[Gretl-users] Re: correlation
Am 02.03.2023 um 13:54 schrieb Sven Schreiber:
...
And I suppose that "parametric correlation" means going to
View/Correlation matrix, where you can choose an arbitrary number of
variables.
...
- About the parametric thing, I'm noting that you get the pairwise
test for n=2, and a (significance) test for non-correlation also for
n>2, but the latter appears to be a joint test with respect to all
off-diagonal correlations. If I understand correctly, you want a
battery of pairwise tests.
Follow-up with a partial correction here: What you get with n>2 is not a
joint test, but gretl gives you a 5% critical value (dependent on the
sample size) that applies to each of the pairwise correlations (under
the null of no correlation, of course). So you have your test battery
already in that case.
It has been discussed that it probably would be good to also provide the
critical values at the 10% and 1% level, so as not to fall into the trap
of the "arbitrary tyranny of the 5% threshold". Perhaps this is going to
go into the next gretl release.
cheers
sven