On the optimality of balanced sampling

In model based sampling for finite populations it is known that, provided a certain variance condition is met and balanced samples are used, the best linear unbiased predictor of the population total is the expansion estimator. This note establishes a condition for balanced sampling to be the optimal sampling strategy, again using the criterion of mean square error. In a sense which is discussed, this condition is both necessary and sufficient.

MoreLess

Year of publication:	1986
Authors:	Tallis, G. M.
Published in:	Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 4.1986, 3, p. 141-144
Publisher:	Elsevier
Subject:	balanced sampling superpopulation prediction approach

More details

Type of publication:	Article
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10005137968