Empirical saddlepoint approximations of the Studentized mean under stratified random sampling

Abstract

We obtain a saddlepoint approximation for the Studentized mean under stratified random sampling. This is of theoretical interest only, since its calculation requires knowledge of the entire population. We also obtain an empirical saddlepoint approximation based on the stratified sample alone. This empirical approximation can be used for tests of significance and confidence intervals for the population mean and can be regarded as a saddlepoint approximation to a bootstrap approximation. We compare the empirical approximation to the true saddlepoint approximation, both theoretically, to obtain results on the relative error of the bootstrap, and numerically, using Monte Carlo methods to compare the true and bootstrap distributions with the saddlepoint approximations.