F-test: Fixed Effects vs Pooled OLS
by Bruno Gobeil
Hello!
I would like to better understand the rational for using the following formula for determining the denominator degrees of freedom for the F-test: n(t-1)-(k-1); where n is the number of regions, t the number of years, k the number of parameters, including the constant. I thought that the use of the de-meaned data approach compared to LSDV approach for the fixed effects model would lead to a higher number of denominator degrees of freedom, as suggested in the Gretl Guide, with the following formula: nt-k.
I would greatly appreciate if someone could shed some light on the matter.
Thank you
Bruno Gobeil
Senior Consultant | Consultant principal
DUNSKY ENERGY CONSULTING | DUNSKY EXPERTISE EN ÉNERGIE
[t] 514.504.9030 *26 [f] 514.289.2665 [e] bruno.gobeil(a)dunsky.com<mailto:bruno.gobeil@dunsky.com>
[cid:654DCECB-E50C-4670-80DE-CDA276C82480@home]<http://www.dunsky.com/>
NOTICE: This message (including any attachments) is intended for the sole use of the individual or entity to which it is addressed, and contains information that is privileged, confidential and exempt from disclosure. Barring explicit indication to the contrary in the text of the message, any copying, distribution or dissemination of any part of this email (including any attachments), by the intended or accidental recipient, is strictly prohibited. If you have received this communication in error, please notify us immediately. Thank You.
AVIS : Ce message (et les pièces qui y sont jointes) est destiné à l'usage exclusif de la personne ou de l'entité à laquelle il est adressé, et renferme des renseignements privilégiés et confidentiels qui ne doivent pas être divulgués. Sauf si le message invite explicitement au contraire, il est strictement interdit de copier, distribuer ou diffuser cette communication (et les pièces qui y sont jointes), en tout ou en partie, cette interdiction s'appliquant tant au récipiendaire voulu qu'involontaire. Si vous avez reçu cette communication par erreur, veuillez nous en aviser immédiatement. Merci.
8 years
Screencasting Tool
by Wingenroth, Thorsten
Hi Clive,
you were looking for a screencasting tool.
I came across ActivePresenter as screencasting tool. It's free as long as you only output video and it's highly professional.
I used it for a couple of screencasts and it is really comfortable. You can split your video, make certain screens appear longer, highlight and do voice recording afterwards and during capture, as you wish.
Here is the link: http://atomisystems.com/
Kind regards,
Thorsten
8 years
new Kalman interface
by Allin Cottrell
For anyone interested in working with state space modela via gretl:
please note that we have now replaced the documentation in chapter 30
of the Gretl User's Guide with an account of the new (bundle-style)
interface. We think the new code is now ready to roll, but if anyone
is able to do some testing before the next release -- which should be
in a week or so -- we'd be grateful.
The old Kalman interface is still available for now (but will likely
be removed in the release following the next one). And you can find
the old doc at
http://gretl.sourceforge.net/papers/kalman_old.pdf
Allin Cottrell
8 years
plottting with --band option: a follow-up
by Allin Cottrell
In my recent posting at
http://lists.wfu.edu/pipermail/gretl-users/2016-June/011934.html
I wasn't precise enough about the handling of the --band option when
the data are supplied in the form of a matrix or matrices.
In fact, this differs depending on whether you are using the
traditional "gnuplot" command or the new(ish) "plot" block. That's
because in the first case the supplied matrix is treated as a mini
dataset (and you index into it for the columns you want), while with
"plot" an initial matrix argument is taken as the data to plot (no
indexing required, or even permitted).
Anyway, here's an illustrative script (based on one devised by Jack
to point out the problem in my exposition, for which thanks):
<hansl>
scalar n = 20
scalar x = 7.5
matrix A = x ~ seq(1,n)' # the data to plot
matrix B = x ~ sqrt(seq(3,n+2)') # the band information
# block style: you need two matrices with --band
plot A
option band=B
end plot --output=display
# traditional style: you select columns of a single matrix
matrix C = A ~ sqrt(seq(3,n+2)')
gnuplot 1 2 --matrix=C --output=display --band=1,3
</hansl>
The two plots should be identical, but they require slightly
different routes for their production.
Allin Cottrell
8 years
Conditional Logit
by Ernesto Armijo
Hi everyone. I was wondering if anybody has coded a routine to estimate conditional logits in Gretl. I teach an introductory course one Econometrics and Gretl works perfectly for everything except this. I don't want to use R (again :-)
REgards,
Ernesto
8 years
