH.1.10.0805251105100.26272@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0965628955615373096=="

--===============0965628955615373096==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Sun, 25 May 2008, Allin Cottrell wrote:

> On Sat, 24 May 2008, Allin Cottrell wrote:
>
>> <script>
>> open bjg.gdt
>> foreign language=R
> ...
>> end foreign
>
> "foreign" now has two options: --send-data, which does what you
> think, and --quiet, which suppresses non-error printed output from
> the foreign program (i.e. R).

This is now in the user's manual (a bit concisely, perhaps) in the R 
chapter.

Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============0965628955615373096==--


From cottrell@wfu.edu Mon May 26 13:14:29 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Mon, 26 May 2008 13:14:25 -0400
Message-ID: <alpine.LRH.1.10.0805261313440.20845@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.DEB.1.10.0805242335320.18194@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2809734177033489311=="

--===============2809734177033489311==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Sat, 24 May 2008, Riccardo (Jack) Lucchetti wrote:

> We have an open problem: if some sort of subsampling is in 
> effect on the gretl side, mostl likely the append mechanism 
> won't work...

Is this now better with current CVS?

Allin.

--===============2809734177033489311==--


From cottrell@wfu.edu Mon May 26 13:18:00 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Gretl and R integration
Date: Mon, 26 May 2008 13:17:39 -0400
Message-ID: <alpine.LRH.1.10.0805261316510.20845@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.DEB.1.10.0805261457380.14689@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2725410220778126970=="

--===============2725410220778126970==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Mon, 26 May 2008, Riccardo (Jack) Lucchetti wrote:

> On Sun, 25 May 2008, Allin Cottrell wrote:
> > 
> > "foreign" now has two options: --send-data, which does what you
> > think, and --quiet, which suppresses non-error printed output from
> > the foreign program (i.e. R).
> 
> This is now in the user's manual (a bit concisely, perhaps) in 
> the R chapter.

Very nice.  Like the sample R-calling function.

Allin.

--===============2725410220778126970==--


From r.lucchetti@univpm.it Tue May 27 15:48:03 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] Proposal for a syntax change
Date: Tue, 27 May 2008 21:48:01 +0200
Message-ID: <alpine.DEB.1.10.0805272140350.6994@ec-4.econ.univpm.it>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2612663307906511311=="

--===============2612663307906511311==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit


Hi all,

since release is near, I feel it may be a good time to propose a 
backward-incompatible change that IMHO sooner or later we'll have to make: 
originally, we had "|" and "&" for logical operations. Then, when matrices 
appeared, we used the same operators for concatenation (mostly, because 
that's what everybody else uses). At the same time, we also support "||" 
and "&&" for logical operators. This, I think, creates an ugly ambiguity 
which should be resolved by specialising "|"-"~" for concatenation and 
"||"-"&&" for logical operations.

Do you agree?


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============2612663307906511311==--


From r.lucchetti@univpm.it Tue May 27 15:58:18 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Proposal for a syntax change
Date: Tue, 27 May 2008 21:58:16 +0200
Message-ID: <alpine.DEB.1.10.0805272156110.6994@ec-4.econ.univpm.it>
In-Reply-To: alpine.DEB.1.10.0805272140350.6994@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============7242750305532100724=="

--===============7242750305532100724==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

On Tue, 27 May 2008, Riccardo (Jack) Lucchetti wrote:

>
> Hi all,
>
> since release is near, I feel it may be a good time to propose a=20
> backward-incompatible change that IMHO sooner or later we'll have to make: =

> originally, we had "|" and "&" for logical operations. Then, when matrices =

> appeared, we used the same operators for concatenation (mostly, because=20
> that's what everybody else uses). At the same time, we also support "||" an=
d=20
> "&&" for logical operators. This, I think, creates an ugly ambiguity which =

> should be resolved by specialising "|"-"~" for concatenation and "||"-"&&" =

> for logical operations.

Sorry to reply to myself, but by re-reading my own message perhaps it's=20
not clear that the ambiguity is only about "|" vs "||". However, for=20
consistency, "&" should be banned in favour of "&&".


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Universit=C3=A0 Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============7242750305532100724==--


From cottrell@wfu.edu Tue May 27 17:06:35 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Proposal for a syntax change
Date: Tue, 27 May 2008 17:06:28 -0400
Message-ID: <alpine.LRH.1.10.0805271701450.24106@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.DEB.1.10.0805272140350.6994@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============3994662153976864189=="

--===============3994662153976864189==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Tue, 27 May 2008, Riccardo (Jack) Lucchetti wrote:

> since release is near, I feel it may be a good time to propose a 
> backward-incompatible change that IMHO sooner or later we'll 
> have to make...

Since we already have some backward-incompatible changes this 
time, I'm inclined to agree the time is right for

OR:  "|" -> "||"
AND: "&" -> "&&"

If we do this I'll make an effort to provide what Sven very 
reasonably requested some time back, namely good error messages in 
case "|" or "&" seem to be used in their old sense, with a 
"redirect" to the currently accepted syntax, e.g.

 for AND, please use "&&"

Allin.

--===============3994662153976864189==--


From svetosch@gmx.net Wed May 28 10:42:16 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Proposal for a syntax change
Date: Wed, 28 May 2008 16:42:11 +0200
Message-ID: <20080528144211.43550@gmx.net>
In-Reply-To: alpine.LRH.1.10.0805271701450.24106@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4614357051127361114=="

--===============4614357051127361114==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


-------- Original-Nachricht --------
> Datum: Tue, 27 May 2008 17:06:28 -0400 (EDT)
> Von: Allin Cottrell <cottrell(a)wfu.edu>
> An: r.lucchetti(a)univpm.it, Gretl development <gretl-devel(a)lists.wfu.edu>
> Betreff: Re: [Gretl-devel] Proposal for a syntax change

> On Tue, 27 May 2008, Riccardo (Jack) Lucchetti wrote:
> 
> > since release is near, I feel it may be a good time to propose a 
> > backward-incompatible change that IMHO sooner or later we'll 
> > have to make...
> 
> Since we already have some backward-incompatible changes this 
> time, I'm inclined to agree the time is right for
> 
> OR:  "|" -> "||"
> AND: "&" -> "&&"
> 
> If we do this I'll make an effort to provide what Sven very 
> reasonably requested some time back, namely good error messages in 
> case "|" or "&" seem to be used in their old sense, with a 
> "redirect" to the currently accepted syntax, e.g.
> 
>  for AND, please use "&&"
> 

I believe that what I suggested was to have a warning period before the change is actually introduced. In this case it would mean to get the warnings into 1.7.5, and then some months later (1.7.6?) make the actual syntax change. 

Of course, nobody committed to my proposal, so I'm not claiming that it must be done that way. I still think it would be more user-friendly, though.

cheers,
sven

--===============4614357051127361114==--


From cottrell@wfu.edu Wed May 28 11:01:41 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Proposal for a syntax change
Date: Wed, 28 May 2008 11:01:33 -0400
Message-ID: <alpine.LRH.1.10.0805281053530.18423@ricardo.ecn.wfu.edu>
In-Reply-To: 20080528144211.43550@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0618249700521647366=="

--===============0618249700521647366==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Me, agreeing with Jack:

> > Since we already have some backward-incompatible changes this 
> > time, I'm inclined to agree the time is right for
> > 
> > OR:  "|" -> "||"
> > AND: "&" -> "&&"
> > 
> > If we do this I'll make an effort to provide what Sven very 
> > reasonably requested some time back, namely good error 
> > messages...

Sven:

> I believe that what I suggested was to have a warning period 
> before the change is actually introduced. In this case it would 
> mean to get the warnings into 1.7.5, and then some months later 
> (1.7.6?) make the actual syntax change.

In general that is good procedure, I agree.  On the other hand, in 
this case the "||" and "&&" forms for OR and AND (and only those) 
have been in the manual for over a year, and none of the sample 
scripts packaged with gretl use the old forms.

I did find one of the separately-packaged Stock/Watson scripts 
that used '&' for AND; I've now replaced that.

Allin.


--===============0618249700521647366==--


From andreas.rosenblad@ltv.se Thu May 29 06:05:16 2008
From: andreas.rosenblad@ltv.se
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] missing documentation
Date: Thu, 29 May 2008 12:05:12 +0200
Message-ID: <OF19476567.C3FF7759-ONC1257458.00366AC1-C1257458.00375FE6@ltv.se>
In-Reply-To: OF68E931FE.2D4EC052-ONC1257449.006AAF49@LocalDomain
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5813361441403073270=="

--===============5813361441403073270==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


The gretl changelog says "User-defined functions: allow a descriptive
string to accompany each parameter, for use in GUI". However, this function
is not documented in the User's Guide.

Best regards
Andreas


--===============5813361441403073270==--


From svetosch@gmx.net Thu May 29 11:53:35 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] Proposal for a syntax change
Date: Thu, 29 May 2008 17:53:28 +0200
Message-ID: <483ED178.6030807@gmx.net>
In-Reply-To: alpine.LRH.1.10.0805281053530.18423@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============8817783386458009109=="

--===============8817783386458009109==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Am 28.05.2008 17:01, Allin Cottrell schrieb:
> Me, agreeing with Jack:
> 
>>> Since we already have some backward-incompatible changes this 
>>> time, I'm inclined to agree the time is right for
>>>
>>> OR:  "|" -> "||"
>>> AND: "&" -> "&&"
>>>
>>> If we do this I'll make an effort to provide what Sven very 
>>> reasonably requested some time back, namely good error 
>>> messages...
> 
> Sven:
> 
>> I believe that what I suggested was to have a warning period 
>> before the change is actually introduced. In this case it would 
>> mean to get the warnings into 1.7.5, and then some months later 
>> (1.7.6?) make the actual syntax change.
> 
> In general that is good procedure, I agree.  On the other hand, in 
> this case the "||" and "&&" forms for OR and AND (and only those) 
> have been in the manual for over a year, and none of the sample 
> scripts packaged with gretl use the old forms.

Has it been a year already? Time goes by fast...

I'm not really opposed to this change, all I'm saying is that at some 
point gretl's syntax needs to stabilize if you want gretl scripting to 
become widespread. Or can you imagine such syntax changes in Matlab? But 
since matrix manipulation in gretl is still fairly new, I guess I agree 
that ongoing development is still more important than user-land stability.

So -- are there any other changes that are planned for after 1.7.5, such 
that warnings should be introduced now?

-sven




--===============8817783386458009109==--


From futur.dorko@gmail.com Fri May 30 05:06:23 2008
From: Stefano Balietti <futur.dorko@gmail.com>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] strange bug with loop and submatrixes
Date: Fri, 30 May 2008 11:06:17 +0200
Message-ID: <1c7d4a100805300206y79182f04x5696d5928f50f1f1@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5639031822711970674=="

--===============5639031822711970674==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

The operation of selection of submatrixes causes a seg-fault if
performed within a loop.

--

function getData ( matrix results, scalar request[0:8:0], bool verbose[0])

	matrix treat = {}

	coeff = cols(results)

	coeff

	# init values
	if request = 0
		treat = results[2,]

	# coeff
    elif request = 1
        treat = results[3,]

	# stderr
    elif request = 2
		treat = results[4,]

    # vcv
	elif request = 3
		treat = results[5:4+coeff,]		
    endif

	return matrix treat

end function

nulldata 10

matrix a = zeros(20,5)

cc = cols(a)

matrix aa = a[5:4+cc,]	

aa

aa = getData( a,3)

aa

# SEGFAULT
loop for i=0..3
	aa = getData( a, i)

	aa
end loop

--

Cheers

--===============5639031822711970674==--


From r.lucchetti@univpm.it Fri May 30 05:38:38 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] strange bug with loop and submatrixes
Date: Fri, 30 May 2008 11:38:10 +0200
Message-ID: <alpine.DEB.1.10.0805301133370.29465@ec-4.econ.univpm.it>
In-Reply-To: 1c7d4a100805300206y79182f04x5696d5928f50f1f1@mail.gmail.com
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0770971201423988002=="

--===============0770971201423988002==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Fri, 30 May 2008, Stefano Balietti wrote:

> The operation of selection of submatrixes causes a seg-fault if
> performed within a loop.

Not quite. The problem exists, but is more subtle than that. For example, 
if you substitute

 	loop for i=0..3
         	aa = getData( a, i)
         	aa
 	end loop

with

 	loop for i=0..3
         	aa = getData( a, i)
 	        print aa
 	end loop

or with

 	loop for i=0..3
         	aa = getData( a, $i)
 	        aa
 	end loop

your script works alright. I have no time to work on this today, but 
clearly something funny's going on here. Thanks for spotting this.


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============0770971201423988002==--


From r.lucchetti@univpm.it Fri May 30 05:44:00 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] missing documentation
Date: Fri, 30 May 2008 11:43:58 +0200
Message-ID: <alpine.DEB.1.10.0805301140590.29465@ec-4.econ.univpm.it>
In-Reply-To: OF19476567.C3FF7759-ONC1257458.00366AC1-C1257458.00375FE6@ltv.se
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============3386903784419606609=="

--===============3386903784419606609==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Thu, 29 May 2008, andreas.rosenblad(a)ltv.se wrote:

>
> The gretl changelog says "User-defined functions: allow a descriptive
> string to accompany each parameter, for use in GUI". However, this function
> is not documented in the User's Guide.

It is. It's the fifth-from-last paragraph in section 10.1. Maybe it should 
be a little more conspicuous or less concise; suggestions?

Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============3386903784419606609==--


From andreas.rosenblad@ltv.se Fri May 30 05:53:15 2008
From: andreas.rosenblad@ltv.se
To: gretl-devel@gretlml.univpm.it
Subject: Re: Re: [Gretl-devel] missing documentation
Date: Fri, 30 May 2008 11:53:10 +0200
Message-ID: <OF4C342EBC.D026D0D2-ONC1257459.0035EB10-C1257459.00364E5A@ltv.se>
In-Reply-To: OFC01F5B97.C63DC349-ONC1257459.0035ABB7@LocalDomain
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============3005297805689365343=="

--===============3005297805689365343==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


r.lucchetti(a)univpm.it @ INTERNET skrev 2008-05-30 11:46:13 :

> On Thu, 29 May 2008, andreas.rosenblad(a)ltv.se wrote:
>
> >
> > The gretl changelog says "User-defined functions: allow a descriptive
> > string to accompany each parameter, for use in GUI". However, this
function
> > is not documented in the User's Guide.
>
> It is. It's the fifth-from-last paragraph in section 10.1. Maybe it
should
> be a little more conspicuous or less concise; suggestions?
>

Sorry, I see it now. However, it would be good if it was given as example
of the use of this, e.g. function myfunc(series x "descriptive string
1",scalar y "descriptive string 2")

Also, an example function of this in chapter 10.5 would be nice.

Andreas


--===============3005297805689365343==--


From r.lucchetti@univpm.it Fri May 30 06:42:47 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: Re: [Gretl-devel] missing documentation
Date: Fri, 30 May 2008 12:42:45 +0200
Message-ID: <alpine.DEB.1.10.0805301242290.29465@ec-4.econ.univpm.it>
In-Reply-To: OF4C342EBC.D026D0D2-ONC1257459.0035EB10-C1257459.00364E5A@ltv.se
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============7304217428337696821=="

--===============7304217428337696821==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Fri, 30 May 2008, andreas.rosenblad(a)ltv.se wrote:

> Sorry, I see it now. However, it would be good if it was given as example
> of the use of this, e.g. function myfunc(series x "descriptive string
> 1",scalar y "descriptive string 2")
>
> Also, an example function of this in chapter 10.5 would be nice.

Done.

Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============7304217428337696821==--


From cottrell@wfu.edu Fri May 30 10:13:45 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: Re: [Gretl-devel] missing documentation
Date: Fri, 30 May 2008 10:13:36 -0400
Message-ID: <alpine.LRH.1.10.0805301012590.1768@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.DEB.1.10.0805301242290.29465@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============7624315304336454762=="

--===============7624315304336454762==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 30 May 2008, Riccardo (Jack) Lucchetti wrote:

> On Fri, 30 May 2008, andreas.rosenblad(a)ltv.se wrote:
> 
> > Sorry, I see it now. However, it would be good if it was given as example
> > of the use of this, e.g. function myfunc(series x "descriptive string
> > 1",scalar y "descriptive string 2")
> > 
> > Also, an example function of this in chapter 10.5 would be nice.
> 
> Done.

Great.  I've now modified the earlier section too, making it a 
little more explicit.

Allin.

--===============7624315304336454762==--


From svetosch@gmx.net Fri May 30 10:37:01 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] omit bug
Date: Fri, 30 May 2008 16:38:26 +0200
Message-ID: <48401162.3030209@gmx.net>
In-Reply-To: alpine.LRH.1.10.0805241503340.2831@ricardo.ecn.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5574944673237107006=="

--===============5574944673237107006==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Allin Cottrell schrieb:
> On Sat, 24 May 2008, Sven Schreiber wrote:
> 
> [ first, about cating 1x1 matrices to scalars, to which I'll reply 
> later, then... ]
> 

any news here about the 1x1 matrix treatment?

cheers,
sven


--===============5574944673237107006==--


From cottrell@wfu.edu Fri May 30 10:52:32 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] strange bug with loop and submatrixes
Date: Fri, 30 May 2008 10:52:22 -0400
Message-ID: <alpine.LRH.1.10.0805301051380.1768@ricardo.ecn.wfu.edu>
In-Reply-To: alpine.DEB.1.10.0805301133370.29465@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============7118507923516379102=="

--===============7118507923516379102==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 30 May 2008, Riccardo (Jack) Lucchetti wrote:

> On Fri, 30 May 2008, Stefano Balietti wrote:
> 
> > The operation of selection of submatrixes causes a seg-fault if
> > performed within a loop.
> 
> Not quite. The problem exists, but is more subtle than that...

Ah, this was a residual GENCOMPILE bug.  Fixed in CVS.

Allin.

--===============7118507923516379102==--


From cottrell@wfu.edu Fri May 30 14:32:29 2008
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] omit bug
Date: Fri, 30 May 2008 14:32:14 -0400
Message-ID: <alpine.LRH.1.10.0805301424580.1768@ricardo.ecn.wfu.edu>
In-Reply-To: 48401162.3030209@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============1799249470139987874=="

--===============1799249470139987874==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

On Fri, 30 May 2008, Sven Schreiber wrote:

> Allin Cottrell schrieb:
> > On Sat, 24 May 2008, Sven Schreiber wrote:
> > 
> > [ first, about cating 1x1 matrices to scalars, to which I'll reply later,
> > then... ]
> > 
> any news here about the 1x1 matrix treatment?

The current status is:

* If a genr expression "naturally" produces a matrix result, but 
the matrix is 1x1, and the user has not forced the issue by 
specifying a result-type of "scalar" or "matrix", then we leave 
the result as a 1x1 matrix.  (As per Stefano's suggestion.)

* But we've tried to make inter-conversion between 1x1 matrices 
and scalars (inside genr) more flexible, so that the type of
the result should make less difference to the user.

Allin.



--===============1799249470139987874==--


From svetosch@gmx.net Fri May 30 17:25:59 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] bug when using gmm
Date: Fri, 30 May 2008 23:27:24 +0200
Message-ID: <4840713C.3020401@gmx.net>
Content-Type: multipart/mixed; boundary="===============6791131252145228738=="

--===============6791131252145228738==
Content-Type: application/x-gretldata
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="ppi100excerpt.gdt"
MIME-Version: 1.0
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--===============6791131252145228738==
Content-Type: application/x-gretlscript
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="bug2.inp"
MIME-Version: 1.0

c2V0IGVjaG8gb2ZmDQpzZXQgbWVzc2FnZXMgb2ZmDQpzZXQgYmZnc190b2xlciAxZS03DQpzZXQg
aGFjX2xhZyBudzEJCQkJCSMgZW5zdXJlIGdyZXRsIGRlZmF1bHQNCnNldCBoYWNfa2VybmVsIGJh
cnRsZXR0ICMgZGl0bw0Kc2V0IGhhY19wcmV3aGl0ZW4gb2ZmCQkjIGRpdG8NCg0Kb3BlbiBwcGkx
MDBleGNlcnB0DQoNCiMgc29tZSBwcmVsaW1pbmFyeSBkZWZpbml0aW9uczoNCnNlcmllcyBsZF9p
cHNxdWFyZSA9IGxkX2lwXjINCg0KbGlzdCBlbmRvcmhzID0gcGlzcXVhcmUgcGljcm9zcyBwaWFu
biANCmxpc3QgZF9wcmVkcmhzID0gY29uc3QgcnBpdnN0YW5kKC0xKSBwaWFubigtMSkgcGlzcXVh
cmUoLTEpIA0KDQojIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMNCiMg
d2hhdCBhcmUgdGhlIGluc3RydW1lbnRzPyANCg0KIyBnZW5lcmFsIG5ldyBzZXR1cDsgd294ID0g
d2l0aG91dCBwcmUtZGV0LiByaHMtdmFyaWFibGVzDQpsaXN0IGQyX3dveCA9IHBpYW5uKC0yKSBw
aWNyb3NzKC0xKSBycGl2c3F1YXJlKC0yKSBycGl2Y3Jvc3MoLTEpIHJwaXZjcm9zcygtMikgbGRf
aXBzcXVhcmUoLTEpIGZmcigtMSkgZmZyKC0yKQ0KbGlzdCBkbmV3MiA9IGQyX3dveCBkX3ByZWRy
aHMNCiMjIyBlbmQgaW5zdHJ1bWVudCBkZWZpbml0aW9uIA0KDQojIEdNTSBwcmVsaW1pbmFyaWVz
Og0KbWF0cml4IGRuZXcybWF0ID0geyBkbmV3MiB9DQptYXRyaXggZG5ldzJ3ZWlnaHRzID0gaW52
KGRuZXcybWF0J2RuZXcybWF0KQ0KDQojIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMj
IyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMNCg0KcHJpbnQgIi0tLS0tLS0tLS0tLS0tLSBHTU0gbGlu
ZWFyaXplZCwgZGVsdGEgaW5zdC4gMiAtLS0tLS0tLS0tLS0tLS0tLS0tIg0KIyBzdGFydGluZyB2
YWx1ZXMgDQpzZXJpZXMgbXlyZXNpZHMgPSAwDQpzY2FsYXIgYTEgPSAwLjINCnNjYWxhciBhMiA9
IDANCnNjYWxhciBhMyA9IDAuMw0Kc2NhbGFyIGE0ID0gMA0Kc2NhbGFyIGE1ID0gMQ0Kc2NhbGFy
IGE2ID0gMQ0Kc2NhbGFyIGEwID0gLTENCg0KZ21tDQpzZXJpZXMgbXlyZXNpZHMgPSBycGl2c3Rh
bmQgIC0gYTAgLSBhMSpycGl2c3RhbmQoLTEpIFwgDQogIC0gYTIqcGlzcXVhcmUgLSBhMypwaWNy
b3NzIC0gYTQqcGlhbm4gXA0KICAtIGE1KnBpc3F1YXJlKC0xKSAtIGE2KnBpYW5uKC0xKSAgDQpv
cnRob2cgbXlyZXNpZHMgOyBkbmV3Mm1hdA0Kd2VpZ2h0cyBkbmV3MndlaWdodHMNCnBhcmFtcyBh
MCBhMiBhMyBhNCBhMSBhNSBhNiANCmVuZCBnbW0gLS10d28tc3RlcA0KDQoNCnByaW50ICItLS0t
LS0tLS0tLS0tLS0tIEdNTSBub24tbGluZWFyLCBkZWx0YSBpbnN0LiAyIC0tLS0tLS0tLS0tLS0t
Ig0KIyBzdGFydGluZyB2YWx1ZXMgDQpzZXJpZXMgbXlyZXNpZHMgPSAwDQpzY2FsYXIgYzEgPSAw
LjINCnNjYWxhciBjMiA9IDANCnNjYWxhciBjMyA9IDAuMw0Kc2NhbGFyIGM0ID0gMA0Kc2NhbGFy
IGMwID0gLTENCg0KZ21tDQpzZXJpZXMgbXlyZXNpZHMgPSBycGl2c3RhbmQgIC0gYzAgLSBjMSpy
cGl2c3RhbmQoLTEpIFwgDQogIC0gYzIqKChwaWFubiAtIGMzKnBpYW5uKC0xKSAtIGM0KV4yKSAg
DQpvcnRob2cgbXlyZXNpZHMgOyBkbmV3Mm1hdA0Kd2VpZ2h0cyBkbmV3MndlaWdodHMNCnBhcmFt
cyBjMCBjMiBjMyBjNCBjMSANCmVuZCBnbW0gLS10d28tc3RlcA0K

--===============6791131252145228738==--


From r.lucchetti@univpm.it Fri May 30 17:47:33 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] bug when using gmm
Date: Fri, 30 May 2008 23:47:30 +0200
Message-ID: <alpine.DEB.1.10.0805302341450.13425@ec-4.econ.univpm.it>
In-Reply-To: 4840713C.3020401@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============6142329261695749638=="

--===============6142329261695749638==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Fri, 30 May 2008, Sven Schreiber wrote:

> Hello,
>
> either this post contains a relatively stupid question or I've stumbled
> upon a strange bug in gretl. this is with the released 1.7.4 on windows
> (no snapshot). Since you would have asked me for it anyway, I'm
> providing a testcase right away.
>
> The attached script contains two gmm blocks, which AFAICS should be
> independent and not affecting each other. The funny thing is, if I
> comment out the first gmm block, the results of the second gmm
> estimation change. As I'm writing this, tired and everything, I feel
> sure that somebody is going to point out a programming mistake and make
> me feel embarrassed, but well, maybe it's a gretl bug after all.

Fortunately, it isn't.

When you do 2 or more steps in GMM, the weights matrix gets updated. 
Hence, in your script the second GMM blocks starts with the weights matrix 
out of the first one. If you insert the statement

matrix dnew2weights = inv(dnew2mat'dnew2mat)

between the two gmm blocks, you get that the second estimate is precisely 
equal to what would have been if you had skipped the first gmm block.


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============6142329261695749638==--


From svetosch@gmx.net Sat May 31 08:12:17 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] bug when using gmm
Date: Sat, 31 May 2008 14:13:40 +0200
Message-ID: <484140F4.4000705@gmx.net>
In-Reply-To: alpine.DEB.1.10.0805302341450.13425@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4521698563649380466=="

--===============4521698563649380466==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Riccardo (Jack) Lucchetti schrieb:

> 
> Fortunately, it isn't.
> 
> When you do 2 or more steps in GMM, the weights matrix gets updated. 
> Hence, in your script the second GMM blocks starts with the weights 
> matrix out of the first one. If you insert the statement
> 
> matrix dnew2weights = inv(dnew2mat'dnew2mat)
> 
> between the two gmm blocks, you get that the second estimate is 
> precisely equal to what would have been if you had skipped the first gmm 
> block.
> 

First of all, thanks a lot for the quick explanation!

However, I'm wondering whether the updated weight matrix shouldn't 
rather be stored in some internal gretl variable instead of what the 
user passed as the *initial* weights. Are there any other cases where a 
user-defined variable is implicitly altered by gretl?

IMHO at least it should be stated explicitly in the manual -- but I 
confess I haven't checked and I can very well imagine that it is already 
mentioned there.

thanks,
sven

--===============4521698563649380466==--


From r.lucchetti@univpm.it Sat May 31 09:18:49 2008
From: Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] bug when using gmm
Date: Sat, 31 May 2008 15:18:46 +0200
Message-ID: <alpine.DEB.1.10.0805311501120.14478@ec-4.econ.univpm.it>
In-Reply-To: 484140F4.4000705@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============7612581932258568596=="

--===============7612581932258568596==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

On Sat, 31 May 2008, Sven Schreiber wrote:

> However, I'm wondering whether the updated weight matrix shouldn't 
> rather be stored in some internal gretl variable instead of what the 
> user passed as the *initial* weights.

If you need something like that, you can do it "by hand":

matrix W0 = (whatever)
matrix W = W0

gmm
    ...
   weights W
    ...
end gmm --iterate

> Are there any other cases where a user-defined variable is implicitly 
> altered by gretl?

Lots. For example,

scalar a = 0
mle ...
    params a
end mle

Same goes for gmm and nls.

> IMHO at least it should be stated explicitly in the manual -- 
> but I confess I haven't checked and I can very well imagine that 
> it is already mentioned there.

Thanks for the idea. I'd already planned to do some work on the manual 
this weekend. I'll throw this in.

Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti

--===============7612581932258568596==--


From svetosch@gmx.net Sat May 31 11:46:58 2008
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] bug when using gmm
Date: Sat, 31 May 2008 17:48:23 +0200
Message-ID: <48417347.5070800@gmx.net>
In-Reply-To: alpine.DEB.1.10.0805311501120.14478@ec-4.econ.univpm.it
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4228489323116495990=="

--===============4228489323116495990==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Riccardo (Jack) Lucchetti schrieb:
> On Sat, 31 May 2008, Sven Schreiber wrote:
> 

> 
>> Are there any other cases where a user-defined variable is implicitly 
>> altered by gretl?
> 
> Lots. For example,
> 
> scalar a = 0
> mle ...
>    params a
> end mle
> 
> Same goes for gmm and nls.

well of course, there it's very obviuos -- so I meant _except_ with 
'params' :-)

BTW, this got me thinking: in some sense it might be more consistent to 
use the accessor $coeff also for gmm, mle, nls, and so forth. But since 
I'm not in favor of breaking backwards-compatibility (as you know), I 
guess I'm not proposing it seriously.

> 
>> IMHO at least it should be stated explicitly in the manual -- but I 
>> confess I haven't checked and I can very well imagine that it is 
>> already mentioned there.
> 
> Thanks for the idea. I'd already planned to do some work on the manual 
> this weekend. I'll throw this in.

thanks for the work!

-sven


--===============4228489323116495990==--