Deep Integro-Difference Equation Models for Spatio-Temporal Forecasting
Integro-difference equation (IDE) models describe the conditional dependence between the spatial process at a future time point and the process at the present time point through an integral operator. Nonlinearity or temporal dependence in the dynamics is often captured by allowing the operator param...
Main Authors: | , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
arXiv
2019
|
Subjects: | |
Online Access: | https://dx.doi.org/10.48550/arxiv.1910.13524 https://arxiv.org/abs/1910.13524 |
id |
ftdatacite:10.48550/arxiv.1910.13524 |
---|---|
record_format |
openpolar |
spelling |
ftdatacite:10.48550/arxiv.1910.13524 2023-05-15T17:31:23+02:00 Deep Integro-Difference Equation Models for Spatio-Temporal Forecasting Zammit-Mangion, Andrew Wikle, Christopher K. 2019 https://dx.doi.org/10.48550/arxiv.1910.13524 https://arxiv.org/abs/1910.13524 unknown arXiv https://dx.doi.org/10.1016/j.spasta.2020.100408 arXiv.org perpetual, non-exclusive license http://arxiv.org/licenses/nonexclusive-distrib/1.0/ Machine Learning stat.ML Machine Learning cs.LG Methodology stat.ME FOS Computer and information sciences article-journal Article ScholarlyArticle Text 2019 ftdatacite https://doi.org/10.48550/arxiv.1910.13524 https://doi.org/10.1016/j.spasta.2020.100408 2022-03-10T16:32:56Z Integro-difference equation (IDE) models describe the conditional dependence between the spatial process at a future time point and the process at the present time point through an integral operator. Nonlinearity or temporal dependence in the dynamics is often captured by allowing the operator parameters to vary temporally, or by re-fitting a model with a temporally-invariant linear operator in a sliding window. Both procedures tend to be excellent for prediction purposes over small time horizons, but are generally time-consuming and, crucially, do not provide a global prior model for the temporally-varying dynamics that is realistic. Here, we tackle these two issues by using a deep convolution neural network (CNN) in a hierarchical statistical IDE framework, where the CNN is designed to extract process dynamics from the process' most recent behaviour. Once the CNN is fitted, probabilistic forecasting can be done extremely quickly online using an ensemble Kalman filter with no requirement for repeated parameter estimation. We conduct an experiment where we train the model using 13 years of daily sea-surface temperature data in the North Atlantic Ocean. Forecasts are seen to be accurate and calibrated. A key advantage of our approach is that the CNN provides a global prior model for the dynamics that is realistic, interpretable, and computationally efficient. We show the versatility of the approach by successfully producing 10-minute nowcasts of weather radar reflectivities in Sydney using the same model that was trained on daily sea-surface temperature data in the North Atlantic Ocean. : 22 pages, 10 figures Article in Journal/Newspaper North Atlantic DataCite Metadata Store (German National Library of Science and Technology) |
institution |
Open Polar |
collection |
DataCite Metadata Store (German National Library of Science and Technology) |
op_collection_id |
ftdatacite |
language |
unknown |
topic |
Machine Learning stat.ML Machine Learning cs.LG Methodology stat.ME FOS Computer and information sciences |
spellingShingle |
Machine Learning stat.ML Machine Learning cs.LG Methodology stat.ME FOS Computer and information sciences Zammit-Mangion, Andrew Wikle, Christopher K. Deep Integro-Difference Equation Models for Spatio-Temporal Forecasting |
topic_facet |
Machine Learning stat.ML Machine Learning cs.LG Methodology stat.ME FOS Computer and information sciences |
description |
Integro-difference equation (IDE) models describe the conditional dependence between the spatial process at a future time point and the process at the present time point through an integral operator. Nonlinearity or temporal dependence in the dynamics is often captured by allowing the operator parameters to vary temporally, or by re-fitting a model with a temporally-invariant linear operator in a sliding window. Both procedures tend to be excellent for prediction purposes over small time horizons, but are generally time-consuming and, crucially, do not provide a global prior model for the temporally-varying dynamics that is realistic. Here, we tackle these two issues by using a deep convolution neural network (CNN) in a hierarchical statistical IDE framework, where the CNN is designed to extract process dynamics from the process' most recent behaviour. Once the CNN is fitted, probabilistic forecasting can be done extremely quickly online using an ensemble Kalman filter with no requirement for repeated parameter estimation. We conduct an experiment where we train the model using 13 years of daily sea-surface temperature data in the North Atlantic Ocean. Forecasts are seen to be accurate and calibrated. A key advantage of our approach is that the CNN provides a global prior model for the dynamics that is realistic, interpretable, and computationally efficient. We show the versatility of the approach by successfully producing 10-minute nowcasts of weather radar reflectivities in Sydney using the same model that was trained on daily sea-surface temperature data in the North Atlantic Ocean. : 22 pages, 10 figures |
format |
Article in Journal/Newspaper |
author |
Zammit-Mangion, Andrew Wikle, Christopher K. |
author_facet |
Zammit-Mangion, Andrew Wikle, Christopher K. |
author_sort |
Zammit-Mangion, Andrew |
title |
Deep Integro-Difference Equation Models for Spatio-Temporal Forecasting |
title_short |
Deep Integro-Difference Equation Models for Spatio-Temporal Forecasting |
title_full |
Deep Integro-Difference Equation Models for Spatio-Temporal Forecasting |
title_fullStr |
Deep Integro-Difference Equation Models for Spatio-Temporal Forecasting |
title_full_unstemmed |
Deep Integro-Difference Equation Models for Spatio-Temporal Forecasting |
title_sort |
deep integro-difference equation models for spatio-temporal forecasting |
publisher |
arXiv |
publishDate |
2019 |
url |
https://dx.doi.org/10.48550/arxiv.1910.13524 https://arxiv.org/abs/1910.13524 |
genre |
North Atlantic |
genre_facet |
North Atlantic |
op_relation |
https://dx.doi.org/10.1016/j.spasta.2020.100408 |
op_rights |
arXiv.org perpetual, non-exclusive license http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
op_doi |
https://doi.org/10.48550/arxiv.1910.13524 https://doi.org/10.1016/j.spasta.2020.100408 |
_version_ |
1766128915591462912 |