Dear gretl users:
You might be interested to hear about my new paper entitled "Should
Economists Use Open Source Software for Doing Research?" published
this month in Computational Economics. The paper investigates
econometric software reliability and advocates the use of open source
software by taking gretl as a case study and showing how responsive
and transparent its development process is. I think many people here
might find it an interesting read.
More information and the download link is available here:
I can send a working paper version if you do not have access to the above.
A. Talha Yalta
“Remember not only to say the right thing in the right place, but far
more difficult still, to leave unsaid the wrong thing at the tempting
moment.” - Benjamin Franklin (1706-1790)
I have 2 feature requests which would be nice to have in gretl.
1. Plotting the error-correction-terms: At the moment one has to save
the ec-term(s) before one can plot them as series. It would be much more
convenient to have the possibility to plot them out of the model menu as
possible for IRFs or residuals. This could be implemented under the
"Graphs" menu in the vecm model menu: "Graphs --> ECM-Plot" or "Graphs
--> Combined ECM-Plot" -- of course if not too hard to implement.
2. Also atm one can only compute the confidence intervals for an
unrestricted VECM but not for a restricted one. In this case one has
unfortunately to switch to different software programs.
@Riccardo: Are there intentions to implement this feature into your SVAR
I am sure this is more difficult to program than the first request.
Best wishes and happy new year!
Hello gretl community
I've got a problem obtaining the ECM-term from a restricted VECM via the
Consider the following example:
1 <- vecm 2 1 LRM LRY IBO
ecm = $ec
2 <- restrict --full
ecm_r = $ec
ecm = ecm ~ ecm_r
gnuplot 1 2 --with-lines --time-series --matrix=ecm --output=display
The obtained ECM-term from the restricted model looks weird. If one gets
the ECM-term from the restricted model manually and not via the script,
it looks totally different and much more plausible.
I use gretl 1.9.3 cvs (build date 2010-12-22) on ubuntu.
Best wishes and happy new year! ;)
1. There is a longer term alternative to the suggestion made by
John. It appears that you are at the Free University in
Berlin. Most academic libraries and/or computing services have
access to IMF, EU and other data for use by students and staff, often
through data services provided by consortia of universities. There
is one in the UK and I would expect that there is something similar
in Germany. If the Free University doesn't have such access, then
you should lobby for it. Academic subscriptions to international
datasets are usually very cheap.
2. There is no particular advantage to using data from the IFS. It
is just a compilation of data from national sources. Since you are
dealing with a few OECD countries, you can extract data from the
OECD's online databank without charge or, indeed, from the relevant
national sources. I am not sure about Japan but it is certainly
possible to get what you want for the other countries.
>Date: Tue, 28 Dec 2010 10:25:55 +0100
>From: Leon Unger <pindar(a)zedat.fu-berlin.de>
>Subject: [Gretl-users] WPI series - Has somebody access to IMF's
> International Financial Statistics?
>To: Gretl list <gretl-users(a)lists.wfu.edu>
>Content-Type: text/plain; charset="iso-8859-15"
>I'm working on a paper concerning the link between international parity
>and granger causality. In order to built PPP the WPI series from 1975M03
>on from USA, UK, Germany, Japan and Canada would be of great help for
>me. After several days of search I had to draw the conclusion that the
>series from IMF's International Financial Statistics are the only chance
>to get comparable information. However, I have not access to them. Does
>someone can help me out?
I try to estimate a logit model using gretl according to the user's
I used an embedded example for the calculation (credscore.gdt).
The dependent variable (Y) = 0 or 1.
Four independent variables: MDR, Age, Selfempl, Income.
The package performs calculations using maximum likelihood method.
But I would like to use an ols method.
How can I do it?
Variable sustitution does not work:
y* = ln (P / 1-P) = a + bx
but y* is unobserved variable (latent).
P^ = 1 / (1 + exp (-y*))
ln (1/P^ -1) = a + bx
May I use Y (0 or 1) instead of P^?
Any help will be appreciated.
With best wishes for Christmas and the New Year,
I'm working on a paper concerning the link between international parity
and granger causality. In order to built PPP the WPI series from 1975M03
on from USA, UK, Germany, Japan and Canada would be of great help for
me. After several days of search I had to draw the conclusion that the
series from IMF's International Financial Statistics are the only chance
to get comparable information. However, I have not access to them. Does
someone can help me out?
On Tue, 21 Dec 2010, Yangbo Du wrote:
> In gretl 1.9.2-cvs, I encountered the following error message when
> attempting to save a box plot in PNG format: graph.9: No such file or
> directory. Is there a bug that needs to be fixed?
Maybe. Thanks, I got the data file that you posted (though it was
too big for the list). Using those data I can plot and save as PNG
the boxplot that you describe, without an error. However, the
"graph.9" business suggests that you got the error when saving the
PNG from within a gretl session containing several graphs. If you
could send me the sesson file (off list) I'll take a look and see
if there's something wrong there.
I will be out of the office starting 12/23/2010 and will not return until
Hi All, I will be out of the office on PTO from Dec 23 to Jan 2, returning
Monday, Jan 3. Please, if you have an immediate concern regarding Cost
Model or Construction Economics, please email or contact Bradley A Njus,
510 625 4595, with your questions.
Best Regards, and Merry Christmas and Happy New Year to all.