questions about the next release
by Allin Cottrell
It's time we put out a new gretl release (this will be 1.9.13) and
we're pretty much ready to do that. But first I'd like to ask a
couple of questions, about the gretl packages for MS Windows and Mac
OS X, and see what people have to say.
MS Windows:
For some time now I've been putting up snapshots with a 64-bit build
of gretl for Windows; see
http://sourceforge.net/projects/gretl/files/windows-testing/
I've tested this myself on Windows 8 and found no problems. (There
was a problem with gnuplot at first but that is long since fixed.)
Since 64-bit Windows is becoming more widespread I think it would be
worth moving this out of testing and offering it as a regular
download option. Anyone see any reason why not? (Since the 32-bit
version runs OK on win64 but not vice versa we could say something
like, "If you're not sure which version you need, choose the 32-bit
one.")
Mac OS X:
For some time I've kept the snapshot gretl-quartz.dmg.gz up to date;
see
http://sourceforge.net/projects/gretl/files/osx-testing/
Unlike our "regular" OS X packages this one does not require that
X11 be installed, and neither does it require a separate download of
the GTK "framework" -- it is self-contained. In addition it is a
64-bit build and it is somewhat more "Mac-like" in its behavior than
our other packages.
Anyway, I'm thinking we should probably make this "quartz" version
of gretl the default download for Intel macs from the next release
onward (while keeping the X11 version available as an option).
My main reason for saying this is that various changes in OS X since
10.6 have made it more difficult for the "average" user to install
any third-party software supplied in the form of a disk image (dmg),
as gretl is -- let alone to install three pieces of software (X11,
GTK, gretl) as required by the X11 build. We really should supply
totally explicit instructions (preferably with screenshots), and
this is much more feasible for an all-in-one package.
There are a couple of reasons why some users might prefer to stick
with the X11 version, but I don't think these will apply to most Mac
users:
(1) keyboard navigation of the GUI is not as well supported in the
quartz version at present (this may fix itself as the
gtk-mac-integration library improves over time); and
(2) to avoid dependence on X11, the interactive gnuplot "terminal"
shipped in the quartz package is "aquaterm", which looks quite nice
but does not support interactive manipulation of "3-dimensional"
plots. Note that this has nothing to do with most of the plots
displayed by gretl, which don't use any gnuplot "terminal"; it's
only 3-D plots we're talking about.
So my question here is: does anyone have reason to think we
shouldn't switch to 64-bit gretl-quartz as our default offering for
the Mac?
Allin Cottrell
12 years
system of equations
by valentina colombo
Thanks. However, the command "series res= $uhat" works only whether I estimate my system only equation by equation. To be clear, it works whether I estimate 4 different equations by ols and then collect step by step the residual of each regression by the above command and the RSS. It may work, but then I need the matrix residual sum of squares of the system of equations to pass at 2nd step to compute. Because at end I have to compute the test statistic
LM=T(p-tr{RRS0^(-1) RRS1}
Has somebody any suggestion to obtain in a simple way the matrix residual sum of squares?Thanks
12 years
MVGARCH with DCC
by Fazeel Jaleel
Dear Gretl Users,
Could anyone tell how to use MVGARCH with DCC or CCC in GRETL.
Thank you,
Fazeel
12 years
non stationary panel
by Alessia Via
Hi all.
I'm working with a non-stationary panel model. I would like to know if
Gretl has any routine for:
Pedroni cointegration test
PANIC (common factor) methodology
Thanks.
a.
12 years
How to post some question
by valentina colombo
Dear Gretl-users,I have to test by LM this vector of variables Y=[g t y n]'. Thus, I create a system with 4 equations, each one includes some lagged variable of one of the variables in Y. 1st step: I run an ols and save the residual by "genr res=$yhat", also I have to save the matrix residual sum of squares. To do the latter I use to transform into X'X my residual.
2nd step: I need to regress the residuals on some other variables (i.e. g*c)At this step I have a problem: I cannot regress the residual on covariates because "the residual is not a variable".
Can you suggest me how to solve this problem in Gretl?
Thanks
> Subject: Welcome to the "Gretl-users" mailing list (Digest mode)
> From: gretl-users-request(a)lists.wfu.edu
> To: valentina81c(a)hotmail.it
> Date: Thu, 10 Oct 2013 10:33:54 -0400
>
> Welcome to the Gretl-users(a)lists.wfu.edu mailing list!
>
> To post to this list, send your email to:
>
> gretl-users(a)lists.wfu.edu
>
> General information about the mailing list is at:
>
> http://lists.wfu.edu/mailman/listinfo/gretl-users
>
> If you ever want to unsubscribe or change your options (eg, switch to
> or from digest mode, change your password, etc.), visit your
> subscription page at:
>
> http://lists.wfu.edu/mailman/options/gretl-users/valentina81c%40hotmail.it
>
>
> You can also make such adjustments via email by sending a message to:
>
> Gretl-users-request(a)lists.wfu.edu
>
> with the word `help' in the subject or body (don't include the
> quotes), and you will get back a message with instructions.
>
> You must know your password to change your options (including changing
> the password, itself) or to unsubscribe. It is:
>
> stefania84
>
> Mailman will remind you of your lists.wfu.edu mailing list
> subscriptions by e-mail once a semester, although you can disable this
> action if you prefer. This reminder will include instructions on how
> to unsubscribe or change your account options.
>
> There is also a button on your options page that will email your
> current password to you.
12 years
Re: [Gretl-users] Question concerning paths to files
by Pindar
Cool!
Thanks for the manifold hints and examples :-)
Cheers
Leon
>
>
>>> Am 09.10.2013 um 16:49 schrieb "Riccardo (Jack) Lucchetti" <r.lucchetti(a)univpm.it>:
>>>
>>> On Wed, 9 Oct 2013, Pindar Os wrote:
>>>
>>> Hi there,
>>>
>>> I found some infos how to use the getenv() function in the manuel for e.g.
>>> 'USER'. Did you mean that Jack?
>>> I think in terms of an automatic parse fucntion the getline() fucntion
>>> would be needed and sscanf().
>>
>> I mean: suppoose you have the following hansl script called foo.inp
>>
>> <hansl>
>> string foo = getenv("A")
>> scalar bar = ngetenv("N")
>> printf "%g kilograms of %s\n", bar, foo
>> </hansl>
>>
>> in a Unix shell you can execute the following
>>
>> <bash>
>> A=wood N=23.5 gretlcli -b foo.inp
>> </bash>
>>
>> or
>>
>> <bash>
>> A = stone; export A
>> N=33.3 gretlcli -b foo.inp
>> </bash>
>>
>> and get output like
>>
>> <output>
>> 23.5 kilograms of wood
>> </output>
>>
>> Of course you could enclose this into a shell script; I guess something similar can be done with a Windows "batch file", or whatever they're called.
>>
>> This said, in the current CVS version, which will soon become 1.9.13, we have quite a few new functions that make it possible to exactly what you say: read a text file and grab numerical or string values from it, given a proper syntax.
>>
>> -------------------------------------------------------
>> Riccardo (Jack) Lucchetti
>> Dipartimento di Scienze Economiche e Sociali (DiSES)
>>
>> Università Politecnica delle Marche
>> (formerly known as Università di Ancona)
>>
>> r.lucchetti(a)univpm.it
>> http://www2.econ.univpm.it/servizi/hpp/lucchetti
>> -------------------------------------------------------
12 years
Question concerning paths to files
by Pindar
Hi there,
I have some problems with gretl (snapshop on windows 8) automatically
adding its 'shelldir' to a file path:
<hansl>
eval shelldir
open abdata.gdt
list xList = YEAR IND CAP
store "testme.csv" --csv xList # writes file in
'shelldir' -> ok
string readme = readfile("testme2.csv") # reads file in 'shelldir' -> ok
print readme
open "testme.csv" # *fails *to
read file in 'shelldir' ->*why?*
string file = shelldir
file +="\testme.csv"
open @file # works
<hansl>
Cheers
Leon
12 years
Console command - choice of ordering for Cholesky decomposition for VAR IRF
by Felipe L. Bhering
Dear list,
1. My question is about choice of ordering for Cholesky decomposition for
VAR Impulse response function in Var/VEC.
in Gretl 1.9 version this started working for Gui - enable choice of
ordering for Cholesky decomposition for VAR impulse response plots
I wanted to know how to change order of cholesky decomposition via console
command.
simple example, making a shock in a var
#####################
matrix shock = irf(1,1)
gnuplot 1 {set title "Response of variable 1 to a 1 s.d. shock in
variable 1";} --time-series --with-lines --matrix=shock
--output=display
#####################
where this command changes?
2. Another thing is that my gretl just stopped recognizing the \ char at
the end of the line. It has happened before to you?
Thanks,
Felipe
12 years
Heteroskedasticity test
by Rezart Hoxhaj
Dear all,
I am trying to find any method to make a heteroskedasticity test
for an interval regression estimation. Do anybody have any clue on
how to do it?
Best,
Rezart
12 years
issue defining a panel with sub-sample data
by Artur T.
Hi all,
I've got an issue with the following. It used to work in the past, but
doesn't do so anymore. I want to estimate a panel model for a sub-sample
dataset. I am using current cvs on linux.
<hansl>
open abdata.gdt
# Restriction for which panel structure remains
smpl YEAR > 1977 --restrict
smpl full
# Restriction for which panel structure doesn't remain
# -> Undated dataset
smpl YEAR > 1977 && EMP > 40 --restrict
# Redefinition to panel structure is not allowed acc. to gretl
setobs unit YEAR --panel-vars
<\hans>
Any hints what's wrong here?
Artur
12 years
