4:35 a.m.
although Jack is indeed always right and the SVAR package is obviously a
more natural place to do IRFs, my question was prompted by the need to
demonstrate the basics of VAR estimation in class - for this purpose the
GUI command and associated outoput menus "Tests", "Analysis" etc is
perfectly fit, except that I feel uncomortable to show confidence
intervals with unclear characteristics :-)
thanks!
Stefano
--
________________________________________________________________________
Stefano Fachin
Professore Ordinario di Statistica Economica
Dip. di Scienze Statistiche
"Sapienza" Università di Roma
P.le A. Moro 5 - 00185 Roma - Italia
Tel. +39-06-49910834
fax +39-06-49910072
web http://stefanofachin.site.uniroma1.it/
--
________________________________________________________
Le informazioni
contenute in questo messaggio di posta elettronica sono strettamente
riservate e indirizzate esclusivamente al destinatario. Si prega di non
leggere, fare copia, inoltrare a terzi o conservare tale messaggio se non
si è il legittimo destinatario dello stesso. Qualora tale messaggio sia
stato ricevuto per errore, si prega di restituirlo al mittente e di
cancellarlo permanentemente dal proprio computer.
The information
contained in this e mail message is strictly confidential and intended for
the use of the addressee only. If you are not the intended recipient,
please do not read, copy, forward or store it on your computer. If you have
received the message in error, please forward it back to the sender and
delete it permanently from your computer system.
6:04 a.m.
Am 22.11.2019 um 10:35 schrieb Stefano:
although Jack is indeed always right and the SVAR package is
obviously
a more natural place to do IRFs, my question was prompted by the need
to demonstrate the basics of VAR estimation in class - for this
purpose the GUI command and associated outoput menus "Tests",
"Analysis" etc is perfectly fit, except that I feel uncomortable to
show confidence intervals with unclear characteristics :-)
Nevertheless, my proposal is to (a) increase the number of bootstrap
iterations in there from 999 to 1999, (b) add a note about this to the
help text "Impulse response plots" that you get when you click on Help
in the IRF specification dialog. Anybody objecting?
(I don't think adding an option for the user is high priority because of
the existence of the SVAR addon.)
But while we're at the built-in IRF GUI, I don't find the current
structure of menu entries optimal. I'm talking about the VAR model
window, and there you have menu Plots/Impulse responses (various
entries), and secondly Analysis/Impulse responses. In the second item
you just get the numbers of the point estimates, so in a way it's OK
that it's not under "Plots". However, there doesn't seem to be any way
to get the numbers of the bootstrapped confidence bands which is a pity.
So if there were an option in the IRF Plot dialog to save the numbers to
a matrix, then the second entry under Analysis could be scrapped I guess.
thanks
sven
2:23 p.m.
On Fri, 22 Nov 2019, Stefano wrote:
although Jack is indeed always right and the SVAR package is
obviously a more
natural place to do IRFs, my question was prompted by the need to demonstrate
the basics of VAR estimation in class - for this purpose the GUI command and
associated outoput menus "Tests", "Analysis" etc is perfectly fit,
except
that I feel uncomortable to show confidence intervals with unclear
characteristics :-)
Hi Stefano,
I'm now working on this -- in the first instance, allowing the user
to specify a number of bootstrap iterations (plus increasing the
default, as Sven has suggested).
It's many years since this particular chunk of the gretl C code was
written. Meanwhile personal computer speeds and memory have of
course increased dramatically, so it's indeed time for an update.
Allin
4:51 a.m.
Prof. Allin
Please, we would appreciate it if you could also make the type of impulse
response flexible just as that of Eviews. I mean, selecting impulse types
such generalised, unit, standard error, cholesky.... etc.
Thanks.
On Fri, Nov 22, 2019, 8:23 PM Allin Cottrell <cottrell(a)wfu.edu> wrote:
On Fri, 22 Nov 2019, Stefano wrote:
> although Jack is indeed always right and the SVAR package is obviously a
more
> natural place to do IRFs, my question was prompted by the need to
demonstrate
> the basics of VAR estimation in class - for this purpose the GUI command
and
> associated outoput menus "Tests", "Analysis" etc is perfectly
fit,
except
> that I feel uncomortable to show confidence intervals with unclear
> characteristics :-)
Hi Stefano,
I'm now working on this -- in the first instance, allowing the user
to specify a number of bootstrap iterations (plus increasing the
default, as Sven has suggested).
It's many years since this particular chunk of the gretl C code was
written. Meanwhile personal computer speeds and memory have of
course increased dramatically, so it's indeed time for an update.
Allin
_______________________________________________
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/
1777
days inactive
1778
days old
3 comments
4 participants
participants (4)
-
Allin Cottrell -
Olasehinde Timmy -
Stefano -
Sven Schreiber