Re: [Gretl-devel] difftest
Am 29.11.2017 um 08:42 schrieb Sven Schreiber:
Am 29.11.2017 um 01:34 schrieb Allin Cottrell:
> On Tue, 28 Nov 2017, Sven Schreiber wrote:
>> Furthermore, for small samples n=5..8, gretl appears to
calculate...
>> W+ and W-.
> We're looking for a z-statistic, and that's NA for the given sample
> size. If anyone wishes to pursue tiny-sample statistics in this area,
> that's fine by me.
Fair enough. It's just that my impression was that gretl already wanted
to go in that direction, because it is reporting some relevant critical
values and the W+/W- thingies.
Basically what I have in mind is this kind of change in nonparametric.c:
lines 1120-1122/3 are currently:
} else if (n > 5) {
pprintf(prn, " 5%% critical values: %d (two-tailed), %d
(one-tailed)\n", rank5[n-6][0], rank5[n-6][1]);
}
and would become something like this:
} else if (n > 5) {
if (!quiet) {
pprintf(prn, " 5%% %s: %d %s, %d %s\n",
_("critical values"), _("(two-tailed)"), _("(one-tailed)"),
rank5[n-6][0], rank5[n-6][1]);
}
z = (wp < wm) ? wp : wm;
}
where z later gets stored and returned as the test statistic. The
various inserted _(...) things are supposed to mark those strings for
translation, not sure if I got the syntax right there.
If this is not blatantly wrong I could try to test it locally.
cheers,
sven