Re: [Gretl-devel] 2017d release?
On Tue, 24 Oct 2017, Sven Schreiber wrote:
Am 23.10.2017 um 21:28 schrieb Allin Cottrell:
> I think we have amassed enough bug-fixes and enhancements that a gretl
> 2017d release is now in order. Any issues that people reckon should be
> addressed before we go ahead?
Nothing that is super-pressing I guess. Nevertheless...
-- What about bug #195?
https://sourceforge.net/p/gretl/bugs/195/
(Quote from my own latest comment: "How can one retrieve the test results of
the automatically printed tests (Breusch-Pagan idiosyncratic variance etc.)
in a script way?")
eval $model.bp_test
(Inspecting a $model bundle and seeing what extra accessors it
contains is something that needs to be documented.)
-- A part (or diversion) of bug #215:
https://sourceforge.net/p/gretl/bugs/215/
The new unbalanced option is only available via command line, not GUI;
Allin's reply: "Should be added at some point. It's on my to-do list."
That's now done.
-- Feature request #102 seems easy if there is a consensus, it would
be just
a matter of changing the default:
https://sourceforge.net/p/gretl/feature-requests/102/
(" make the "exact" unbalanced RE method of Swamy/Arora the default for
gretl. Currently, it is an option. ")
Actually I don't want to make it the default. I don't think the
Baltagi-Chang modification is particularly helpful, and in light of
that I'd rather keep our default in line with Stata's.
http://ricardo.ecn.wfu.edu/~cottrell/gretl/random-effects.pdf
That's all for now; I'll try to get to Sven's other points later.
Allin