Generalised additive point process models for natural hazard occurrence
Article This is the author accepted manuscript. The final version is available from Wiley via the DOI in this record. Point processes are a natural class of model for representing occurrences of various types of natural hazard event. Flexibly implementing such models is often hindered by intractable...
Published in: | Environmetrics |
---|---|
Main Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Wiley
2017
|
Subjects: | |
Online Access: | http://hdl.handle.net/10871/26714 https://doi.org/10.1002/env.2444 |
id |
ftunivexeter:oai:ore.exeter.ac.uk:10871/26714 |
---|---|
record_format |
openpolar |
spelling |
ftunivexeter:oai:ore.exeter.ac.uk:10871/26714 2024-09-15T18:23:36+00:00 Generalised additive point process models for natural hazard occurrence Youngman, BD Economou, T 2017 http://hdl.handle.net/10871/26714 https://doi.org/10.1002/env.2444 en eng Wiley Vol. 28, Iss. 4, June 2017, e2444 doi:10.1002/env.2444 http://hdl.handle.net/10871/26714 1180-4009 1099-095X Environmetrics 2018-05-04 Publisher's policy. Point process Generalised additive model Restricted maximum likelihood Nonhomogeneous Article 2017 ftunivexeter https://doi.org/10.1002/env.2444 2024-07-29T03:24:15Z Article This is the author accepted manuscript. The final version is available from Wiley via the DOI in this record. Point processes are a natural class of model for representing occurrences of various types of natural hazard event. Flexibly implementing such models is often hindered by intractable likelihood forms. Consequently, rates of point processes tend to be reduced to parametric forms, or the processes are discretised to give data of readily modelled `count-per-unit' type. This work proposes generalised additive model forms for point process rates. The resulting low-rank spatio-temporal representations of rates, coupled with the Laplace approximation, makes the restricted likelihood relatively tractable, and hence inference for such models possible. The models can also be interpreted from a regression perspective. The proposed models are used to estimate di erent types of Cox process and then spatio-temporal variation in European windstorms. Through a combination of thin plate and cubic regression splines, and their tensor product, established relationships between where windstorms occur and the state of the North Atlantic Oscillation are con rmed, and then expanded to bring detailed understanding of within-year variation, which has otherwise not been possible with count-based models. Willis Research Network Article in Journal/Newspaper North Atlantic North Atlantic oscillation University of Exeter: Open Research Exeter (ORE) Environmetrics 28 4 e2444 |
institution |
Open Polar |
collection |
University of Exeter: Open Research Exeter (ORE) |
op_collection_id |
ftunivexeter |
language |
English |
topic |
Point process Generalised additive model Restricted maximum likelihood Nonhomogeneous |
spellingShingle |
Point process Generalised additive model Restricted maximum likelihood Nonhomogeneous Youngman, BD Economou, T Generalised additive point process models for natural hazard occurrence |
topic_facet |
Point process Generalised additive model Restricted maximum likelihood Nonhomogeneous |
description |
Article This is the author accepted manuscript. The final version is available from Wiley via the DOI in this record. Point processes are a natural class of model for representing occurrences of various types of natural hazard event. Flexibly implementing such models is often hindered by intractable likelihood forms. Consequently, rates of point processes tend to be reduced to parametric forms, or the processes are discretised to give data of readily modelled `count-per-unit' type. This work proposes generalised additive model forms for point process rates. The resulting low-rank spatio-temporal representations of rates, coupled with the Laplace approximation, makes the restricted likelihood relatively tractable, and hence inference for such models possible. The models can also be interpreted from a regression perspective. The proposed models are used to estimate di erent types of Cox process and then spatio-temporal variation in European windstorms. Through a combination of thin plate and cubic regression splines, and their tensor product, established relationships between where windstorms occur and the state of the North Atlantic Oscillation are con rmed, and then expanded to bring detailed understanding of within-year variation, which has otherwise not been possible with count-based models. Willis Research Network |
format |
Article in Journal/Newspaper |
author |
Youngman, BD Economou, T |
author_facet |
Youngman, BD Economou, T |
author_sort |
Youngman, BD |
title |
Generalised additive point process models for natural hazard occurrence |
title_short |
Generalised additive point process models for natural hazard occurrence |
title_full |
Generalised additive point process models for natural hazard occurrence |
title_fullStr |
Generalised additive point process models for natural hazard occurrence |
title_full_unstemmed |
Generalised additive point process models for natural hazard occurrence |
title_sort |
generalised additive point process models for natural hazard occurrence |
publisher |
Wiley |
publishDate |
2017 |
url |
http://hdl.handle.net/10871/26714 https://doi.org/10.1002/env.2444 |
genre |
North Atlantic North Atlantic oscillation |
genre_facet |
North Atlantic North Atlantic oscillation |
op_relation |
Vol. 28, Iss. 4, June 2017, e2444 doi:10.1002/env.2444 http://hdl.handle.net/10871/26714 1180-4009 1099-095X Environmetrics |
op_rights |
2018-05-04 Publisher's policy. |
op_doi |
https://doi.org/10.1002/env.2444 |
container_title |
Environmetrics |
container_volume |
28 |
container_issue |
4 |
container_start_page |
e2444 |
_version_ |
1810463836672622592 |