interpretation var lag selection
by JOSE FRANCISCO PERLES RIBES
Dear all.
I'm estimating a Unrestricted VAR in gretl via the GUI, and previosly I
check the best order of the VAR with the option selection menu. I'm a
little surprised to see how AIC BIC and HQC criteria obtained are different
in this option, which then get really when estimating the VAR.
For example, I find that selecting a maximum of 4 lags, the optimal lag
function gives me 1 lag in AIC, BIC and HQC. But when after I estimated the
model with 1 or 2 lags, AIC, BIC and HQC are lower for 2 than for 1 lags, plus
get a better R2 with 2 lags. I have found that the criteria obtained by
estimating the VAR are identical to those of Eviews.
Am I doing something wrong? or is there a problem with the menu
selection optimum
lag for vars? Because the results and interpretations (eg in the
Granger Causality
test) are very different as choose one lag or two lags for the var
estimation.
Thanks in advance and sorry for any inconvenience.
José F. Perles
University of Alicante
Spain
12 years, 2 months
- 3
- 3
feauture request: boxplot of vectors
by artur tarassow
Dear gretl programmers,
I just found out that it is not possible to apply the boxplot command to
vectors/matrices. Would it be possible to activate this feature? This would
also increase the consistency, as it is already possible to use the gnuplot
command for matrices. Maybe one can use a similiar syntax as for the
gnuplot command:
boxplot 1 2 3 --matrix=test --output=display
Additionally, in the case that columns are named by 'colnames' before, it
would be helpful to plot these strings as well.
I hope this is not to demanding.
Best,
Artur
12 years, 2 months
- 3
- 6
Fwd: [ESS] Gretl mode for ESS
by ahmadou dicko
I forwarded the discussion about a Gretl mode in ESS for those who are not
on the ESS <http://ess.r-project.org/> mailing list.
I'm really happy to see that we'll have soon a Gretl mode in ESS
Ahmadou
---------- Forwarded message ----------
From: Vitalie Spinu <spinuvit(a)gmail.com>
Date: Mon, Oct 1, 2012 at 10:17 AM
Subject: Re: [ESS] Gretl mode for ESS
To: ahmadou dicko <dicko.ahmadou(a)gmail.com>
Cc: ess-help <ess-help(a)stat.math.ethz.ch>, gretl-devel(a)lists.wfu.edu
>> ahmadou dicko <dicko.ahmadou(a)gmail.com>
>> on Mon, 1 Oct 2012 00:03:27 +0000 wrote:
ad> Thank's a lot Vitalie for your quick reply.
ad> I have never heard of gretl. How big is the community? And how does it
>> compare to R, ox and eviews?
>>
ad> Gretl is a software for econometric analysis ( also made by
ad> econometrician ) but you can do some statistical analysis too
ad> (even if I prefer R for that).
ad> Nowaday, It's widely known among econometrician and it's viable
ad> alternative to Stata and Eviews.
Yes, I have looked into it a bit closer. It's a pretty mature
kit. Detailed manual of 330 pages and very easy to use GUI. Not to
mention a huge set of scripts and datasets for Ramanathan, R. (2002) and
Greene, W. H. (20xx). It is indeed very focused and pretty much
complementary to R in terms of time series.
Help system is good. It will be very easy to integrate with ESS
help. Argument completion and eldoc are also pretty straightforward as
their help pages are well structured and easy to scrap (though it would
be good to have an internal function to retrieve arguments and command
options directly).
Gretl has a pretty nice menu interface. I wonder if there is a way to
start a GUI dialog directly from gretl console. Would be nice to
automatically generate all gretl menus in emacs such that users familiar
with GUI can feel at home.
I am pretty convinced now that gretl would be a nice addition to ESS,
especially in the light of it being GNU licensed.
Vitalie
ad> I think that R and Gretl can play nice together because Gretl is
ad> really fast (really fast) and have some nice time series library
ad> (I just started to write a R package to exchange easily data
ad> between R and Gretl https://github.com/dickoa/gretlReadWrite)
ad> Comparing to R and I think that the community is smaller but It
ad> growing very fast (active mailing list, yearly
ad> conference<http://www.gretlconference.org/>and some book about the
ad> software).
>> It's not that difficult to add a new interaction mode to ESS. The main
>> problem is not with adding a mode, but rather with maintaining it
>> latter on.
>>
>> I can look into it, but not earlier than in a couple of months time.
>>
ad> Great !!!
>> You can look into it yourself. Sources for inspiration are
>> ess-julia.el, which you probably should start with, and also
>> ess-r-d.el for a more complete picture. Once the editing mode is
>> there, the main step is to construct ess-gretl-customize-alist with
>> all the defaults needed for the interaction with the subprocess,
>> and then write 'gretl function for starting a subprocess.
>>
ad> Thank's a lot...ESS 12.09 is very nice and I already tried to play
ad> with Julia (which is really interesting too). So, I'll look into
ad> those files and try to understand how their works
ad> Thank's again
ad> On Sun, Sep 30, 2012 at 10:00 PM, Vitalie Spinu <spinuvit(a)gmail.com>
wrote:
>> >> ahmadou dicko <dicko.ahmadou(a)gmail.com>
>> >> on Sun, 30 Sep 2012 13:04:40 +0000 wrote:
>>
>> > Hi everyone,
>> > I would like to know if adding a GRETL <http://gretl.sourceforge.net/>>
mode
>> > to ESS is planned ?
>>
>> I have never heard of gretl. How big is the community? And how does it
>> compare to R, ox and eviews?
>>
>> > There's already a basic gretl
>> > mode<
>>
http://fossies.org/unix/misc/gretl-1.9.9.tar.gz:a/gretl-1.9.9/utils/emacs...>
>inemacs
>> > but it lacks several features availables in ESS R-mode (sending
>> > command line, buffer, autocompletion, etc)
>>
>> It's not that difficult to add a new interaction mode to ESS. The main
>> problem is not with adding a mode, but rather with maintaining it
latter
>> on.
>>
>> I can look into it, but not earlier than in a couple of months time.
>>
>> > So my question is if it's not planned, where to start ? (I would like
>> to
>> > help even if my elisp programming skillz is weak)
>>
>> You can look into it yourself. Sources for inspiration are
ess-julia.el,
>> which you probably should start with, and also ess-r-d.el for a more
>> complete picture. Once the editing mode is there, the main step is to
>> construct ess-gretl-customize-alist with all the defaults needed for
the
>> interaction with the subprocess, and then write 'gretl function for
>> starting
>> a subprocess.
>>
>>
>> Vitalie
>>
12 years, 2 months
- 1
- 0