[Gretl-users] GUI points
svetosch at gmx.net
Sat Nov 12 11:16:50 EST 2016
Am 11.11.2016 um 04:28 schrieb Allin Cottrell:
> 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.)
Thanks for this feature.
Before it's released, let me just ask whether this output might instead
be added to the output of "correlation matrix" in the same context menu.
Correlation is also a measure of linear association, and so it would
feel like a natural place IMO.
This would save the extra entry in the context menu, and instead of
asking the user whether or not to include a constant term, both variants
could just be printed out simultaneously. (Currently there's only a few
lines of output in the window when one selects "correlation matrix".)
While we're on the subject of the condition number and collinearity, I
have a question about the following example: Open the example data
hall.gdt and select the two variables "consrat" and "ewr"; the
correlation output (again, right-click and then select from the context
menu) then shows a corr coeff of just 0.16. The new collinearity
analysis gives a whopping 634, an order of magnitude greater than the
rule-of-thumb value 50. The bkw.gfn package confirms this value.
This strikes me as qualitatively very different, and spontaneously I'm
not sure why that is so. Any ideas?
More information about the Gretl-users