panel data issues
by Sven Schreiber
Hello all panel-interested people,
while using gretl for teaching with panel data (which I hadn't done much
before) I noticed the following, let's say, interface nuisances compared
to the usual luxury gretl offers for time series:
1: The sample and/or range in the main window (bottom) are given as pure
index numbers, even if "panel data marker strings" (cf. user guide p.23)
are defined. At least for the time dimension it would be useful to show
the sample periods in a human-readable form (through the markers). Also,
I noticed that the period numbers shown do not always coincide with the
values of the "time" index variable, if subsampling is in effect. (Seen
in the CEL.gdt dataset after applying the sample restriction year>1970
for example.)
1b: A slightly more general suggestion, also for non-panel data: The
active sample restriction criterion could be shown next to the resulting
active sample in the main window. (At least for simple restrictions,
maybe not for complex, multiple ones.)
2: Menu Sample -> Set range: Only the group range can be chosen, not the
periods. Actually, given the often arbitrary ordering of groups, this is
really the less useful dimension to choose a contiguous range from. (I
know I can use "set sample based on criterion" for periods, but that's
not the point.)
3: About pshrink(): A version that returns a full panel series (with
repeated values like pmean() etc.) could be useful -- practical example:
in growth regressions one needs the initial value of output-per-worker
as a regressor. Also maybe it should be called "pfirst()" or something
instead.
4: Time-constant variables: I'm not sure how to create variables that
only vary along the cross-section, like it is done with the built-in
pmean() etc. functions. Or how to append them (like the user guide p.114
"adding a time series", but along the other panel dimension).
5: Constant in a fixed-effects regression: I don't understand what gretl
reports as the global constant term in a fixed-effects model, and it
doesn't seem to be defined in the guide. It's also confusing that gretl
complains if one wants to discard the constant in the specification
dialog (when fixed effects are selected). (But obviously gretl
estimates the right thing as the comparison with explicit LSDV
regression shows, just the constant is mysterious -- even if it's the
average of the fixed effects it's not clear where the standard errors
come from.)
6: Lags not showing in model spec dialog when sample is restricted to a
single period: If I restrict the CEL.gdt data with year==1985, I cannot
include any previously created lags (of y for example) in the
regression, because they don't show up in the variable selector. Because
the subsampled dataset is now treated as "undated", there's also no
"lags..." button in the dialog. -- Actually I don't understand why gretl
"temporarily forgets" the panel structure of the dataset when a single
period is active. It would seem less problematic to treat even a T=1
sample as a special case of panel data if the underlying dataset has a
panel structure; especially in conjunction with point 1 above about
showing the selected periods in the sample.
Ok, that was a long post, sorry, but still necessary I think.
Cheers,
Sven
10 years, 6 months
Bug in Gretl Reference :)
by Hélio Guilherme
Hi Allin,
I was learning about Matrices and I noticed that "Table 15.1: Valid
functions on an empty matrix, A" is placed after "15.3 Selecting
sub-matrices" , being confused with "... where rows can take any of these
forms:".
At page 110
---
Took a look to matrices.tex and the text flow was correct. Although is not
the correct thing to do, I have inserted at line 179:
\break %% Forcing the table before next section
This way, the section 15.3 starts at page 110 after table 15.1.
Cheers,
Hélio
10 years, 9 months
problem with CSV import procedure in gretl (v. 1.9.14)
by Evgenii Gilenko
Dear developers of gretl!
Thanks again for the wonderful software! It helps a lot both in scientific
research and in educational process!
Unfortunately, I've discovered a problem with importing data from a CSV file
to a gretl data file.
Specifically, when I try to import daily data from a CSV file (I attach the
CSV-file to this letter), I receive the following:
+++++++++++++++++++++++++++++++++++++++++
parsing D:\aapl.csv...
using delimiter ','
longest line: 56 characters
first field: 'date'
seems to be observation label
number of columns = 7
number of variables: 6
number of non-blank lines: 250
scanning for variable names...
line: date,Open,High,Low,Close,Volume,AClose
scanning for row labels and data...
first row label "014-03-25", last label "2013-04-01"
treating these as undated data
+++++++++++++++++++++++++++++++++++++++++
For some reasons gretl cuts off the first digit of the date in the "first
row label" line (I stressed in red).
And thus it interprets the data as cross-sectional, not as time-series.
I'm using gretl of version 1.9.14. I tried on version 1.9.13 - the same
problem.
But, on version 1.9.12 there was no such problem with importing data from
the attached CSV file - I double-checked it.
Could you please give any comments on how to override the problem on version
1.9.14?
Downgrade to version 1.9.12 is not an option for me now.
Thanks,
Evgenii Gilenko
10 years, 9 months
Problems with MPI supp. on Win 7 machine
by Artur T.
First of all, thank you very much for the implementation of MPI support! I
am really forward-looking to use it in future for heavy computing tasks.
However, atm I am experiencing some an issue with the installation on a Win
7 machine. I have installed MS-MPI and set the directory to:
"C:\Program Files\Microsoft MPI\Bin\mpiexec.exe"
but "eval $havempi$" returns zero.
I use current cvs (32-bit) with no admin rights. Could this be a problem?
Best,
Artur
10 years, 9 months
Re: [Gretl-devel] Gretl-devel Digest, Vol 86, Issue 10
by David Voňka
Thank you. Removing MPICH2 (packages libmpich2-3 and libmpich2-dev) helped.
David Vonka
Message: 5
> Date: Fri, 14 Mar 2014 22:54:11 +0000 (GMT)
> From: Allin Cottrell <cottrell(a)wfu.edu>
> Subject: Re: [Gretl-devel] Compilation on Ubuntu 13.10, MPI
> To: Gretl development <gretl-devel(a)lists.wfu.edu>
> Message-ID: <alpine.LNX.2.03.1403142248050.1246(a)wfu.edu>
> Content-Type: text/plain; charset="utf-8"
>
> On Fri, 14 Mar 2014, David Vo?ka wrote:
>
> > This is the result of make after simple ./configure
> >
> > libtool: link: mpicc -mavx -Winline -o .libs/gretlcli-mpi gretlcli-mpi.o
> > ../lib/.libs/libgretl-1.0.so -llapack -lblas -lgfortran -lm -ldl -lz
> -lxml2
> > -lglib-2.0 -lgmp -lfftw3 /usr/lib/x86_64-linux-gnu/libcurl-gnutls.so
> > -pthread
> > gretlcli-mpi.o: In function `main':
> > /home/vonkad/dev/gretl/cli/gretlcli-mpi.c:318: undefined reference to
> > `ompi_mpi_comm_world'
> > /home/vonkad/dev/gretl/cli/gretlcli-mpi.c:319: undefined reference to
> > `ompi_mpi_comm_world'
> > /home/vonkad/dev/gretl/cli/gretlcli-mpi.c:320: undefined reference to
> > `ompi_mpi_errors_return'
> > /home/vonkad/dev/gretl/cli/gretlcli-mpi.c:320: undefined reference to
> > `ompi_mpi_comm_world'
> > collect2: error: ld returned 1 exit status
>
> It looks as if your MPI installation is broken. The MPI compiler, mpicc,
> is not linking to the right MPI library, and/or the gretl build has
> included the wrong mpi.h header.
>
> Have you by any chance installed both OpenMPI and MPICH2 at some point?
>
> Allin Cottrell
>
>
>
>
> ------------------------------
>
> _______________________________________________
> Gretl-devel mailing list
> Gretl-devel(a)lists.wfu.edu
> http://lists.wfu.edu/mailman/listinfo/gretl-devel
>
> End of Gretl-devel Digest, Vol 86, Issue 10
> *******************************************
>
--
David Voňka
Programátor
mobil: +420 608 916 031
10 years, 9 months
Patch against readline proposal
by Martino
Hello everyone.
I had problems compiling gretl on Arch Linux getting these errors:
> gcc -c -march=native -O2 -pipe -fstack-protector --param=ssp-buffer-size=4 -I.. -I.. -I../lib/src -DLOCALEDIR=\"/usr/share/locale\" -DHAVE_CONFIG_H complete.c
> complete.c: In function 'initialize_readline':
> complete.c:129:41: error: 'CPPFunction' undeclared (first use in this function)
> rl_attempted_completion_function = (CPPFunction *) gretl_completion;
> ^
> complete.c:129:41: note: each undeclared identifier is reported only once for each function it appears in
> complete.c:129:54: error: expected expression before ')' token
> rl_attempted_completion_function = (CPPFunction *) gretl_completion;
> ^
> Makefile:65: set di istruzioni per l'obiettivo "complete.o" non riuscito
> make[1]: *** [complete.o] Errore 1
> make[1]: uscita dalla directory "/tmp/yaourt-tmp-gunther/aur-gretl/src/gretl-1.9.9/cli"
> Makefile:54: set di istruzioni per l'obiettivo "cli" non riuscito
> make: *** [cli] Errore 2
As pointed out here:
http://lists.gnu.org/archive/html/bug-readline/2014-03/msg00003.html
(see the previous message also) the CPPFunction type has been removed as
it was deprecated.
So I just deleted the cast to CPPFunction* in cli/complete.c and now
gretl works fine for me (patch attached). I updated the old package in
the AUR repository for ArchLinux (x86_64 architecture only as of today).
Hope this is someway useful.
Keep up the good work! Cheers
10 years, 10 months
Compilation on Ubuntu 13.10, MPI
by David Voňka
Hello,
I just checkoed out the source and try to compile it on my ubuntu 13.10
64bit box. The problem seems to be the MPI. configure works ok,
./configure --without-mpi
but when I call make, the compilation fails with
make[1]: Leaving directory `/home/vonkad/dev/gretl/lib'
make -C cli
make[1]: Entering directory `/home/vonkad/dev/gretl/cli'
mkdir .deps
gcc -c -g -O2 -I.. -I.. -I../lib/src -I/usr/include/glib-2.0
-I/usr/lib/x86_64-linux-gnu/glib-2.0/include
-DLOCALEDIR=\"/usr/local/share/locale\" -DHAVE_CONFIG_H gretlcli.c
gcc -g -O2 -I.. -I.. -I../lib/src -I/usr/include/glib-2.0
-I/usr/lib/x86_64-linux-gnu/glib-2.0/include
-DLOCALEDIR=\"/usr/local/share/locale\" -DHAVE_CONFIG_H -MM gretlcli.c >
.deps/gretlcli.d
gcc -c -g -O2 -I.. -I.. -I../lib/src -I/usr/include/glib-2.0
-I/usr/lib/x86_64-linux-gnu/glib-2.0/include
-DLOCALEDIR=\"/usr/local/share/locale\" -DHAVE_CONFIG_H complete.c
gcc -g -O2 -I.. -I.. -I../lib/src -I/usr/include/glib-2.0
-I/usr/lib/x86_64-linux-gnu/glib-2.0/include
-DLOCALEDIR=\"/usr/local/share/locale\" -DHAVE_CONFIG_H -MM complete.c >
.deps/complete.d
../libtool --mode=link gcc -mavx -Winline -o gretlcli gretlcli.o
complete.o \
../lib/libgretl-1.0.la -lreadline -lncurses
libtool: link: gcc -mavx -Winline -o .libs/gretlcli gretlcli.o complete.o
../lib/.libs/libgretl-1.0.so -llapack -lblas -lgfortran -lm -ldl -lz -lxml2
-lglib-2.0 -lgmp -lfftw3 /usr/lib/x86_64-linux-gnu/libcurl-gnutls.so
-lreadline -lncurses -pthread
../lib/.libs/libgretl-1.0.so: undefined reference to `set_mpi_variant'
collect2: error: ld returned 1 exit status
make[1]: *** [gretlcli] Error 1
make[1]: Leaving directory `/home/vonkad/dev/gretl/cli'
make: *** [cli] Error 2
a similar problem appears when I call ./configure without hte parameter,
but the undefined referece is to another object, not set_mpi_variant. I
have installed the mpi-default-dev package.
Thanks for any help,
David Vonka
--
David Voňka
Programátor
mobil: +420 608 916 031
10 years, 10 months
gretl conference 2015
by Allin Cottrell
Hello all,
It's maybe not too soon to be thinking about the next biennial gretl
conference. Before getting to that, however, let me thank Lee and
Kathy Adkins again for their wonderful hospitality in Oklahoma City
in June 2013. If you weren't there, you missed out! (Particularly
James MacKinnon's talk on bootstrapping and the fine dinner at the
Petroleum Club, among other highlights.)
I'm not sure if this was announced at last summer's conference
(though I think so), but Sven Schreiber (who unfortunately couldn't
make it) has volunteered to host the 2015 conference in Berlin.
We're happy to take Sven up on this offer. So we're thinking of some
date in June 2015. Sven tells us that Helmut Lütkepohl has agreed to
give an invited talk -- good news!
I'll let Sven and others take it from here: expect some sort of
polling for the best date and preparations for forming a scientific
committee.
Allin
10 years, 10 months
gretl + MPI
by Allin Cottrell
We have an exciting new development to announce: support for
parallelization via MPI in gretl. This is in CVS, the Windows snapshot,
and the quartz snapshot for OS X. There's preliminary documentation
with examples at
http://users.wfu.edu/cottrell/tmp/gretl-mpi.pdf
Here are the first two paragraphs just to give you an idea:
<exerpt>
As of March 2014 gretl CVS includes support for MPI. In brief, this takes
the form of (a) a new, dedicated command-line program called gretlcli-mpi,
which may be used in stand-alone fashion, plus (b) a facility for calling
this program from within a regular hansl script (e.g. as opened in the GUI
script editor). These things are described in detail below.
What is MPI?
To get the full story, see open-mpi.org or mpich.org; here we just
describe the basics as they apply to gretl. MPI (Message Passing
Interface) is a de facto standard that supports running a given (C or
Fortran) program simultaneously on several cores in a given computer
and/or on several networked computers. MPI provides means for dividing
labor among the multiple nodes and for communicating data between them. It
thereby supports a very flexible sort of parallelism...
</exerpt>
The hansl interface to MPI is not yet "set in stone"; we welcome comments.
Allin and Jack
10 years, 10 months
Compilation is broken
by Hélio Guilherme
Hi,
In gui2/library.c at line 7215 we have: shown = 1;
(variable is not used)
Adding int shown = 0; in line 7181 solves compilation error.
Hélio
10 years, 10 months
