Name
Abstract | ||
|---|---|---|
The basic idea underpinning the theory of spatially balanced sampling is that units closer to each other provide less information about a target of inference than units farther apart. Therefore, it should be desirable to select a sample well spread over the population of interest, or a spatially balanced sample. This situation is easily understood in, among many others, environmental, geological, biological, and agricultural surveys, where usually the main feature of the population is to be geo-referenced. Since traditional sampling designs generally do not exploit the spatial features and since it is desirable to take into account the information regarding spatial dependence, several sampling designs have been developed in order to achieve this objective. In this paper, we present the R package Spbsampling, which provides functions in order to perform three specific sampling designs that pursue the aforementioned purpose. In particular, these sampling designs achieve spatially balanced samples using a summary index of the distance matrix. In this sense, the applicability of the package is much wider, as a distance matrix can be defined for units according to variables different than geographical coordinates. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.18637/jss.v103.c02 | JOURNAL OF STATISTICAL SOFTWARE |
Keywords | DocType | Volume |
finite population, spatial dependence, spatial balance index, MCMC | Journal | 103 |
Issue | ISSN | Citations |
CN2 | 1548-7660 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Francesco Pantalone | 1 | 0 | 0.34 |
| Roberto Benedetti | 2 | 0 | 0.34 |
| Federica Piersimoni | 3 | 0 | 0.34 |