Spbsampling: an R package for spatially balanced sampling
Spbsampling: an R package for spatially balanced sampling
The basic idea underpinning the theory of spatial ly 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 spatial ly 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.
MCMC, finite population, spatial balance index, spatial dependence
Pantalone, Francesco
c1b85bef-a71c-4851-9807-7776bc0b5ded
Benedetti, Roberto
1197c065-613f-4074-b103-4cbd5dd9bec8
Piersimoni, Federica
38b51a17-1b20-4c0f-b18a-5cae8e255d57
24 August 2022
Pantalone, Francesco
c1b85bef-a71c-4851-9807-7776bc0b5ded
Benedetti, Roberto
1197c065-613f-4074-b103-4cbd5dd9bec8
Piersimoni, Federica
38b51a17-1b20-4c0f-b18a-5cae8e255d57
Pantalone, Francesco, Benedetti, Roberto and Piersimoni, Federica
(2022)
Spbsampling: an R package for spatially balanced sampling.
Journal of Statistical Software, 103.
(doi:10.18637/jss.v103.c02).
Abstract
The basic idea underpinning the theory of spatial ly 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 spatial ly 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.
Text
v103c02
- Version of Record
More information
Accepted/In Press date: 2 March 2022
Published date: 24 August 2022
Additional Information:
Publisher Copyright:
© 2022, American Statistical Association. All rights reserved.
Keywords:
MCMC, finite population, spatial balance index, spatial dependence
Identifiers
Local EPrints ID: 474371
URI: http://eprints.soton.ac.uk/id/eprint/474371
ISSN: 1548-7660
PURE UUID: 4383c907-c643-4e9c-ad27-d7713fb2f075
Catalogue record
Date deposited: 21 Feb 2023 17:33
Last modified: 17 Mar 2024 04:10
Export record
Altmetrics
Contributors
Author:
Francesco Pantalone
Author:
Roberto Benedetti
Author:
Federica Piersimoni
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics