Message-ID: <536103FF.4020904@gmail.com>
Date: 2014-04-30T14:09:03Z
From: Gabor Bakos
Subject: arima.c bug?
In-Reply-To: <5360F6E1.5030508@gmail.com>
On 2014. April 30. 15:13:05, Gabor Bakos wrote:
> Hello,
>
> Reading the code I have found that in the line 653 (Pnew[i + r * j] =
> tmp;) of http://svn.r-project.org/R/trunk/src/library/stats/src/arima.c
> (latest as of now) the multiplier for j is r instead of rd (which is the
> dimension of the matrix). Was it intentional? That seems at least worth
> a comment why it is the case instead of Pnew[i + rd * j] = tmp;
> Thanks and Kind Regards, gabor
Sorry, I missed the if (d == 0) condition for that block. In that case
rd == d, so this do not cause bugs, just a bit confusion for a short
time. :)
Thanks, gabor