Re: [Gretl-devel] [Gretl-users] multivariate lrvar?
On Mon, 16 Apr 2018, Sven Schreiber wrote:
Am 16.04.2018 um 18:33 schrieb Allin Cottrell:
> I think it would be more intuitive if the option took the form of a request
> for action rather than for inaction. So that a nonzero second argument
> would mean, "please demean X first" rather than "hey, X is already
> demeaned".
Yes, why not.
And your take on the default? Action or no action? I tend to think that for
generic (multiple) variables there's no reason to expect they're demeaned
already (thus default "yes do demean them first"), but I'm not sure.
I'd be OK with making "true" (non-zero) the default for the
do-demeaning argument to lrcovar() -- assuming we make the flip that
I claimed as more intuitive above, which is not yet the case in git!
So, just to be explicit, the idea would then be:
V = lrcovar(X) # does de-meaning
V = lrcovar(X, 1) # does de-meaning (more explicitly)
V = lrcovar(X, 0) # cancels de-meaning
Is this preferable to what's in place right now? Which is:
V = lrcovar(X) # does de-meaning
V = lrcovar(X, 1) # cancels de-meaning
V = lrcovar(X, 0) # does de-meaning
Allin