8:07 a.m.
Hi again, let me put this information about new features in the
development (= not yet released) version into this separate message.
(And no, there's no schedule yet for when the next version will be
released.)
I'm not going to repeat the entire changelog which I linked to in the
previous message; this is just about the stuff that might be most
interesting for the end user:
- A hansl (scripting) function distance(): "evaluate the distance
between points on various metrics such as Euclidean, Manhattan, Hamming,
Chebyshev, Cosine and Mahalanobis"
- Also for scripting: you can now use the keyword "end" for indexing
into multi-element objects, to denote the last entry or item. I'm not
sure if this is already documented, by the way, but basically "end" can
stand in for rows(x), cols(x), nelem(x), depending on the context.
- Panel data time dimension: This handling has been much improved. For
example, creating a new panel dataset in the GUI is now much easier, you
can directly enter the wanted n and T dimensions. And restricting the
sample in the GUI is now also possible in the time dimension.
Again, this only concerns the next version (and current snapshots)!
thanks
sven
8:12 a.m.
Am 23.11.2021 um 09:07 schrieb Sven Schreiber:
Hi again, let me put this information about new features in the
development (= not yet released) version into this separate message.
(And no, there's no schedule yet for when the next version will be
released.)
I'm not going to repeat the entire changelog which I linked to in the
previous message; this is just about the stuff that might be most
interesting for the end user:
Hi everybody, here's another one: In dynamic panel (dpanel) GMM
estimation there will be a new option to "collapse" the
instruments/moment conditions, like Stata's xtabond2 package does. As
with all new features, testing is always good, so if you're interested
you could run the current git version and do that. (For
self-compilation, I believe new snapshots are not out yet, but usually
will follow soon.)
cheers
sven
8:23 a.m.
Sir,
Please, in addition to the new features, we are thirsty for the generalized
impulse response functions in the vector model.
We hope that it comes along with the upcoming version.
Thanks.
On Mon, Dec 6, 2021, 9:12 AM Sven Schreiber <svetosch(a)gmx.net> wrote:
Am 23.11.2021 um 09:07 schrieb Sven Schreiber:
> Hi again, let me put this information about new features in the
> development (= not yet released) version into this separate message.
> (And no, there's no schedule yet for when the next version will be
> released.)
>
> I'm not going to repeat the entire changelog which I linked to in the
> previous message; this is just about the stuff that might be most
> interesting for the end user:
Hi everybody, here's another one: In dynamic panel (dpanel) GMM
estimation there will be a new option to "collapse" the
instruments/moment conditions, like Stata's xtabond2 package does. As
with all new features, testing is always good, so if you're interested
you could run the current git version and do that. (For
self-compilation, I believe new snapshots are not out yet, but usually
will follow soon.)
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/
4:33 p.m.
Am 06.12.2021 um 09:23 schrieb Olasehinde Timmy:
Sir,
Please, in addition to the new features, we are thirsty for the
generalized impulse response functions in the vector model.
We hope that it comes along with the upcoming version.
This remains unlikely; the main reason (at least from my point of view)
being the following argument from Kilian & Lütkepohl (2017, SVAR
Analysis, chapter 18, pp. 617-618), with which I agree:
"... Pesaran and Shin (1998) explicitly advocate the use of the
generalized impulse responses ... as being more general than traditional
structural VAR impulse responses... They suggest that studying the
response of the model variables to reduced-form shocks is more general
because it avoids the use of economic identifying assumptions that may
be controversial. This argument is not persuasive because reduced-form
shocks have no economic interpretation and violate the ceteris paribus
assumption required for quantifying causal relationships in the data. At
best, these responses may serve as a description of the properties of
the data. They are not informative about the structural model. Moreover,
if we were actually interested in responses to reduced-form shocks in
linear VAR models, such responses could easily have been generated using
the standard tools..."
In gretl's context, these standard tools are for example the vma() and
irf() functions which allow you to calculate the reduced-form impulse
responses. However, it does seem to be true that there is no direct
option to get the reduced-form IRFs from the GUI. (Because typically
they are not of direct interest.)
If you really must see the reduced-form IRF related to the shock
variable x, here's what you can do in the GUI as a workaround: Use the
Choleski scheme and move x to the extreme position such that its shock
only affects its own equation on impact, nothing else.
cheers
Sven
1105
days inactive
1124
days old
3 comments
2 participants
participants (2)
-
Olasehinde Timmy -
Sven Schreiber