... and talking about this, may be a good chance to review what´s the sense of gretl
to show Theil statistic in forecast output, with cross-section data. 🙂
Cheers
Juan

De: Sven Schreiber <svetosch@gmx.net>
Enviado: mércores, 3 de marzo de 2021 10:55
Para: Gretl development <gretl-devel@gretlml.univpm.it>
Asunto: [Gretl-devel] Theil's U1 for fcstats (reloaded)
 
Hi,
an old issue resurfaced here recently, and I dug out this old thread:
(hope the link works, it starts on May 6th, 2017:)

https://gretlml.univpm.it/hyperkitty/list/gretl-devel@gretlml.univpm.it/thread/FRXL3GIUXIP6POSIDMPN2SGQJAPTZSLG/#TUD5DBGI5KZCIGPRGRQ3FAIZABHSCGXH

The point is that we might want to have Theil's U1 measure in addition
to the U2 measure which is in fcstats() and in the guide. U1 has a
different interpretation, but also works if some values are zero.

The guide says in a footnote: "This statistic is sometimes called U 2,
to distinguish it from a related but different U defined in an earlier
work by Theil (1961). It seems to be generally accepted that the later
version of Theil’s U is a superior statistic, so we ignore the earlier
version here."
But I don't think it's purely a matter of superior or not, it's just
different.

(For U1 I'm quoting myself:
"U1 being basically, using some obvious notation:
numerator = sqrt(mean((my-mfc).^2))
denom = sqrt(mean(mfc.^2)) + sqrt(mean(my.^2))
out = numerator/denom
-- This only breaks down if all observations are zero and if those are
all correctly and exactly forecast.")

Would it be OK to simply extend the result of fcstats()?

thanks
sven
_______________________________________________
Gretl-devel mailing list -- gretl-devel@gretlml.univpm.it
To unsubscribe send an email to gretl-devel-leave@gretlml.univpm.it
Website: https://gretlml.univpm.it/postorius/lists/gretl-devel.gretlml.univpm.it/