|
From: | Peter Johansson |
Subject: | Re: [Bug-gsl] Error in sample code |
Date: | Wed, 27 Feb 2013 09:33:58 +1000 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:10.0.12) Gecko/20130108 Thunderbird/10.0.12 |
Hi, On 02/27/2013 02:57 AM, Frantisek Kluknavsky wrote:
are you sure that "buggy" and correct behavior are not confused here? I do not know, but look at http://en.wikipedia.org/wiki/Bias_of_an_estimator and search for "usual definition of sample variance". If I understand it correctly, the "buggy" formula was unbiased, while now it is biased again.The biased estimate is 21.5/n = 4.2984. The unbiased estimate is 21.5/(n-1) = 5.373. Note, that neither is buggy; both estimators are used and it's a matter of taste which one is preferred. One is unbiased, but the other one is more efficient.
HTH, Peter
[Prev in Thread] | Current Thread | [Next in Thread] |