3:28 p.m.
Dear Admin,
I am trying to plot the impulse response(irf) after estimating a 5-variable
VAR/VEC model, however, I realized that the combined irf graphics option
disappeared. But when I reduced the variables to four, it reappeared.
Was this a bug or something? Then I would like to know how to go about it.
Regards
Timmy.
5:51 p.m.
Am 14.01.2023 um 16:28 schrieb Olasehinde Timmy:
Dear Admin,
I am trying to plot the impulse response(irf) after estimating a
5-variable VAR/VEC model, however, I realized that the combined irf
graphics option disappeared. But when I reduced the variables to four,
it reappeared.
Was this a bug or something? Then I would like to know how to go about it.
This is by design, because it was thought that more than a 2x2 plot
wouldn't make sense. (At least I think that was the rationale.) I guess
one could argue that with today's larger screens this decision could be
reconsidered.
If scripting is OK with you, you could use the "multiplot" contributed
function package to combine the pre-produced individual plots.
cheers
sven
6:03 p.m.
Thanks for the swift response, sir. First, it would be appreciated if this
function could be reconsidered for us who don't have much
scripting experience.
I remembered vividly that this limitation wasn't present in some previous
versions.
You can as well put me through how to use the multiplot because I have
tried many times using it but to futile.
I patiently await your response.
Thanks.
On Sat, 14 Jan 2023, 6:51 pm Sven Schreiber, <sven.schreiber(a)fu-berlin.de>
wrote:
Am 14.01.2023 um 16:28 schrieb Olasehinde Timmy:
Dear Admin,
I am trying to plot the impulse response(irf) after estimating a
5-variable VAR/VEC model, however, I realized that the combined irf
graphics option disappeared. But when I reduced the variables to four, it
reappeared.
Was this a bug or something? Then I would like to know how to go about it.
This is by design, because it was thought that more than a 2x2 plot
wouldn't make sense. (At least I think that was the rationale.) I guess one
could argue that with today's larger screens this decision could be
reconsidered.
If scripting is OK with you, you could use the "multiplot" contributed
function package to combine the pre-produced individual plots.
cheers
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/
10:56 p.m.
Am 14.01.2023 um 19:03 schrieb Olasehinde Timmy:
Thanks for the swift response, sir. First, it would be appreciated if
this function could be reconsidered for us who don't have much
scripting experience.
I remembered vividly that this limitation wasn't present in some
previous versions.
Are you sure? I don't actively remember, but I also don't think anything
has changed in that area. When I previously wrote about a 2x2 plot, this
was rubbish, for 4 variables of course you already get a 4x4 plot, and
5x5 would be quite big indeed. But again, I agree that today's larger
screens could accommodate a little more.
Perhaps an idea would be to enable a middle road between an individual
plot and all N^2 plots at once, namely N row-wise joint plots or
something like that.
Anyway, with respect to the multiplot package, it so happens that we've
almost finished a slightly updated version, and once that's ready we
might give an explicit example with VAR IRFs.
BTW, I'm seeing there's a bug with 5 variables when I try to see the
multiple residual plot (not an IRF plot): there are only three panels
and some plots are overlaid over others.
cheers
sven
7:36 a.m.
Thanks for the feedback, sir. I believe you are in a position to help work
on that area of graphics just like in Eviews and STATA where freedom is
given to configure the number of variables to plot after estimation, and
even the kind of decomposition.
As you can as well see that there is a bug in the package already, I
believe it is high time to help us work jointly on these areas.
Moreover, in the SVAR package, the option for a joint graph would be a good
thing if can be made available in the next version.
Thanks.
On Sun, 15 Jan 2023, 11:56 pm Sven Schreiber, <sven.schreiber(a)fu-berlin.de>
wrote:
Am 14.01.2023 um 19:03 schrieb Olasehinde Timmy:
Thanks for the swift response, sir. First, it would be appreciated if this
function could be reconsidered for us who don't have much
scripting experience.
I remembered vividly that this limitation wasn't present in some previous
versions.
Are you sure? I don't actively remember, but I also don't think anything
has changed in that area. When I previously wrote about a 2x2 plot, this
was rubbish, for 4 variables of course you already get a 4x4 plot, and 5x5
would be quite big indeed. But again, I agree that today's larger screens
could accommodate a little more.
Perhaps an idea would be to enable a middle road between an individual
plot and all N^2 plots at once, namely N row-wise joint plots or something
like that.
Anyway, with respect to the multiplot package, it so happens that we've
almost finished a slightly updated version, and once that's ready we might
give an explicit example with VAR IRFs.
BTW, I'm seeing there's a bug with 5 variables when I try to see the
multiple residual plot (not an IRF plot): there are only three panels and
some plots are overlaid over others.
cheers
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/
710
days inactive
712
days old
5 comments
3 participants
participants (3)
-
atecon -
Olasehinde Timmy -
Sven Schreiber