Network Field Technician
by ashaikfe＠gmail.com
A Network Field Technician is responsible for network development and maintenance in the field for any organization. The individual works with the organization’s technical team and its clients to install, configure, maintain, and fix all LAN/WAN and other equipment issues, ensuring efficient network functionality.
The technician maintains computer equipment, mobile devices, phones, cabling, routers, switches, and printers, among others, to address all of the client’s networking requirements either remotely or at the clients’ premises.
Read More: https://www.fieldengineer.com/skills/network-field-technician
2 years, 1 month
New packages for autoregressive Probit and Hamilton filter
by Sven Schreiber
Hi everybody,
just a quick notice that recently two new contributed function packages
have seen the light of the gretl world:
- 'ARprobit', which implements the Kauppi&Saikkonen approach to estimate
a dynamic Probit model including an autoregressive term (in addition to
lags of the dependent dummy).
- 'hamilcycle', which applies the simple but already popular proposal by
Hamilton to do a regression-based filter of a time series to obtain a
cyclical component (based on an idea by Jack a year ago)
As always (sorry for being repetitive) you can get these and other
contributed packages online from within gretl ("fx" button at the
bottom, or through the menu File/Function packages/On server).
cheers
sven
2 years, 5 months
issues with the graph
by tpanag＠uom.edu.gr
Dear All,
I have installed gretl on two machines (desktop and laptop).
Graphs in my desktop appear fine.
In my laptop there is an issue with the graphs. The scale of the axis is an issue with numbers being too big and the graph too small.
Is there anything I need to do to fix it?
I have tried the options from the preferences , default graph scale.
I have also tried uninstalling and re installing.
Thanks for the help.
regards
Theo
2 years, 5 months
gretl virtual conference, June 3-4
by Allin Cottrell
Apologies if you get this more than once, but we wanted to reach
everyone.
We've posted on this topic before, but here's a follow-up. The 2021
gretl summer (virtual) conference has shaped up quite nicely. We
hope you'll agree if you take a look at the program: see
www.gretlconference.org .
This will take place over Zoom, on June 3 and 4, and our license
restricts us to 100 participants. 40+ people have registered so far,
so if you're interested this would be a good time to go ahead and
sign up (no fee), at the site mentioned above.
For the gretl team,
Allin
2 years, 6 months
Invitation to join the gretl Association
by Sven Schreiber
Dear all,
it is a great pleasure to introduce the gretl Association to the gretl
community. (The official name is "gretl Association e.V." because it was
founded under German law and e.V. stands for registered association.)
The gretl Association is formally independent from the gretl project,
and so we have a separate initial website online here:
https://gretl-assoc.github.io/en/
But of course we share the same spirit!
By browsing through the information on that web page you will see that
we already founded the association last year, but due to the pandemic it
took a while to get it up and running. Right now the elected
chairpersons are Artur Tarassow and yours truly.
Of course we welcome new members, so please consider if you could join.
Again, more information is on the web page above, but the most important
bit is probably that it costs €18 per year (or €12 for students). You do
not have to live in Germany (or have a specific citizenship), anybody
can join. Currently you can pay by bank transfer or via Paypal.
Also feel free to circulate this invitation in your networks.
Thank you and have a nice weekend
Sven
2 years, 6 months
the ketvals package
by Riccardo (Jack) Lucchetti
(with apologies for cross-posting)
Dear all,
I'd like to announce a new package, that's available on the gretl official
repository, written by Francesco Valentini and myself. It's called
"ketvals", and performs estimation of linear models with time-varying
parameters by using the non-parametric approach developed by Liudas
Giraitis, George Kapetanios and other co-authors.
Just to whet your appetite, I'm attaching one of the plots produced by the
package sample script.
Of course, comments are very welcome. Download and enjoy!
-------------------------------------------------------
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
-------------------------------------------------------
2 years, 6 months
"Data Error" from kpss_test/adf_test C/C++ api
by riles＠triton.net
Hello
The following code produces this error "Data error"
int list[2] = {1,0};
int kpss_error = ::kpss_test(2,list,dset,OPT_N,prn);
if (kpss_error!=0) {
errmsg(adf_error, prn);
}
When I use the same data set in the GUI everything works just fine.
From what I can tell it seems like the problem is because of 0 in list[2] = {1,0};
or int list[13] = {12,0,1,2,3,4,5,6,7,8,9,10,11};
Due I need to split the data set in half and have a data set with two variables were
one variable is the first half of the data and the second the second half of the data?
Because in using a dset with more than one variable and setting list[3] = {1,1} seems to work
The error comes from the following lines in (line 2097)static int real_kpss_test(int order, int varno, DATASET *dset,gretlopt opt, kpss_info *kinfo, PRN *prn)
(line 2114) if (varno <= 0 || varno >= dset->v) {
return E_DATA;
}
were varno is set at (line 2435) in the kpss_test(....) function
these are in adf_kpss.c
I also get the same error with ::adf_test(2,list,dset,OPT_N,prn);
were varno is set at line 2435 in the kpss_test(....) function
I also get the same error with ::adf_test(2,list,dset,OPT_N,prn);
Thank you
2 years, 6 months
gretl snapshots for macOS
by Allin Cottrell
A heads-up for Mac users. As you may have noticed, we've now figured
out how to "notarize" Gretl.app, so as to suppress the warning from
Gatekeeper when you go to install gretl. That's nice, but
notarization requires in-person access to a recent Mac, which I have
only intermittently.
From now on we'll be sure to get official releases notarized, but
it's likely that most of our snapshots will not be notarized (though
they'll always be signed). If you know the routine to work around
Gatekeeper (see http://gretl.sourceforge.net/osx.html ), hopefully
that won't be a big problem.
At some point I may acquire a Mac, in which case we'll be able to
notarize snapshots too.
Allin
2 years, 6 months
gretl 2021b released
by Allin Cottrell
See http://gretl.sourceforge.net/
Mac users note: we now have a specific macOS package for arm64, aka
"Apple silicon" or "M1".
Here's the change log:
2021-05-04 version 2021b
- New function trigamma(): second derivative of lngamma
- New function midasmult(): gives MIDAS multipliers
- New command "bds": BDS nonlinearity test
- gdtb data format: make the new variant the default
- tighten up on defbundle() syntax: no empty arguments
allowed
- array definition: enforce the specific choice of type
- "wls" command, with 0/1 weights: compute $yhat and $uhat
for the excluded observations
- "dpanel" estimation via the GUI: make the --dpdstyle flag
configurable
- "dpanel" --verbose switch: provide some information on the
instruments used
- "adf" command: support more accurate inference with
improved p-values or critical values, for the GLS case
in particular
- "xtab" command: clarify that the $result (matrix) accessor
is just for the bivariate case
- "clear" command: add --functions option to remove all hansl
functions from memory
- "restrict" command: add "inject" keyword to support use of
an array of strings to specify restrictions
- "param_names" keyword for nls, mle, gmm: accept an array of
strings as argument
- binary probit and logit: offer Estrella's pseudo R-squared
as an alternative to McFadden, via --estrella option
- The functions mean, sum, sd, var, wmean, wsd and wvar now
accept an optional boolean argument; specific to the
cross-sectional statistics produced when list arguments
are given, this allows skipping of missing values.
- hdprod(): introduce "shorthand" syntax (one argument only)
- dbnomics addon: speed up retrieval of data
- user-defined functions: allow overloading of numeric
arguments and return value (scalar, series or matrix)
- user-defined function parameters: enforce the "const" flag
more consistently
- gretl timer ("stopwatch"): provide a separate timer for each
level of function execution
- GUI usage of geoplot addon: make it easier to save an edited
map as GeoJSON
- GUI model specification dialog: enable interactive search in
lists of series
- GUI, saving contents of bundles: enable saving of integer
members
- GUI, handle dependencies when installing function packages
- GUI script editor: improve implementation of "smart tab"
- Fix bug: breakage in computing out-of-sample forecasts from
VECMs
- Fix bug: possible crash when displaying a matrix with very
long column names in the GUI program
- Fix bug: broken covariance matrix from mols() with a large
number of regressors
- Fix bug: crash on applying postfix "++" to a submatrix
- macOS: first release for arm64
--
Allin Cottrell
Department of Economics
Wake Forest University
2 years, 6 months
