Premium
Inference under pivotal sampling: Properties, variance estimation, and application to tesselation for spatial sampling
Author(s) -
Chauvet Guillaume,
Le Gleut Ronan
Publication year - 2021
Publication title -
scandinavian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.359
H-Index - 65
eISSN - 1467-9469
pISSN - 0303-6898
DOI - 10.1111/sjos.12441
Subject(s) - sampling (signal processing) , sampling design , estimator , mathematics , statistics , inference , simple random sample , context (archaeology) , variance (accounting) , statistical inference , selection (genetic algorithm) , econometrics , computer science , artificial intelligence , population , demography , accounting , filter (signal processing) , sociology , business , computer vision , paleontology , biology
Unequal probability sampling is commonly used for sample selection. In the context of spatial sampling, the variables of interest often present a positive spatial correlation, so that it is intuitively relevant to select spatially balanced samples. In this article, we study the properties of pivotal sampling and propose an application to tesselation for spatial sampling. We also propose a simple conservative variance estimator. We show that the proposed sampling design is spatially well balanced, with good statistical properties and is computationally very efficient.