Re: [Gretl-users] GUI points
My first thought is that "influential observations" should be part of
"Analysis" instead of "Tests". My second thought is that
'Collinearity' should be too. It's a characteristic of the data, as is/are
influential observations, which seems like analysis to me.
fwiw,
PS
Sent from my iPad
On Nov 10, 2016, at 10:28 PM, Allin Cottrell <cottrell(a)wfu.edu>
wrote:
I have one thing to announce and one question about gretl's menus.
Announcement: Following some off-list discussion, there's now a new item in the
right-click menu when you have two or more series selected in the main gretl window,
namely a check on collinearity, in the form of the condition number of a matrix composed
of those series. (You get the choice of including a constant in the matrix.)
Question: In the model menu there's a Tests menu and also an Analysis menu. If we
think of "Tests" as indicating hypothesis tests (of model specification and/or
regarding the properties of the error term) then there are two entries which don't
really belong there, and arguably would be better placed under "Analysis",
namely "Influential observations" and "Collinearity". However, if we
were to move them some users may start asking, "Whatever happened to the XXX
test?" So: should we move them?
Allin Cottrell
_______________________________________________
Gretl-users mailing list
Gretl-users(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users