Gretl.
by Gabriel Carvalho
Como faço para tirar meu e-mail desta lista?
Gabriel Carvalho
12 years, 8 months
Re: [Gretl-users] Gretl-users Digest, Vol 55, Issue 15
by Francisco López Herrera
Hello, by the way, Is it possible to estimate a Multigarch model in Gretl?...flh(Parameswara das)
>________________________________
>De: "gretl-users-request(a)lists.wfu.edu" <gretl-users-request(a)lists.wfu.edu>
>Para: gretl-users(a)lists.wfu.edu
>Enviado: Miércoles, 17 de agosto, 2011 11:00:03
>Asunto: Gretl-users Digest, Vol 55, Issue 15
>
>Send Gretl-users mailing list submissions to
> gretl-users(a)lists.wfu.edu
>
>To subscribe or unsubscribe via the World Wide Web, visit
> http://lists.wfu.edu/mailman/listinfo/gretl-users
>or, via email, send a message with subject or body 'help' to
> gretl-users-request(a)lists.wfu.edu
>
>You can reach the person managing the list at
> gretl-users-owner(a)lists.wfu.edu
>
>When replying, please edit your Subject line so it is more specific
>than "Re: Contents of Gretl-users digest..."
>
>
>Today's Topics:
>
> 1. Re: EGARCH parameter estimation via "gig" GUI
> (Riccardo (Jack) Lucchetti)
> 2. Re: New options for the script editor (Riccardo (Jack) Lucchetti)
> 3. Re: installing gretl in Ubuntu 11.04 (Riccardo (Jack) Lucchetti)
> 4. Re: Variance/covariance matrix via GARCH(1,1)
> (Riccardo (Jack) Lucchetti)
>
>
>----------------------------------------------------------------------
>
>Message: 1
>Date: Wed, 17 Aug 2011 15:43:24 +0200 (CEST)
>From: "Riccardo (Jack) Lucchetti" <r.lucchetti(a)univpm.it>
>Subject: Re: [Gretl-users] EGARCH parameter estimation via "gig" GUI
>To: Gretl list <gretl-users(a)lists.wfu.edu>
>Message-ID: <alpine.DEB.2.02.1108171539310.4497(a)ec-4.econ.univpm.it>
>Content-Type: text/plain; charset="utf-8"
>
>On Mon, 8 Aug 2011, Oscar Soppelsa wrote:
>
>> I was wondering the following: in the Exponential GARCH(1,1) model
>> there?s a parameter in addition to ?, ? and ?, and this is ?. gretl uses
>> the EGARCH(p,q) formulation without ? [? multiplies the residuals in
>> g(?t-1)]. If I need to make a forecast on t + n periods in the future, I
>> would need the ? estimated value: does anyone know how may I get that
>> value?
>
>Hmm; the parametrisation gig uses is reported in the pdf doc file as
>equation 9. I don't quite get what kind of parametrisation for the EGARCH
>model you're referring to. Could you be more explicit, please? If you
>don't feel like writing the specification you have in mind for the
>conditional variance, please provide a reference.
>
>
>Riccardo (Jack) Lucchetti
>Dipartimento di Economia
>Universit? Politecnica delle Marche
>
>r.lucchetti(a)univpm.it
>http://www.econ.univpm.it/lucchetti
>
>------------------------------
>
>Message: 2
>Date: Wed, 17 Aug 2011 15:50:51 +0200 (CEST)
>From: "Riccardo (Jack) Lucchetti" <r.lucchetti(a)univpm.it>
>Subject: Re: [Gretl-users] New options for the script editor
>To: Gretl list <gretl-users(a)lists.wfu.edu>
>Message-ID: <alpine.DEB.2.02.1108171545330.4497(a)ec-4.econ.univpm.it>
>Content-Type: text/plain; charset="iso-8859-15"
>
>On Tue, 9 Aug 2011, amaryl wrote:
>
>> On Mon, 2011-08-08 at 22:04 -0400, Allin Cottrell wrote:
>>> On Tue, 2 Aug 2011, Summers, Peter wrote:
>>>
>>>>> Sometimes a tool to forcibly terminate a running script would come very
>>>> handy, too. Maybe a combined "Start/Stop" button or something alike.
>>>>
>>>> I found myself looking for the "stop" button just yesterday ;-).
>>>
>>>> From a programmer's point of view, the question is: where should we
>>> stick the handlers for aborting a script? Is everyone who's putting
>>> forward this idea thinking in terms of breaking out of a gretl loop?
>>>
>>> Notice that you have a "break" command for loops, so with some
>>> foresight you can arrange a conditional breakout if things are not
>>> going well.
>>
>> Personally, I am not thinking of breaking out of faulty loops.
>> In one of my scripts I had a lot of time-consuming loops. When I wanted
>> to check changes in my script not directly related to the loops, I
>> usually commented them out, but occasionally I forgot. A stop-button
>> would have been useful. "break" obviously doesn't help me in that case.
>>
>> In general, I think a stop-button is useful in cases where you notice
>> that you have lacked foresight ;-)
>
>The problem with a "break" button is that the program has to constantly
>check if the button has been pressed and, if so, take the appropriate
>action. This slows down execution and I personally don't like it very
>much. This said, it has happened to me to long for a break button, like
>for example when you run some estimation command that is known to take
>minutes and you realise just after having pressed "OK" that you forgot to
>add a variable or so.
>
>What we may want to do is think about some mechanism for interrupting the
>current command in the GUI client, and advise users to use the unstoppable
>might of the CLI client if performance is of paramount importance.
>Thoughts?
>
>
>Riccardo (Jack) Lucchetti
>Dipartimento di Economia
>Universit? Politecnica delle Marche
>
>r.lucchetti(a)univpm.it
>http://www.econ.univpm.it/lucchetti
>
>------------------------------
>
>Message: 3
>Date: Wed, 17 Aug 2011 16:00:21 +0200 (CEST)
>From: "Riccardo (Jack) Lucchetti" <r.lucchetti(a)univpm.it>
>Subject: Re: [Gretl-users] installing gretl in Ubuntu 11.04
>To: Gretl list <gretl-users(a)lists.wfu.edu>
>Message-ID: <alpine.DEB.2.02.1108171556140.4497(a)ec-4.econ.univpm.it>
>Content-Type: text/plain; charset="iso-8859-15"
>
>On Thu, 11 Aug 2011, Neil Hepburn wrote:
>
>> Greetings I am trying to install Gretl 1.9.5 in Ubuntu 11.04 as per the
>> note on the Sourceforge site about using the Debian version instead of
>> the Ubuntu version. ?Can someone tell me the repository to use in my
>> /etc/apt/sources.list file?
>
>ftp://ftp.debian.org testing main
>
>should do the trick, I guess. However, IMHO Linux users are almost
>always better off building from source. I have no direct experience with
>ubuntu 11.04, but I believe there shouldn't be any problem.
>
>
>Riccardo (Jack) Lucchetti
>Dipartimento di Economia
>Universit? Politecnica delle Marche
>
>r.lucchetti(a)univpm.it
>http://www.econ.univpm.it/lucchetti
>
>------------------------------
>
>Message: 4
>Date: Wed, 17 Aug 2011 16:03:03 +0200 (CEST)
>From: "Riccardo (Jack) Lucchetti" <r.lucchetti(a)univpm.it>
>Subject: Re: [Gretl-users] Variance/covariance matrix via GARCH(1,1)
>To: Gretl list <gretl-users(a)lists.wfu.edu>
>Message-ID: <alpine.DEB.2.02.1108171601190.4497(a)ec-4.econ.univpm.it>
>Content-Type: text/plain; charset="iso-8859-15"
>
>On Thu, 11 Aug 2011, Oscar Soppelsa wrote:
>
>> May you tell me how may I estimate a variance/covariance matrix using
>> GARCH(1,1) model?
>
>I'm afraid I don't understand what you want to do. Do you need the
>covariance matrix of the estimated parameters or a multivariate garch
>model (or something else altogether)?
>
>
>Riccardo (Jack) Lucchetti
>Dipartimento di Economia
>Universit? Politecnica delle Marche
>
>r.lucchetti(a)univpm.it
>http://www.econ.univpm.it/lucchetti
>
>------------------------------
>
>_______________________________________________
>Gretl-users mailing list
>Gretl-users(a)lists.wfu.edu
>http://lists.wfu.edu/mailman/listinfo/gretl-users
>
>End of Gretl-users Digest, Vol 55, Issue 15
>*******************************************
>
>
>
12 years, 8 months
Dúvida Gretl ( Gretl Doubt)
by Rodrigo Otávio Veiga de Miranda
Olá,
Utilizo o software GRETL (1.9.5cvs) para ajustar sistemas de equações simultâneas, utilizando-se o método dos quadrados ordinários em 2 estágios. Observando-se o resultado fornecido pelo mesmo, nota-se que as estatísticas "R²" e "R²" ajustado são calculadas apenas para a primeira equação do sistema. Desta forma, teria como obter tais estatísticas para a(s) outra(s) equação(s) do sistema? Caso não tenha, fica esta sugestão.
Aguardo resposta.
Hello,
I am an user of GRETL (1.9.5cvs) to adjust simultaneous equations systems with the two stages least squares method (TSLS). By observing the program output, I noticed that the statistics "R²" and "R² adj." are only calculated for the first equation of the system. In this way, is it possible to obtain such statistics for the second (other) equation(s)? In case it isn't possible, it would be nice to see this feature in the next versions of the program.
Awaiting response.
Att.,
Rodrigo Otávio Veiga de Miranda
Mestrando em Ciências Florestais - UNICENTRO - PR
42 9955 6976
38 8813 5790
31 9953 4041
12 years, 8 months
Fixed effects forecast
by Ricardo Gonçalves Silva
Hi,
Can Gretl forecast 1 to 3 periods ahead an estimated panel data(fixed-effects with no iterations) model?
HTH
RIck
12 years, 8 months
Deleting variables: 1) inside a loop; 2) when they have more than N missing values
by Matteo Chinazzi
Dear Gretlers,
I would like to know how to do two specific operations which I am not able
to perform in a proper way:
1) delete variables using loop/for constructs
2) remove variables which contain only missing values and/or have less than
"N" valid observations
1) As far as the first problem is concerned, I partially solved it by
writing a file using outfile and printf but I guess it is not the most
efficient way to do it. That is, for the time being I solved it in the
following way:
loop for (i=1; i<100; i+=1)
> outfile --append foo.inp
> printf "delete %d\n", $i
> outfile --close
> endloop
> run foo.inp
However, I cannot directly use something like this:
loop for (i=1; i<100; i+=1)
> delete $i
> endloop
>
because I receive the following error message: "You cannot delete series in
this context" .
2) Besides the problem at point 1), I would like to know if there is a way
to delete variables using some sort of condition upon the number of
observations they contain. That is, what I have in mind is something like
this (N is a scalar):
loop for (v=1; v<100; v+=1)
> if (sum(missing($v))<N)
> delete $v
> endif
> endloop
However, I cannot do it because if I run "missing($v)" inside a loop it
interprets "$v" as a number and not as a reference to a variable. Therefore,
I would like to know if there is any way to tell Gretl "consider that number
as a reference to a variable, not as a scalar" since this would come handy
also in many other situations. Notice also that, in my case, I cannot build
a list containing all the (string) names of all the variables since I get an
error (probably because Gretl cannot handle two thousands strings... I
guess..).
Thanks in advance for your help,
Matteo
12 years, 8 months
Accessor's Problems
by Henrique Andrade
Dear Gretl Community,
I think the *$fcast* accessor has a problem. I'm trying to save my VAR's
forecast inside series (and not matrices), but all that I can get is an
empty series. Could you please take a look at my scripts?
<hansl>
### Script 1 ###
open australia.gdt
smpl ; 1989:4
var 4 E PAU PUS --quiet
fcast --out-of-sample
series E_hat = $fcast[,1]
smpl --full
print E_hat --byobs
</hansl>
Even if I build a matrix before the *series* command I get the empty series:
<hansl>
### Script 2 ###
open australia.gdt
smpl ; 1989:4
var 4 E PAU PUS --quiet
fcast --out-of-sample
matrix E_HAT = $fcast[,1]
series E_hat = E_HAT
smpl --full
print E_hat --byobs
</hansl>
Thanks in advance,
*Henrique C. de Andrade*
Doutorando em Economia Aplicada
Universidade Federal do Rio Grande do Sul
www.ufrgs.br/ppge
12 years, 8 months
Re: [Gretl-users] Restricted Panel Data Sample
by George Matysiak
*Thanks Leandro. I've worked-around the problem and defined a new restricted
(dependent) variable within the range (I'm looking to exclude outliers) and
imported this into the exiting panel data structure. QED.
George
*
12 years, 8 months
Re: [Gretl-users] Restricted Panel Data Sample
by George Matysiak
*When I use the Sample dialog 'Restrict, based on criterion, for example,
trr>-17 && trr<40.05, the Panel option under Model is greyed out and
unavailable. I'd be grateful if someone could point out why that may be.
Thanks. George
*
12 years, 8 months
KPSS test
by samyto ...
Dear Gretl Users,
My question concerns KPSS unit root test. After selecting my serie and clicking on KPSS test; the default lag is set to 8.How do I find the best lag for my test ? Is there an automatic way to select it ? Is the default lag the best one ?
Thanks a lot for helping !
Samy
12 years, 8 months
