서베이앤헬스폴리시리서치센터

Kim, S. W., Heeringa, S. G., and Solenberger, P.W. (2007). “Sample allocation under a population model and stratified inclusion probability proportional to size sampling,” in Proceedings of the Survey Research Methods Section, American Statistical Association, 3061-3068, Salt Lake City, Utah: Joint Statistical Meetings.

Abstract

Many studies have been conducted on sample allocation for stratified sampling designs. We consider optimal allocation for stratified sample designs in which there is a population regression model for the variable of interest. We present a flexible approach for optimal sample allocation to minimize the expected variance of the Horvitz-Thompson estimator under the population regression model. We show that the presence of the intercept in the model produces a comparatively complicated allocation problem. For one popular IPPS sampling scheme, we show how to determine stratum sample sizes based on optimization theory.

Key words: Superpopulation Regression Model, Stratified sampling, Sample allocation