SVAR Restriction
by Olasehinde Timmy
Dear Prof,
I am aware that the code 'SVAR_restrict(&x, "C", 1, 2, 0)' is to inform
Gretl of the nature of the restriction to estimate. I also know that this
has to be repeated for several restrictions like in the model. However, I
consider this to be rigorous especially when a large matrix is involved.
Although, I know how a matrix can be created and used with the SVAR GUI, I
don't know how it can be used with the SVAR_restrict(&x, ) function.
Please, I need to be enlighten if this is possible or if I have to do the
imputation for each restriction.
Thanks.
11 months, 2 weeks
Uploading the new version of a package
by Paolo Chirico
Dear,
I have a new version of my *stat_match* package ready (Jack already saw
it this summer). Now I would like to update the package uploaded to the
server, but I am not sure how to proceed.
Should I proceed as if I were uploading a new package?
Thanks,
Paolo
--
Paolo Chirico
Università del Piemonte Orientale
Dip. di Giurisprudenza e Scienze Politiche,
Economiche e Sociali (DIGSPES)
Alessandria, Italia
1 year
Hessian with negative/missing diagonal elements ?
by Alecos Papadopoulos
I was discussing with a colleague an empirical application, and he told
me that the only problem he had was that some elements of the Hessian
was "nan" because they came up with the wrong sign. But the R code that
he used did print out the Hessian, with these elements as "nan".
Can we do that in gretl (natively)? I ask because in standard
mle-implementation with the option --hessian we may get "Hessian is not
negative-definite, dropping back to OPG".
What I ask is whether we can instruct gretl to stick with the Hessian,
returning some "nan" values in the diagonal / std errors.
--
Alecos Papadopoulos PhD
Affiliate Researcher
Dpt of Economics, Athens University of Economics and Business
Foundation for Economic and Industrial Research (IOBE)
web: alecospapadopoulos.wordpress.com/
ORCID:0000-0003-2441-4550
1 year
changing the language for gretl on macOS
by Cottrell, Allin
A heads-up for Mac users: in the 2024b release of gretl there's a
problem (at least for recent versions of macOS) with the mechanism for
selecting the language in which gretl appears. (I'm talking about the
item labeled "Language preference" under /Tools/Preferences/General.)
On trying to select a language you're likely to get an error message.
An example of this can be seen at
https://sourceforge.net/p/gretl/bugs/310/
I'm glad to say this problem is fixed in the current gretl snapshots
for macOS, available at https://gretl.sourceforge.net/osx.html
(actually it has been fixed for a few weeks now).
Allin Cottrell
1 year
new gretl build for Windows 10 and higher
by Cottrell, Allin
Hello all,
This may be of interest to users of gretl on Windows. There's a new
snapshot in place with departs from our "traditional" packages in two
main ways: the GUI toolkit is GTK3, not GTK2, and we link against the
"new" Microsoft C library, UCRT (which has been supplied with Windows
since the first release of Windows 10). Several other components of
the package have also been updated. The snapshot can be found at
https://sourceforge.net/projects/gretl/files/snapshots/gretl_install-win6...
Our plan is to replace the previous 64-bit version of gretl for
Windows with this one as of the next release. However, we'd be
grateful if people could test it before then and report any problems.
If you're willing to test and do run into problems, there's an easy
fallback: uninstall the snapshot then reinstall one of the packages
available at https://gretl.sourceforge.net/win32/
Thanks,
Allin
1 year
Re: error evaluating 'if'
by Allin Cottrell
On Fri, 6 Sep 2024, Summers, Peter wrote:
> Hi all,
>
> When trying to use the console to create a new variable, I get an
> error message after typing "if (female == 1)" saying "error
> evaluating 'if'". [...]
To add just a little to what others have said: in hansl a valid
instance of
if <condition>
must produce a scalar result, telling the interpreter to go ahead
(if non-zero) or not (if zero), into a block of statements ended by
"endif".
If "female" is a series with both 0s and 1s then the result is not a
scalar and so can't tell gretl what to do.
However, it's OK for a ternary query such as "(female == 1) ?" to
produce a scalar, series or matrix result, depending on the context.
Allin
1 year, 1 month
Re: error evaluating 'if'
by Sven Schreiber
Am 06.09.2024 um 22:53 schrieb Summers, Peter:
>
> Hi all,
>
> When trying to use the console to create a new variable, I get an
> error message after typing “if (female == 1)” saying “error evaluating
> ‘if’”. I’ve also tried it with a script and get the same result. This
> is on Windows, latest snapshot (office desktop) and also with the
> previous win64.zip snapshot (classroom). I’ve tried spacing/no
> spacing, parentheses & not, all with no change.
>
If 'female' is a series, then 'female==1' will also yield a (dummy)
series, and that won't work with a boolean check, because what should
the result be? So an if-block needs a scalar/unique input for
evaluation. Perhaps you should give more context as to what you are
trying to achieve. For example, a ternary statement is often useful.
cheers
sven
1 year, 1 month
error evaluating 'if'
by Summers, Peter
Hi all,
When trying to use the console to create a new variable, I get an error message after typing "if (female == 1)" saying "error evaluating 'if'". I've also tried it with a script and get the same result. This is on Windows, latest snapshot (office desktop) and also with the previous win64.zip snapshot (classroom). I've tried spacing/no spacing, parentheses & not, all with no change.
Thanks,
Peter
Dr. Peter M. Summers| Associate Professor of Economics
Faculty Advisor, HPU Chess Club
One University Parkway, High Point, NC 27268
Office: 336-841-9650
Choose to be extraordinary!(r)
[Facebook logo]<https://www.facebook.com/HighPointU>[Twitter logo]<https://twitter.com/HighPointU>[Instagram logo]<http://instagram.com/highpointu>[LinkedIn logo]<https://www.linkedin.com/school/28206>
1 year, 1 month
Update: Latextab Package Version 1.2
by Artur T.
Dear gretl users,
I am pleased to announce the release of version 1.2 of the latextab
package. This update includes a crucial bugfix: in cases of empty cells,
the number of ampersands now correctly matches the number of columns,
addressing an issue that previously led to errors.
For more information about the latextab package, please visit:
https://gretl.sourceforge.net/current_fnfiles/latextab.gfn
Best regards,
Artur
1 year, 1 month
Re: [EXTERNAL] Re: bug in test stat calculator
by Sven Schreiber
Am 03.09.2024 um 17:54 schrieb Summers, Peter:
>
> Ok, I’ve sorted it out. It’s not a bug after all, but there is a typo
> in the documentation. Everything works if the restriction is (female =
> 0), but not with (female == 0). The documentation specifies “==” in
> the “var2 op val” line & below.
>
I see what you mean and I actually do think it's a bug. Years ago gretl
switched from "=" to "==" as the required (or sometimes only preferred)
equality testing operator, and this case apparently was overlooked or
got broken afterwards.
>
> I didn’t know this feature existed before today – I think my students
> will really like it!
>
Yeah, it's a hidden gem... (even if documented)
-s
1 year, 1 month
