Thread

Re: Add xicorr(X, Y): support for the xi (ξ) correlation coefficient by Chatterjee

Andrey Borodin <x4mmm@yandex-team.ru> — 2025-09-08T09:03:18Z

> On 8 Sep 2025, at 12:20, Dean Rasheed <dean.a.rasheed@gmail.com> wrote:
> 
> However, even if those issues were addressed, my feeling is that this
> is too specialised to be considered for inclusion in core. The fact
> that it exists in Scipy and not a core python module is a hint at
> that. There are a lot of other Scipy stat functions, the vast majority
> of which aren't included in core Postgres.

Kind of +1.
IMO, it's better to have Spearman and Kendal coefficients before original research from 2019.
I don't see anything wrong with having all them in core eventually, cost of maintaining settled math tools must not be too high.


Best regards, Andrey Borodin.