Abstract submitted to the ESRA Conference, Prague, June 2007

<dd><h3>On standard errors of model-based small-area estimators</h3>

<dd><h4>Presenter:  Nicholas T. Longford, SNTL, Reading, England</h4>

<dd><p>Email: ntl@sntl.co.uk</p>

<p>
An estimator is derived for the mean squared error (MSE) of the empirical
Bayes and composite estimator of the local-area mean in the standard
small-area setting.  The motivation for it is that the inferential target
is a fixed quantity (not altered in hypothetical replications), whereas
its model representation is by a random variable.  The MSE estimator is a
composition of the established estimator based on the conditional
expectation of the random deviation associated with the area and a naive
estimator of the design-based MSE.  Its performance is assessed by
simulations.  Variants of this MSE estimator are explored and some
extensions outlined. The advantage of composition of estimators over
choice among them as alternatives is emphasised throughout.
</p><p>

Based on a paper to appear in <I>Survey Methodology</I> in 2007
<br><br><br>
May 2007