Abstract
Robert J. Casady, Alan H. Dorfman, and Suojin Wang
(1994) "Confidence Intervals for Sub-Domain Parameters When
the Sub-Domain Sample Size is Random," Proceedings of the
Section on Survey Research Methods, American Statistical
Association, forthcoming.
Let be a population sub-domain of interest and assume that
the elements of A cannot be identified on the sampling frame
and the number of elements in A is not known. Further assume
that a sample of fixed size (say n) is selected from the
entire frame and the resulting sub- domain sample size (say
nA) is random. The problem addressed is the construction of a
confidence interval for a sub-domain parameter such as the
sub-domain aggregate TA = ä i xi . The usual approach to
this problem is to redefine xi , by setting xi = 0 if i î A.
Thus, the construction of a confidence interval for the
sub-domain total is recast as the construction of a
confidence interval for a population total which can be
addressed (at least asymptotically in n) by normal theory. As
an alternative, we condition on and construct confidence
intervals which have approximately nominal coverage under
certain assumptions regarding the sub-domain population. We
evaluate the new approach empirically using data from the BLS
Occupational Compensation Survey.
Last Modified Date: July 19, 2008
|