the oddsratios function package will be retired
by Sven Schreiber
Dear all,
now that gretl 2023a is out, the contributed function package
"oddsratios.gfn" will be retired soon, because the functionality is now
included natively in the core gretl program. You can find these results
after a Logit or similar estimation under the Analysis menu, or in case
you are scripting, it's in the $model bundle (as the matrix
$model.oddsratios).
As always, if you need the function package after it is retired from the
server (for example because you cannot upgrade to gretl 2023a
immediately), you should keep the package installed on your local
machine, where it will continue to work. And perhaps make a backup copy
of it.
cheers
sven
2 years
Multivariate GARCH models
by Riccardo (Jack) Lucchetti
Hi all,
as you may know, we've had a function package called MGARCH, that I wrote,
for estimating BEKK models. I'm about to retire it, since two new packages
are now avaliable:
*** a package called BEKK, which can be seen as an updated version of
MGARCH. From the user's point of view, there are very few changes. The
most notable one is that some functions have been renamed. Therefore, if
you have scripts including the line
"include MGARCH.gfn"
you may want to change that to
"include BEKK.gfn"
and check for functions beginning with "Get_": those have now a different
name.
*** a new package called DCC, which, well, deals with DCC models. The
authors are Giulio Palomba, Luca Pedini and myself. It's a brand new
thing, so testing would be much appreciated.
More generally speaking, testing is always good, so if you're into
financial econometrics please give them a go and tell me is you like them.
-------------------------------------------------------
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
correlation
by Alison Loddick
Hi,
I'm sure this is an easy question for econometricians rather than me as a statistician, but when using non-parametric correlation GRETL gives significance values and only one correlation at a time, but when using parametric correlation, we get a matrix but no significance values. Correlation values are impacted by sample size; therefore, I use significance to give an idea of the relationship.
I also ask this question as when teaching using GRETL, a matrix for non-parametric correlation would be a preferred output that students can put in a document rather than creating one themselves from individual answers.
Thank you
Alison
Alison Loddick
BSc MSc PGCE CStat
Learning Development Tutor (Mathematics and Statistics)
Library and Learning Services
DDI +44 (0)1604 893502
[https://static.hobsons.co.uk/northampton/usermedia/Staff%20News/2017/New%...]
northampton.ac.uk
University of Northampton, Waterside Campus,
University Road, Northampton, NN1 5PH United Kingdom
University of Northampton: Transforming Lives and Inspiring Change www.northampton.ac.uk This e-mail is private and may be confidential and is for the intended recipient only. If you are not the intended recipient you are strictly prohibited from using, printing, copying, distributing or disseminating this e-mail or any information contained in it. We virus scan all E-mails leaving The University of Northampton but no warranty is given that this E-mail and any attachments are virus free. You should undertake your own virus checking. The right to monitor E-mail communications through our networks is reserved by us.
Disclaimer
The information contained in this communication from the sender is confidential. It is intended solely for use by the recipient and others authorized to receive it. If you are not the recipient, you are hereby notified that any disclosure, copying, distribution or taking action in relation of the contents of this information is strictly prohibited and may be unlawful.
This email has been scanned for viruses and malware, and may have been automatically archived by Mimecast, a leader in email security and cyber resilience. Mimecast integrates email defenses with brand protection, security awareness training, web security, compliance and other essential capabilities. Mimecast helps protect large and small organizations from malicious activity, human error and technology failure; and to lead the movement toward building a more resilient world. To find out more, visit our website.
2 years
mastodon account
by Artur T.
Dear all,
gretl has a mastodon account now: https://econtwitter.net/@gretl
Username: @gretl@econtwitter.net
Mastodon is part of the fediverse, and an opensource and
community-driven alternative to twitter.
If you're on mastodon: Please follow.
I have also enabled automatic cross-postings from twitter to mastodon
and vice versa.
Our twitter user name is: @gretl_stats
Best,
Artur
P.S.
Self-advertisement: If you're also on mastodon, let's connect! You can
find me under @atecon@mastodon.social
2 years
min Windows version
by Marcin Błażejowski
Hi,
is Windows 7 Pro supported by current snapshot installer? I'm asking
because I got a massage with info that my (windows) version is not
supported by a program.
Marcin
--
Marcin Błażejowski
2 years
Johnson-Neyman plots
by Marcelo Duarte
Hello all,
I would like to know if it is possible to do Johnson-Neyman plots for
interactions on Gretl.
If yes, how to do them.
Best regards,
Marcelo
2 years
Re: Gretl-users Digest, Vol 194, Issue 4
by Marcelo Duarte
>
> On Tue, 7 Mar 2023, Riccardo (Jack) Lucchetti wrote:
>
> > One of the things that would make this easier to do in hansl would be t=
> he=20
> > possibility (now absent, but discussed some time ago in a developer mee=
> ting=20
> > IIRC) to have more than one "band" specification in the gnuplot/plot=20
> > command.
>
> Anyway: if anyone's interested in developing a package for this kind of=20
> plots, here's a rough attempt to a starting point.
>
> <hansl>
> function matrix meff_mat(matrix coef, matrix vcv,
> const series x, scalar n[32])
>
> scalar sdx =3D sd(x)
> x0 =3D min(x) - 0.025*sdx
> x1 =3D max(x) + 0.025*sdx
> matrix X =3D x0 + seq(0, n-1) * ((x1 -x0)/(n-1))
> X =3D 1 ~ X'
> matrix m =3D X * coef
> matrix v =3D sumr(X .* (X*vcv))
> return X[,2] ~ m ~ sqrt(v)
> end function
>
> function void JNplot(const series x, const bundle mod, matrix select,
> scalar n[32])
>
> matrix coef =3D mod.coeff[select]
> matrix V =3D mod.vcv[select, select]
> matrix M =3D meff_mat(coef, V, x)
>
> PLT =3D M[,{2,1}]
> B =3D M[,{2,3}]
>
> plot PLT
> option with-lines
> options band=3DB,1.96 band-style=3Dfill
> end plot --output=3Ddisplay
>
> end function
>
> ### example
>
> set verbose off
> set seed 11080
> nulldata 96
>
> x =3D normal()
> z =3D normal()
> xz =3D x * z
> y =3D 1 + x - z + 0.25*xz + 0.8 * normal()
>
> ols y const x z xz
> sel =3D {2,4}
> JNplot(z, $model, sel)
> </hansl>
>
Thanky so much for your answers.
Currently I'm doing J-N plots on excel, but the time spent and the
probability of errors is very high, so it would be interesting having Gretl
doing the hard work.
This author http://mattgolder.com/interactions provides Stata code for
ploting J-N, both for a simple interaction and for three-way interactions
as well.
Best,
Marcelo
2 years
