Spatial Remainder Sampling

摘要

In this paper, we consider a finite population U for which the units are arranged in a rectangular array of M ≠ ml rows and N ≠ nk columns from which a sample of size mn is to be drawn. New methods called spatial remainder stratified sampling and spatial remainder systematic sampling are introduced. When the remainders of rows and columns are both zero, the suggested method reduces to the usual spatial stratified sampling or spatial systematic sampling procedure. First-order inclusion probabilities are derived for our suggested spatial remainder sampling methods, yielding the Horvitz-Thompson estimator and its variance. The proposed sampling designs are compared with spatial simple random sampling for population exhibiting random order, linear trend and spatially correlated. Finally, a numerical comparison of the procedures is provided using data from a study of geothermal with CO_2 emission in Yellowstone National Park.