gretl crashed while searching for file within GUI
by Fred Engst
Hi Allin,
gretl crashed once again, but for a different reason this time.
It seems that whenever I tried to do a search for files within GUI after “file->Open data->User file...” or “file->Session files->Open session…”, it crashes right after it showed a list of files.
I can’t remember a time when I did a search like this without gretl crashing.
I suspect it has something to do with my character encoding. Even though I use standard English, I have Chinese on my system also.
This is more like a FYI, and it is not urgent, as I know how to get around it easily.
Attached is the crash report if it helps.
Fred
4 years, 10 months
ERROR: sample script gmm
by Artur Tarassow
Hi,
I've just tried to run the gretl sample script "hall_cbapm". It results,
however, in an error for this example:
<hansl>
# iterated estimation, identity matrix for initial weights
gmm e = delta*ewr*consrat^(alpha-1) - 1
orthog e ; inst
weights V0
params alpha delta
end gmm --iterate
</hansl>
The error message on ubuntu 19.04 using latest git is:
<error>
# iterated estimation, identity matrix for initial weights
? gmm e = delta*ewr*consrat^(alpha-1) - 1
? orthog e ; inst
? weights V0
? params alpha delta
? end gmm --iterate
Tolerance = 1.81899e-12
Norm of gradient 7.45054 exceeds maximum of 5
Error executing script: halting
> end gmm --iterate
</error>
As this is an sample script, it should not throw an error I would say.
Best,
Artur
4 years, 12 months
GretlGraphiteConnector
by Artur Tarassow
Hi all,
in case you use the well-known Graphite monitoring tool
(https://en.wikipedia.org/wiki/Graphite_(software)) you may have missed
the possibility to send metrics from gretl to Graphite. This has an end ;-)
I've written a package which sends metrics to Graphite through netcat by
means of executing shell commands (this must be enabled in your gretl
preferences. I use it for daily work and haven't experienced any issues yet.
URL: https://github.com/atecon/graphite_gretl_connector
I write this here, as I think this is not a package which should be part
of the official user-written package stack. But maybe I am wrong.
Enjoy!
Artur
4 years, 12 months
Re: Full Information Maximum Likelihood (VECM)
by macrocs＠juno.com
Thanks, Sven for your helpful suggestion Best, César
---------- Original Message ----------
From: Sven Schreiber <svetosch(a)gmx.net>
To: gretl-users(a)gretlml.univpm.it
Subject: [Gretl-users] Re: Full Information Maximum Likelihood (VECM)
Date: Thu, 11 Jul 2019 16:09:50 +0200
Am 11.07.2019 um 14:49 schrieb macrocs(a)juno.com:
> To whom may it concern
> Do you know the instructions for a Full Information Maximum Likelihood?
> I'd like to run a VECM including short-run restrictions (common cycle
> restrictions) (Vahid and Engle (1993)). Thanks.
Hi Cesar,
I think the best way would be something that PcFiml/PcGive/Oxmetrics
used to do in the old days: estimate the cointegration relations first
and add the estimated level ECTs to a system of differences. In this
second system you can then impose arbitrary restrictions.
(In a 'vecm' in gretl you can only restrict beta and alpha AFAIK.)
See also ch. 31 of the guide (multivariate models).
So roughly:
lagorder = 4
vecm lagorder 1 x y
series ec = $ec # need a loop for rank > 1
diff x y
list dxlags = lags(lagorder-1, d_x)
list dylags = lags(lagorder-1, d_y)
vahidengle <- system
equation d_x const ec(-1) dxlags dylags
equation d_y const ec(-1) dxlags dylags
end system
restrict vahidengle
... # concrete coeff restrictions
end restrict
estimate vahidengle method=fiml
However, I don't remember the type of the restrictions needed in this
case. If they also involve the levels/EC terms, it could be that some
auxiliary identities must also be defined, "explaining" to gretl the
relationship between the EC term and the (lagged) differences. (Using
'identity' lines in the system block as mentioned in the guide.)
hth,
sven
_______________________________________________
Gretl-users mailing list -- gretl-users(a)gretlml.univpm.it
To unsubscribe send an email to gretl-users-leave(a)gretlml.univpm.it
Website: https://gretlml.univpm.it/postorius/lists/gretl-users.gretlml.univpm.it/
____________________________________________________________
Sad News For Meghan Markle And Prince Harry
track.volutrk.com
http://thirdpartyoffers.juno.com/TGL3131/5d31f0f8878d870f85e7est02duc
5 years
gretl input/output with octave
by Allin Cottrell
I wrote a few days ago, in response to Klára Major's report of
problems with gretl-octave I/O on Windows:
<quote>
The functions gretl_loadmat() and gretl_export() have an optional
second argument: set it to zero and gretl will not force use of your
AddData\Roaming subdirectory for input and output.
<hansl>
matrix m = I(3)
mwrite(m, "c:/nice/path/m.mat")
foreign language=octave
mymat = gretl_loadmat("c:/nice/path/m.mat", 0);
mymat *= 2;
gretl_export(mymat, "c:/nice/path/mymat", 0);
end foreign
gotback = mread("c:/nice/path/mymat")
print gotback
</hansl>
</quote>
That's right, however I realized two things:
(1) The business of the optional trailing argument to gretl_loadmat
and gretl_export was not mentioned in the section of the Gretl
User's Guide dealing with Octave (though it was mentioned in the
Python and Julia sections). That's now fixed in git.
(2) Pretty obviously, if the user gives an absolute path then that
should automatically be respected, with no need to append a trailing
0 argument to prevent the prepending of "dotdir". That too is now
fixed in git (and snapshots).
Allin
5 years
using levels and log transformed data in forecasting
by Yusuf Abduwahab Hassan
Good morning all,
Please how can i re-transform the forecasts of the estimations in
log-transformed data back to levels form?
--
*Yusuf Abdulwahab Hassan.Department of Economics and Development
Studies.Federal University of Kashere,Gombe.+234
8036830166.yabdulwahab(a)fukashere.edu.ng <yabdulwahab(a)fukashere.edu.ng>*
5 years
starting octave from gretl
by Klára Major
Hello,
I think I need help to solve this problem. It is very basic, but this is
the first time I wish to run an octave script from gretl.
To learn this I created this very simple example based on Users' manual:
open data9-7
matrix xy = { PRIME, UNEMP }
mwrite(xy, "xy.mat", 1)
foreign language=Octave
xy = gretl_loadmat("xy.mat");
x = xy(:,1);
y = xy(:,2);
h = sum(x.*y);
gretl_export(h, "h.mat");
end foreign
When I try to run this from gretl, I got an error message, this one:
gretl version 2019c
Current session: 2019-07-13 11:39
? open data9-7
Read datafile D:\Programok\gretl\data\ramanathan\data9-7.gdt
periodicity: 4, maxobs: 64
observations range: 1975:1 to 1990:4
Listing 12 variables:
0) const 1) QNC 2) PRICE 3) INCOME 4) PRIME
5) UNEMP 6) STOCK 7) POP 8) WINTER 9) SPRING
10) SUMMER 11) FALL
? matrix xy = { PRIME, UNEMP }
Generated matrix xy
? mwrite(xy, "xy.mat", 1)
? foreign language=Octave
? xy = gretl_loadmat("xy.mat");
? x = xy(:,1);
? y = xy(:,2);
? h = sum(x.*y);
? gretl_export(h, "h.mat");
? end foreign
error: writing file 'C:\Users\MAJOR KLÁRA\.octave_hist': No such file or
directory
error: ignoring const execution_exception& while preparing to exit
I checked, the .octave_hist file is there. Also, when octave is running in
gui mode, it is possible to run gretltmp.m in this directory.
So it seems like the script should work unless this problem remains
unsolved.
So I am anaware, the problem is with gretl or with octave?
Is it possible that the problem is with the path (it has space and an
accented character in it)?
When I set preferences in gretl and define where octave executable is
located, which file should I choose?
octave-cli.exe
octave-cli-5.1.0.exe
Or one of the bat files?
(These are in the bin library of octave. The vbs files in octave root
definitely do not work.)
And finally, I have not found any documentation on these technical details,
this is why I need your help.
Thank you in advance,
Klára
5 years
matlab_utilities updated
by Summers, Peter
Fellow gretl users,
There's a new version of my matlab_utilities function available. It now includes an "eye()" function that mimics the Matlab/Octave version. In particular it allows for eye(r,c), appending rows or columns of zeros to an identity matrix if r > c or r < c, respectively. Note that gretl's native I() function now has this behavior as of a few days ago (thanks, Allin!).
If any of you have ideas for additions to this package that'd make your Matlab/Octave-to-gretl conversion easier, or any other comments on this package, please let me know (off list if you like).
Cheers,
Peter
[High Point University logo]
Dr. Peter M. Summers | Associate Professor of Economics
One University Parkway, High Point, NC 27268
Office: 336-841-9650 | Department of Economics
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>
5 years
Engle and Granger Method (ADF test for cointegration)
by marta.regulez＠ehu.es
Using in gretl the Engle and Granger method to test for cointegration, if you set a maximum lag order allowing gretl to choose the optimum lag order using AIC, you do not get the information on the number of lags actually chosen in the ADF test on the residuals of the cointegrating regression. You have it on the ADF test in the series though.
It would be nice to have this information. Here ther is an example of it:
Etapa 1: contrastando la existencia de una raíz unitaria en Ibexlog
Contraste aumentado de Dickey-Fuller para Ibexlog
contrastar desde 12 retardos, con el criterio AIC
tamaño muestral 310
hipótesis nula de raíz unitaria: a = 1
contraste con constante
incluyendo 0 retardos de (1-L)Ibexlog
modelo: (1-L)y = b0 + (a-1)*y(-1) + e
valor estimado de (a - 1): -0.0100412
Estadístico de contraste: tau_c(1) = -1.52429
valor p 0.5201
Coef. de autocorrelación de primer orden de e: 0.063
Etapa 2: contrastando la existencia de una raíz unitaria en Cambireallog
Contraste aumentado de Dickey-Fuller para Cambireallog
contrastar desde 12 retardos, con el criterio AIC
tamaño muestral 298
hipótesis nula de raíz unitaria: a = 1
contraste con constante
incluyendo 12 retardos de (1-L)Cambireallog
modelo: (1-L)y = b0 + (a-1)*y(-1) + ... + e
valor estimado de (a - 1): -0.0157084
Estadístico de contraste: tau_c(1) = -2.32797
valor p asintótico 0.1631
Coef. de autocorrelación de primer orden de e: 0.014
diferencias retardadas: F(12, 284) = 8.474 [0.0000]
Etapa 3: regresión cointegrante
Regresión cointegrante -
MCO, usando las observaciones 1990:02-2015:12 (T = 311)
Variable dependiente: Ibexlog
Coeficiente Desv. Típica Estadístico t valor p
---------------------------------------------------------------
const 8.43634 2.22064 3.799 0.0002 ***
Cambireallog 0.0895818 0.487804 0.1836 0.8544
Media de la vble. dep. 8.844108 D.T. de la vble. dep. 0.552485
Suma de cuad. residuos 94.61394 D.T. de la regresión 0.553348
R-cuadrado 0.000109 R-cuadrado corregido -0.003127
Log-verosimilitud −256.2467 Criterio de Akaike 516.4935
Criterio de Schwarz 523.9730 Crit. de Hannan-Quinn 519.4832
rho 0.989884 Durbin-Watson 0.013519
Etapa 4: contrastando la existencia de una raíz unitaria en uhat
Contraste aumentado de Dickey-Fuller para uhat
contrastar desde 12 retardos, con el criterio AIC
tamaño muestral 310
hipótesis nula de raíz unitaria: a = 1
modelo: (1-L)y = (a-1)*y(-1) + e
valor estimado de (a - 1): -0.0101159
Estadístico de contraste: tau_c(2) = -1.53386
valor p 0.7501
Coef. de autocorrelación de primer orden de e: 0.067
Hay evidencia de una relación cointegrante si:
(a) La hipótesis de existencia de raíz unitaria no se rechaza para las variables individuales y
(b) La hipótesis de existencia de raíz unitaria se rechaza para los residuos (uhat) de la regresión cointegrante.
5 years
