On Mon, Oct 14, 2024 at 10:17 AM Sven Schreiber
<sven.schreiber(a)fu-berlin.de> wrote:
Am 14.10.2024 um 14:57 schrieb Riccardo (Jack) Lucchetti:
On 14/10/2024 15:53, Cottrell, Allin wrote:
On Mon, Oct 14, 2024 at 9:25 AM Riccardo (Jack) Lucchetti
<p002264(a)staff.univpm.it> wrote:
more intuitive to use the xmin() function, as in "series m = xmin(x,
y)". The function, however, is at present scalar-only.
Just saying, the min() function will do what's wanted here, for 2 or
more series, at the cost of defining a list, as in
<hansl>
open data4-10
series mymin = min(deflist(ENROLL, CATHOL))
</hansl>
Aaaah, very clever, I hadn't thought of that. That said, I still think xmin(a,b)
would probably be a more straightforward solution for the average user. Any objection to
my patch?
Haven't looked at your patch (does it handle NAs as wanted?), but let me put the
xmax/xmin functions in perspective: I tend to find them a bit awkward, allowing only
_exactly_ two comparison arguments. Every time I use them they feel a bit clunky. I
believe their existence predates the generalization of max and min to matrix arguments.
Actually, I formulated my thoughts already 8 years ago, so let me just link to them:
https://gretlml.univpm.it/hyperkitty/list/gretl-devel@gretlml.univpm.it/m...
Thanks for the reminder. I agree that the current situation with
max/min and xmax/xmin is a bit murky (and as you say, not properly
documented). My preference would be to wait till after release then
figure out the best way to proceed.
Allin