Directional spatial autoregressive dependence in the conditional first- and second-order moments
In contrast to classical econometric approaches which are based on prespecified isotropic weighting schemes, we suggest that the spatial weighting matrix in the presence of directional dependencies should be estimated. We identify this direction based on different candidate neighbourhood sets. In th...
Published in: | Spatial Statistics |
---|---|
Main Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
Elsevier
2021
|
Subjects: | |
Online Access: | https://eprints.gla.ac.uk/306121/ |
id |
ftuglasgow:oai:eprints.gla.ac.uk:306121 |
---|---|
record_format |
openpolar |
spelling |
ftuglasgow:oai:eprints.gla.ac.uk:306121 2023-10-09T21:53:58+02:00 Directional spatial autoregressive dependence in the conditional first- and second-order moments Merk, M.S. Otto, Philipp 2021-03 https://eprints.gla.ac.uk/306121/ unknown Elsevier Merk, M.S. and Otto, P. <http://eprints.gla.ac.uk/view/author/74142.html> (2021) Directional spatial autoregressive dependence in the conditional first- and second-order moments. Spatial Statistics <https://eprints.gla.ac.uk/view/journal_volume/Spatial_Statistics.html>, 41, 100490. (doi:10.1016/j.spasta.2020.100490 <https://doi.org/10.1016/j.spasta.2020.100490>) Articles PeerReviewed 2021 ftuglasgow https://doi.org/10.1016/j.spasta.2020.100490 2023-09-21T22:09:44Z In contrast to classical econometric approaches which are based on prespecified isotropic weighting schemes, we suggest that the spatial weighting matrix in the presence of directional dependencies should be estimated. We identify this direction based on different candidate neighbourhood sets. In this paper, we consider two different types of processes – namely spatial autoregressive and spatial autoregressive conditional heteroscedastic processes – and derive the consistency of the corresponding maximum likelihood estimates in the presence of directional dependencies. Moreover, Monte Carlo simulation results indicate that the model’s performance improves with sample size and with smaller neighbourhood subset sizes. Finally, we apply this approach to aerosol observations over the North Atlantic region and show that their spatial dependence matches the direction of the trade winds in this area. Article in Journal/Newspaper North Atlantic University of Glasgow: Enlighten - Publications Spatial Statistics 41 100490 |
institution |
Open Polar |
collection |
University of Glasgow: Enlighten - Publications |
op_collection_id |
ftuglasgow |
language |
unknown |
description |
In contrast to classical econometric approaches which are based on prespecified isotropic weighting schemes, we suggest that the spatial weighting matrix in the presence of directional dependencies should be estimated. We identify this direction based on different candidate neighbourhood sets. In this paper, we consider two different types of processes – namely spatial autoregressive and spatial autoregressive conditional heteroscedastic processes – and derive the consistency of the corresponding maximum likelihood estimates in the presence of directional dependencies. Moreover, Monte Carlo simulation results indicate that the model’s performance improves with sample size and with smaller neighbourhood subset sizes. Finally, we apply this approach to aerosol observations over the North Atlantic region and show that their spatial dependence matches the direction of the trade winds in this area. |
format |
Article in Journal/Newspaper |
author |
Merk, M.S. Otto, Philipp |
spellingShingle |
Merk, M.S. Otto, Philipp Directional spatial autoregressive dependence in the conditional first- and second-order moments |
author_facet |
Merk, M.S. Otto, Philipp |
author_sort |
Merk, M.S. |
title |
Directional spatial autoregressive dependence in the conditional first- and second-order moments |
title_short |
Directional spatial autoregressive dependence in the conditional first- and second-order moments |
title_full |
Directional spatial autoregressive dependence in the conditional first- and second-order moments |
title_fullStr |
Directional spatial autoregressive dependence in the conditional first- and second-order moments |
title_full_unstemmed |
Directional spatial autoregressive dependence in the conditional first- and second-order moments |
title_sort |
directional spatial autoregressive dependence in the conditional first- and second-order moments |
publisher |
Elsevier |
publishDate |
2021 |
url |
https://eprints.gla.ac.uk/306121/ |
genre |
North Atlantic |
genre_facet |
North Atlantic |
op_relation |
Merk, M.S. and Otto, P. <http://eprints.gla.ac.uk/view/author/74142.html> (2021) Directional spatial autoregressive dependence in the conditional first- and second-order moments. Spatial Statistics <https://eprints.gla.ac.uk/view/journal_volume/Spatial_Statistics.html>, 41, 100490. (doi:10.1016/j.spasta.2020.100490 <https://doi.org/10.1016/j.spasta.2020.100490>) |
op_doi |
https://doi.org/10.1016/j.spasta.2020.100490 |
container_title |
Spatial Statistics |
container_volume |
41 |
container_start_page |
100490 |
_version_ |
1779317359124676608 |