Post-Strata Based on Sample Quantiles

Abstract The standard method of creating post-strata is to define the boundaries strata on basis population characteristics auxiliary variables. Estimation treats as for stratified-sample estimation. Samples often contain empty requiring adjustment estimation procedure. To avoid post-strata, we propose using sample distribution function an variable post-strata. We show that large-sample efficiency sample-based post-stratification procedure same equivalently defined population-based In simulation, was slightly more efficient than classical Monte Carlo coverage a nominal 95% interval approximately and 94%