4:15 p.m.
Hi Sven
Many thanks. You have understood my need. I'll try your solution. If it
works it saves me butchering the original xls panel datafile and then
uploading it with 10 panel units.
Wrt the LSDV fixed effects panel model test result for intercepts. It is
helpful up to a point. However, if equal intercepts are rejected, does one
resort to OLS pooled data with DVs to see which differ as this is normally
a matter of interest. I can see no way from the LSDV option itself to
uncover the panel unit dummies relative to the estimated value of the
constant.
On Tue, 6 Feb 2024, 15:45 Sven S., <svetosch(a)users.sourceforge.net> wrote:
Brian, you're replying to a ticket that has been closed for over
two
years, and the topic is unrelated. If the problem persists after my reply,
please re-post your question (or follow-ups) on the mailing list.
Having said that, you would have to choose Sample/Restrict by condition,
and then enter an expression like "unit != 6 && unit != 8" (without
the
quotes). If the series "unit" doesn't exist yet, create it with
"Add/Panel
group index" first (or similar wording).
Whether that qualifies as a "simple way", I don't know.
------------------------------
Sent from sourceforge.net because you indicated interest in
https://sourceforge.net/p/gretl/feature-requests/147/
To unsubscribe from further messages, please visit
https://sourceforge.net/auth/subscriptions/
4:47 p.m.
On 06/02/2024 17:15, Brian Revell wrote:
Hi Sven
Many thanks. You have understood my need. I'll try your solution. If it
works it saves me butchering the original xls panel datafile and then
uploading it with 10 panel units.
A bit of scripting makes this quite painless. Example:
<hansl>
open grunfeld.gdt
to_drop = {1,2,5,7}
smpl !(contains($unit, to_drop)) --restrict
</hansl>
-------------------------------------------------------
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
-------------------------------------------------------
6:26 p.m.
Am 06.02.2024 um 17:15 schrieb Brian Revell:
Wrt the LSDV fixed effects panel model test result for intercepts. It
is helpful up to a point. However, if equal intercepts are rejected,
does one resort to OLS pooled data with DVs to see which differ as
this is normally a matter of interest. I can see no way from the LSDV
option itself to uncover the panel unit dummies relative to the
estimated value of the constant.
OK, this is another question. You can save the unit-specific intercepts
in the fixed-effects model window through Save / Group-specific intercepts.
Typically, though, you would simply run panel FE. It is not normally
needed or wise to try to retain only a subset of the specific
intercepts. (Note that the standard way of estimating does not involve
literal LSDV, but uses the within transformation.)
cheers
sven
6:37 p.m.
Thanks Sven. Clear and helpful.
Brian
On Tue, 6 Feb 2024, 18:26 Sven Schreiber, <sven.schreiber(a)fu-berlin.de>
wrote:
Am 06.02.2024 um 17:15 schrieb Brian Revell:
Wrt the LSDV fixed effects panel model test result for intercepts. It is
helpful up to a point. However, if equal intercepts are rejected, does one
resort to OLS pooled data with DVs to see which differ as this is normally
a matter of interest. I can see no way from the LSDV option itself to
uncover the panel unit dummies relative to the estimated value of the
constant.
OK, this is another question. You can save the unit-specific intercepts in
the fixed-effects model window through Save / Group-specific intercepts.
Typically, though, you would simply run panel FE. It is not normally
needed or wise to try to retain only a subset of the specific intercepts.
(Note that the standard way of estimating does not involve literal LSDV,
but uses the within transformation.)
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/
351
days inactive
351
days old
3 comments
3 participants
participants (3)
-
Brian Revell -
Riccardo (Jack) Lucchetti -
Sven Schreiber