PLACE:  171 Durham

SPEAKER:
Jae Kwang Kim

TITLE:
Replication Variance Estimation for Multi-Phase Stratified Sampling

ABSTRACT:

In the two-phase sampling setup, the second phase sample is often a stratified sample based on the information observed in the first phase sample. For the total of a population characteristic, either the double
expansion estimator or the reweighted expansion estimator can be used.

In this paper, we propose a consistent replication variance estimation that is applicable both to the double expansion estimator and to the reweighted expansion estimator. The first phase sampling design is any design for which an  unbiased replication variance is applicable. The proposed method can be easily extended to the multi-phase sampling. An application to the 2000 Census  A.C.E. survey is illustrated.

Key Words : Variance Estimation, Double Sampling, Double Expansion Estimator

COFFEE: 12:45 a.m,104 Snedecor