|
From: | Michael D Godfrey |
Subject: | Re: Overhaul of statistical distribution functions |
Date: | Wed, 21 Sep 2011 12:05:45 -0700 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:6.0.2) Gecko/20110906 Thunderbird/6.0.2 |
On 09/21/2011 12:00 PM, Rik wrote:
Good. I think that that is settled. I do not like it, but I guess that a Mac-onlyExpected results are [1/4, 1/2, 1/4] or basically p2, p 1, p 2. This really does seem to be something about the Mac OS implementation of lgamma. --Rik tolerance is the only way out. But, the Loader code should really be added... Michael |
[Prev in Thread] | Current Thread | [Next in Thread] |