In the first line of the Relative Efficiency section, the "prime" of T' isn't clear in 2 browsers - IE and Opera. I can only tell that it's there by looking for it. I wonder if a better variable can be used, such as R. 14:18, 5 April 2005 User:22.214.171.124
It was showing up fine for me in Mozilla, but T1 and T2 are much clearer. Cburnett 19:01, Apr 5, 2005 (UTC)
Umm... I'm not much of a statistician, but what's an 'unbiased statistic' as mentioned in the first line of the intro? Shouldn't it say 'an unbiased estimator T of a statistic theta? 126.96.36.199 (talk) 17:41, 19 November 2007 (UTC)
A statistic is a function of the data, so in this context it means more or less the same as estimator. --Zvika (talk) 17:51, 19 November 2007 (UTC)
This is all very interesting, but can someone add a layperson's explanation of "efficiency" to the article? When an estimator of a statistic is "efficient," does it mean that it is very accurate at estimating the various values in the sample? As opposed to an estimator that is "inefficient," which would seem to guess wrong a whole lot. Can someone confirm this and add that to the head of the article? Thanks -R —Preceding unsigned comment added by Docfink (talk • contribs) 16:51, 26 September 2008 (UTC)
The last sentence in Examples says: "Many workers prefer the sample median as an estimator of the mean, holding that the loss in efficiency is more than compensated for by its enhanced robustness in terms of its insensitivity to outliers." This is a letdown. It implies that instead of a rigorous definition of efficiency as an estimator, we are left with personal judgements as to the choice of estimator. I think this needs further development. I would expect an efficient estimator to also be robust and to work for all sample sizes. —Preceding unsigned comment added by Parveson (talk • contribs) 19:57, 21 February 2009
Good point. I changed the last paragraph. How do you like it? --Zvika (talk) 07:40, 22 February 2009 (UTC)
Thank you. I just added "...unity (100%)" for clarity. Parveson (talk) 16:35, 23 July 2009 (UTC)
The following discussion is an archived discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.
The result of the move request was: page moved per unopposed request. GTBacchus(talk) 16:41, 16 May 2010 (UTC)
The new title seems to outline the topic more precisely. One can tell what an efficient estimator is, but it’s hard to define the concept of efficiency itself. // stpasha » 00:33, 7 May 2010 (UTC) // stpasha » 00:33, 7 May 2010 (UTC)
The above discussion is preserved as an archive of a requested move. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.
The reference - Arcones M.A. "Bahadur efficiency of the likelihood ratio test" preprint - is just not good. I read first 2 pages, and found 2 mistakes: page 1, 2nd last sentence, should be "...versus H1: theta in Theta1", rather than "... not in...". Page 2, 3rd sentence, "For theta1 in Theta1 we would like that the p value converges to 0 as fast as possible." p-value is not defined for any theta in Theta1. so why do we use this reference??? Jackzhp (talk) 15:08, 13 March 2011 (UTC)