I have 20 quarter observations from 1981:1 to 1990:4
When I run an OLS-estimation with three dummy variables yt = ß0 + ß1*t + ß2
*D2 + ß3 *D3 + ß4 *D4 + ut
I get a result with the titel « Model x : OLS, using observations 1986
:2-1990 :2 (T17). »
It is not clear to me why Gretl takes 17 observations instead of 20 to run
the OLS estimation.
Thank you in advance
Em 22 de setembro de 2015, Obasa disse:
Yes on a window Pc window 7 ultimate 32 bit
> I am currently using it for a research as an indepedent local consulttantSent
> from my BlackBerry® wireless handheld from Glo Mobile.
Please keep the list in this thread. Someone else could be facing a problem
like yours and this messages could help these people too ;-) So let me try
(1) To install packages: At the main window (the first one you see when
open Gretl) go to: Tools -> Function Packages -> On server. Select the
function you want to install and click on the diskette icon.
(2) By default they will be saved to the folloing folder: "C:\Program
Files\gretl\functions" (if you use Windows 64 bit) or "C:\Program Files
(x86)\gretl\functions" (if you use Windows 32 bit).
I hope this could help you.
I have found a discussion from last December about performing the
Bai-Perron test for structural changes in Gretl.
Eventually it was suggested to exploit Gretl’s link to R and do this test
using the strucchange package for R. Could someone help me in doing this? I
am far to be defined an expert in coding and programming… Thank you so much
(I copy and paste below the discussion that I have found in the archive – I
was not part of the mailing list at the time).
On Sun, 28 Dec 2014, Sven Schreiber wrote:
>* Am 28.12.2014 um 23:11 schrieb Riccardo (Jack) Lucchetti:*
>>* On Sun, 28 Dec 2014, Allin Cottrell wrote:*
>>>* In the meantime, one could use gretl's link to R and do this test*
>>>* using the strucchange package for R.*
>>* This has come up before. I've got a complete implementation in Ox,
>>* is IMO much better than the GAUSS cose by the original authors. I never*
>>* found the time to translate it to hansl; it's a long job because my
>>* is very ox-ish in style (heavily object-oriented). If someone is*
>>* interested, I can send the whole thing and possibly help with little*
>>* things, but it's quite low on my priority list, so if you wait until I*
>>* do it myself you may be disappointed (QUITE disappointed).*
>* I cannot realistically promise to complete a port to hansl, but I have*
>* recently started to explore the gretl-to-Ox link for another real-world*
>* scenario. So I could give it a try to make a hansl wrapper around your*
>* (Jack's) Ox code. Caveat: I have noticed that the possible interactions*
>* between gretl and Ox within the 'foreign language=Ox' framework seem to*
>* be more limited than with other languages, so I might come up with some*
>* corresponding feature requests along the way.*
>* Of course that requires Ox to be installed, but since it is costless for*
>* academics I guess it might acceptable for the medium term. BTW, what's*
>* the relationship between R's 'strucchange' and Jack's Ox code in terms*
>* of features?*
I have no idea. :D
I wrote the code a while back, used it for what I needed it, and then
become abandonware. I'll send you a zipfile privately.
Pls how can I add the gretl extra packages to the gretl application istalled on my computer.
I attempt to installed but its not loading. Which folder in the gretl directtory should I install the extras into?
Sent from my BlackBerry® wireless handheld from Glo Mobile.
please ask questions on this gretl mailing list. As you can see, I am
sending my answer to the list, one of the reasons is that I do not know
(Apparently Raul has some files on Microsoft's/Windows' "OneDrive". If I
understand correctly, gretl's file-open dialog doesn't offer an access
to that even though it's visible from Windows' Explorer.)
My guess would be that within Windows the OneDrive location has to be
mapped to a windows-typical drive letter. At least that's how I know it
from other network drive locations.
In principle this would not be a gretl-related question. However,
perhaps with all this cloud-computing out there perhaps there is some
programming interface in the OSes that programs like gretl could use, I
Am 21.09.2015 um 15:14 schrieb Raul Gimeno:
> Guten Tag Herr Schreiber
> Ich bin momentan ziemlich verzweifelt mit Gretl, da ich keinen Zugang
> auf meine Dateien auf OneDrive habe. Meine Dateien befinden sich auf
> OneDrive (rechts) aber im Gretl wird nur das Verzeichnis C: gezeigt.
> Wie kann ich das Verzeichnis OneDrive im Gretl haben?
> Vielen Dank für Ihre Antwort.
> Vielen Dank im Voraus
> Dr. Raul Gimeno
out of curiosity, why does the "eval" command exists?
If I define a matrix "A" and I type "A" in the console, the
contents of A is displayed.
If I calculate A*B, nothing is displayed, I need to use eval.
What's the reason behind it? Matlab does give the answer
directly, but I am sure you guys have a reason why it works
this way in Gretl.
Dear fellow gretl users,
I have two questions about deterministic trends in VEC models.
gretl offers the five traditional approaches to VEC modelling, either
omitting all deterministic terms or including constants and trends in
restricted and unrestricted combinations. I was wondering if there are
any good "rules of thumb" for when to choose which approach? The EViews
manual, for example, suggests choosing a model with constant +
restricted trend if some of the level variables appear to oscillate
around a linear trend, I believe (although my memory may serve me wrong
Also, from the gretl manual I learned that the specification of a VEC
model with a constant and *restricted* trend looks like this (LaTeX
I was wondering how the restriction of alpha being orthogonal to mu is
achieved in practice?
Thanks in advance for your help!
section 15.4 of the guide enumerates matrix operators, and there is an
elementwise equality comparison (.=), but no corresponding inequality
comparison (no .!=). And indeed gretl doesn't accept .!=, but why? I
know this can be worked around, but the same argument would hold for .>
and .<, for example.