cvs builds on ubuntu edgy
by Sven Schreiber
Hi,
I upgraded from ubuntu dapper to edgy and rebuilt gretl. However,
checkinstall-ing didn't work anymore for some (unknown) reason, so I
just did the standard make install (of today's cvs checkout). So for the
moment everything is fine, but I wonder what I should do once I decide
to install a newer version without getting any conflicts. Any hints?
Thanks,
Sven
17 years, 11 months
'store' issues
by Sven Schreiber
Hi,
playing around with the 'store' command, I came across the following (on
1.6.0 self-compiled on ubuntu edgy):
* store saves data in ~, even though ~/gretl is activated in prefs
* with data from example file data2-3.gdt (annual), store saves dates as
literally "'1995" (including the double quotes); reopening works fine,
but why all these quotes, including the single quote?
* store --csv saves first column with heading "obs", although in the
guide it says that for reading the column must have header "date" ??
(well actually haven't double-checked that last assumption, sorry, am in
a rush) But reopening with ascii works, so just a documentation bug?
* gui importing from csv seems to presuppose .csv extension, must use
ascii import for csv files with different extension -- is that necessary?
* 'open' seems to rely on heuristics only; why no possibility to specify
e.g. --csv as in store?
Sorry if anything has been obsoleted in newer versions,
Sven
18 years
Answer to Peter
by Javier García
I thank you so much your clarification, John. Merry christmas and a happy
new year ;P
Javi
18 years
Answer to Peter
by javier garcia enriquez
Peter Summers wrote:
Javi,
What you really need to do is consult an econometrics textbook, not the gretl list.
Sorry, but I think youre wrong. If you have read the Johansens article and other autors articles, you can see the topic that I question isn´t clear in the literature, especially with the lambda max test. Depending what software do you use, the null hypothesis is different and, in statistics, the interpretation depends completely of the null hypothesis. So, to interpretate Gretl´s or other programs output we really need the null hypothesis that they use, dont you agree?
Thanks to all, especially to Allin and Merry Christmas
Javi
PD: sorry for my bad English :p
---------------------------------
LLama Gratis a cualquier PC del Mundo.
Llamadas a fijos y móviles desde 1 céntimo por minuto.
http://es.voice.yahoo.com
18 years
Johansen(now with the table well)
by Javier García
Hi;
Before I put a Gretl's example. Data set which I study is monthly. I have
101 observations and, if I apply the Johansen's tests in it, I obtain
following output:
Rank Eigenvalue trace p-value l-max p-value
0 0.11637 13.736 0.3151 11.01 0.2600
1 0.030163 2.7258 0.6392 2.7258 0.6380
Then I question to my self: if two hypothesis are accepted, why we must
conclude that there aren't any cointegration relationships??? To conclude
that I think we have to interpretate sequently, that is, starting with the
less rank, the first rank we accept it will be the rank of cointegration,
won´t it? In this case rank zero, doesn´t it?
If the null hypothesis in L-max were that there are exactly "r0" ( and no
r<=r0 as the trace test)) cointegration vectors, it would impossible the
interpretation, because the rank would be, at the same time, zero and one.
In a lot of articles there is that distintion between both tests.
I continue thinking that this isn't very clear in Gretl
Cheerssssss and thanksss
Javi
18 years
johansen
by Javier García
And then the interpretation is sequentialy, isn't it?? That is, first we
look at the rank zero and, if it is refused, we look at the rank one...until
one of them doesn't be refused??? or, on the contrary, we start with the
biggest rank and we goes down??
Thanks
Javi
18 years
Johansen
by Javier García
Sven, in other programs, like as Eviews, the null hypothesis are specified.
I want know which are the null hypothesis that Gretl uses, not the
interpretation. Depending of that the interpretation is diferent, isn't it?
Feliz Navidad/Merry Christmas and thanks for your time
Javi
18 years
Suggestions for improvements of gretl
by Andreas Karlsson
Hi,
I have been using gretl in my teaching, and it has worked well. The
students seem to like the software.
However, I have a few suggestions for improvements of gretl, making
it possible for me to use it in other courses:
1. A possibility to generate random binomial variables, for example
by adding a menu entry Add > Random variable > Binomial
2. For frequency plots of continual data (i.e., histograms), amke it
possible for the user to
a). Choose bin width
b). Choose number of bins
3. For discrete data, make it possible to create pie charts
These are the most important improvements. What are the chances of
these suggestions being implemented soon?
Some other suggestions, not necessary for using in the courses I am
thinking about, but which would be quite useful and nice to have in
gretl (in order of importance):
1. Support for Multinomial logistic regression
2. Forward/backward stepwise selection for regressions
3. Direct import of R format, SPSS format and OpenOffice.org Calc format files
4. Quantile regression
5. Ordinal logistic regression
Are there any chances of these suggestions being implemented?
Best regards,
Andreas
18 years
Re: [Gretl-users] Suggestions for improvements of gretl
by Riccardo (Jack) Lucchetti
On Thu, December 21, 2006 22:33, Andreas Karlsson wrote:
>>I'm personally inclined to say that from now on everything that can
>> reasonably
>>done via a user function should be done that way. I say "reasonably" because
our scripting infrastructure is, as of now, not sophisticated enough to
support computationally demanding methods.
>
> Hi,
>
> When I use gretl in my teaching for students in statistics or
> econometrics at a basic level, it is as a replacement for Minitab or EViews.
To let them do the tasks I am giving them, it would be to hard for them to
write scripts and use other programming
> skills. They want to have a point-and-click GUI like the one in EViews or
Minitab, which I think the gretl GUI can be compared with in usability. For
this purpose, script based user function is thus not enough, but it has to
be a GUI for these tasks. That's why I have asked for inclusion of some
features.
Foreword: Allin the merciful and compassionate has already provided a "point
and click" interface for generating binomials, which is is CVS right now,
together with a few histogram fixes.
This said, I do realise that one of gretl's selling points is ease of use,
maybe the biggest one. But I also think that if you try to cram too many
things in a GUI interface, all you get is information overload and easy things
become difficult simply because you have too many items to choose from. I
personally don't use MS Word, but people who do tell me that this is exactly
what happens to them (which is why, by the way, some MS programs now have this
"feature" of just showing you only some menu items --- bizarre if you ask me
but, hey, I'm a linux user, so I'm not really bothered).
If you want to manage a decent level of complexity, there is no alternative to
a text interface. I don't mean to interefere with your style of teaching, but
IMHO giving the students the idea that a text interface can be a powerful tool
(which, incidentally, prevents carpal tunnel syndrome) is not that bad.
And if I haven't convinced you so far, it's ok: as Sven said, write your own
function and give it a GUI interface: the manual will tell you all about it.
> Best regards, and thanks for a great software
Thank you for your ideas!
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Facoltà di Economia "G. Fuà"
Ancona
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Facoltà di Economia "G. Fuà"
Ancona
18 years
gretl translation update
by Allin Cottrell
Hello all translators,
I've now committed to CVS a revised .pot file, in preparation for
the next gretl release.
I've also added the new Portuguese translations, and built these
into a new Windows snapshot. Thanks Hélio!
--
Allin Cottrell
Department of Economics
Wake Forest University, NC
18 years
