From svetosch@gmx.net Thu Feb 18 10:15:16 2010
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: [Gretl-devel] gmm restrict error
Date: Thu, 18 Feb 2010 16:14:25 +0100
Message-ID: <4B7D5951.1080007@gmx.net>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============8178847923643854408=="

--===============8178847923643854408==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

Hi,
I can't seem to use the new restrict after gmm feature.

<script>
open denmark
series y = normal()
series x = normal()

/* initialize stuff */
series e = 0
scalar beta = 0
matrix V = I(1)
/* proceed with estimation */
gmm
series e = y - x*beta
orthog e ; x
weights V
params beta
end gmm

restrict
    beta = 3
end restrict
</script>

gives

? restrict
? beta = 3
? end restrict
Restriktion:
  = 3.05173
Nicht genügend Speicher!

Fehler bei Skriptausführung: Stopp
> end restrict

= out of memory error!

thanks,
sven



--===============8178847923643854408==--


From cottrell@wfu.edu Thu Feb 18 17:11:09 2010
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] gmm restrict error
Date: Thu, 18 Feb 2010 17:11:08 -0500
Message-ID: <Pine.A41.4.58.1002181706590.2625858@f1n11.sp2net.wfu.edu>
In-Reply-To: 4B7D5951.1080007@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============5044517169116199016=="

--===============5044517169116199016==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


On Thu, 18 Feb 2010, Sven Schreiber wrote:

> I can't seem to use the new restrict after gmm feature.
>
> gmm
> series e = y - x*beta
> orthog e ; x
> weights V
> params beta
> end gmm
>
> restrict
>     beta = 3
> end restrict

You must refer to the parameters in the canonical way, as set out
in the help for "restrict".  In this case,

restrict
  b[1] = 3
end restrict

The error message was misleading, this is now fixed in CVS.

Allin.

--===============5044517169116199016==--


From svetosch@gmx.net Fri Feb 19 04:59:38 2010
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] gmm restrict error
Date: Fri, 19 Feb 2010 10:59:08 +0100
Message-ID: <4B7E60EC.6040700@gmx.net>
In-Reply-To: Pine.A41.4.58.1002181706590.2625858@f1n11.sp2net.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============8850681820688682490=="

--===============8850681820688682490==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Allin Cottrell schrieb:
> On Thu, 18 Feb 2010, Sven Schreiber wrote:
> 
>> I can't seem to use the new restrict after gmm feature.
>>
>> gmm
>> series e = y - x*beta
>> orthog e ; x
>> weights V
>> params beta
>> end gmm
>>
>> restrict
>>     beta = 3
>> end restrict
> 
> You must refer to the parameters in the canonical way, as set out
> in the help for "restrict".  In this case,
> 
> restrict
>   b[1] = 3
> end restrict
> 

Hm, ok, but given that here gretl's syntax allows (actually: forces) the
user to specify the names of the parameters, I think those names should
be valid throughout. Maybe b[beta] would be a compromise. Or the use of
b[i] is enforced even in the formulation of the equation (that's what
Eviews does [did?]).

I guess that applies also to NLS, and MLE.

> The error message was misleading, this is now fixed in CVS.
> 

thanks!

-sven

--===============8850681820688682490==--


From cottrell@wfu.edu Fri Feb 19 12:46:23 2010
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] gmm restrict error
Date: Fri, 19 Feb 2010 12:46:22 -0500
Message-ID: <Pine.A41.4.58.1002191238470.2621880@f1n11.sp2net.wfu.edu>
In-Reply-To: 4B7E60EC.6040700@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2645767767345109474=="

--===============2645767767345109474==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


On Fri, 19 Feb 2010, Sven Schreiber wrote:

> Allin Cottrell schrieb:
> > On Thu, 18 Feb 2010, Sven Schreiber wrote:
> >
> >> I can't seem to use the new restrict after gmm feature.
> >>
> >> gmm
> >> series e = y - x*beta
> >> orthog e ; x
> >> weights V
> >> params beta
> >> end gmm
> >>
> >> restrict
> >>     beta = 3
> >> end restrict
> >
> > You must refer to the parameters in the canonical way, as set out
> > in the help for "restrict".  In this case,
> >
> > restrict
> >   b[1] = 3
> > end restrict
>
> Hm, ok, but given that here gretl's syntax allows (actually: forces) the
> user to specify the names of the parameters, I think those names should
> be valid throughout. Maybe b[beta] would be a compromise.

b[beta] works OK. I'll think about how we might permit use of the
original (scalar) parameter names for NLS, MLE and GMM.

> Or the use of b[i] is enforced even in the formulation of the
> equation (that's what Eviews does [did?]).

That would be too constraining: you couldn't write the equation
using matrix notation.

Allin.

--===============2645767767345109474==--


From svetosch@gmx.net Fri Feb 19 15:07:46 2010
From: Sven Schreiber <svetosch@gmx.net>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] gmm restrict error
Date: Fri, 19 Feb 2010 21:07:02 +0100
Message-ID: <4B7EEF66.3060008@gmx.net>
In-Reply-To: Pine.A41.4.58.1002191238470.2621880@f1n11.sp2net.wfu.edu
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============6263145464759935766=="

--===============6263145464759935766==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

Allin Cottrell schrieb:
> On Fri, 19 Feb 2010, Sven Schreiber wrote:
> 
>> Allin Cottrell schrieb:
>>> On Thu, 18 Feb 2010, Sven Schreiber wrote:
>>>
>>>> I can't seem to use the new restrict after gmm feature.
>>>>
>>>> gmm
>>>> series e = y - x*beta
>>>> orthog e ; x
>>>> weights V
>>>> params beta
>>>> end gmm
>>>>
>>>> restrict
>>>>     beta = 3
>>>> end restrict
>>> You must refer to the parameters in the canonical way, as set out
>>> in the help for "restrict".  In this case,
>>>
>>> restrict
>>>   b[1] = 3
>>> end restrict
>> Hm, ok, but given that here gretl's syntax allows (actually: forces) the
>> user to specify the names of the parameters, I think those names should
>> be valid throughout. Maybe b[beta] would be a compromise.
> 
> b[beta] works OK. I'll think about how we might permit use of the
> original (scalar) parameter names for NLS, MLE and GMM.
> 

well as I said, b[beta] seems like an ok solution, so if that works
already, and with a more informative error message, maybe no need to
touch that.

-sven

--===============6263145464759935766==--


From cottrell@wfu.edu Fri Feb 19 15:27:31 2010
From: Allin Cottrell <cottrell@wfu.edu>
To: gretl-devel@gretlml.univpm.it
Subject: Re: [Gretl-devel] gmm restrict error
Date: Fri, 19 Feb 2010 15:27:31 -0500
Message-ID: <Pine.A41.4.58.1002191524140.2507198@f1n11.sp2net.wfu.edu>
In-Reply-To: 4B7EEF66.3060008@gmx.net
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============2764535444465431466=="

--===============2764535444465431466==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


On Fri, 19 Feb 2010, Sven Schreiber wrote:

> Allin Cottrell schrieb:
> > On Fri, 19 Feb 2010, Sven Schreiber wrote:
> >
> >> Allin Cottrell schrieb:
> >>> On Thu, 18 Feb 2010, Sven Schreiber wrote:
> >>>
> >>>> I can't seem to use the new restrict after gmm feature.
> >>>>
> >>>> gmm
> >>>> series e = y - x*beta
> >>>> orthog e ; x
> >>>> weights V
> >>>> params beta
> >>>> end gmm
> >>>>
> >>>> restrict
> >>>>     beta = 3
> >>>> end restrict
> >>> You must refer to the parameters in the canonical way, as set out
> >>> in the help for "restrict".  In this case,
> >>>
> >>> restrict
> >>>   b[1] = 3
> >>> end restrict
> >> Hm, ok, but given that here gretl's syntax allows (actually: forces) the
> >> user to specify the names of the parameters, I think those names should
> >> be valid throughout. Maybe b[beta] would be a compromise.
> >
> > b[beta] works OK. I'll think about how we might permit use of the
> > original (scalar) parameter names for NLS, MLE and GMM.
>
> well as I said, b[beta] seems like an ok solution, so if that works
> already, and with a more informative error message, maybe no need to
> touch that.

In fact, it was not too difficult to enable the use of the
original names of scalar parameters for MLE, NLS and GMM, in the
context of a "restrict" block. That's now in CVS. It looks better
that way, since when gretl prints the restriction (so you can
verify that it has been recognized correctly) it uses the
parameter names.

Allin

--===============2764535444465431466==--