Re: [Gretl-devel] gretl 2018a?
Am 03.02.2018 um 09:54 schrieb Riccardo (Jack) Lucchetti:
On Fri, 2 Feb 2018, Allin Cottrell wrote:
> I think we've now accumulated enough bug fixes and refinements to make
> it worth hoisting 2017e out of limbo as release 2018a.
>
> http://gretl.sourceforge.net/ChangeLog.html
>
> Are there any issues that people would really like to see addressed
> before we do that?
I'm perfectly fine with this.
I, too. (This time I really should update the German translation, I guess.)
- We should finally get rid of bug 195
(https://sourceforge.net/p/gretl/bugs/195/) I think; this just means to
mention Baltagi/Li in the manual ch. 20 for the Breusch-Pagan test of
individual effects under "Testing panel models". An extra bonus would be
to explicitly distinguish that Breusch-Pagan test in the printed
estimation output from the one of heteroskedasticity.
(Baltagi/Li (1990), A lagrange multiplier test for the error components
model with incomplete panels, Econometric Reviews, 9, pp. 103-107)
- Feature request 103
(https://sourceforge.net/p/gretl/feature-requests/103/) is about the
documentation of estimation of unbalanced panels. Do I understand
correctly that Allin's gretl working paper No. 4 was about that? Then
all that's needed is to put a reference to that working paper into the
guide.
- Finally let me mention request 101
(https://sourceforge.net/p/gretl/feature-requests/101/) because in the
accompanying mailing list discussion in September Allin answered "No
doubt there should [be a scripting access]". But there hasn't been much
demand apparently, so not very urgent it seems.
cheers,
sven