Talk:Efficient estimator

Re User:Sieluzyc: I understand your point that "efficient estimator does not have to be unbiased". However, the article must define this notion. "Estimator whose variance is the lower bound in the Rao–Cramér inequality" is a definition (though maybe an obsolete one). "A sufficient statistic for the parameter to be estimated" is not. If you want to improve the article please provide the new definition rather than just deleting the old one. --Boris Tsirelson (talk) 21:13, 24 October 2012 (CEST)