Message-ID: <27058.40857.762563.725573@stat.math.ethz.ch>
Date: 2026-03-12T11:12:25Z
From: Martin Maechler
Subject: Typo/bug in ar.yw.default, ar.ols, ar.burg.default
In-Reply-To: <fake-VM-id.de6d212f5024b0393ce4a3f4031ef156@talos.iv>
>>>>> Simone Giannerini
>>>>> on Mon, 9 Mar 2026 12:43:07 +0100 writes:
> Dear all,
> I think there is a (small) bug in
> ar.yw.default (line 142 of the source file ar.R)
> ar.ols (line 429 of the source file ar.R)
> ar.burg.default (line 570 of the source file ar.R)
> maic <- min(aic)
> should be
> maic <- min(xaic)
> Note that, apparently, this typo/bug does little harm since the
> subsequent line of code
> xaic <- setNames(if(is.finite(maic)) xaic - min(xaic) else
> ifelse(xaic == maic, 0, Inf), 0L:order.max)
> does not use maic but min(xaic) directly. In any case, after fixing,
> one could replace min(xaic) with maic to avoid recomputing min(xaic).
> In absence of comments I will file a bug report.
I'm looking into this, thank you, Simone.
This is not really a bug, e.g., in the sense of
https://www.R-project.org/bugs.html ,
as you mention yourself that the code works (flawlessly).
Still, of course, (if you are right which I assume currently)
it is a code infelicity we will be happy to fix.
Best,
Martin
> Kind regards
> Simone
> Tested against R-devel 2026-03-06
> ___________________________________________________
> Simone Giannerini
> Dipartimento di Scienze Economiche e Statistiche
> Universita' di Udine
> Via Tomadini 30/A - 33100 Udine, ITALY
> Tel: +39 0432 249577
> https://simonegiannerini.net/