Modeling Spatial Extremes via Ensemble-of-Trees of Pairwise Copulas
Assessing the risk of extreme events in a spatial domain, such as hurricanes, floods, and droughts, presents a unique significance in practice. Unfortunately, the existing extreme-value statistical models are typically not feasible for practical large-scale problems. Graphical models, on the other h...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/84387 http://hdl.handle.net/10220/43582 |
| _version_ | 1811695674997080064 |
|---|---|
| author | Yu, Hang Uy, Wayne Isaac T. Dauwels, Justin |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Yu, Hang Uy, Wayne Isaac T. Dauwels, Justin |
| author_sort | Yu, Hang |
| collection | NTU |
| description | Assessing the risk of extreme events in a spatial domain, such as hurricanes, floods, and droughts, presents a unique significance in practice. Unfortunately, the existing extreme-value statistical models are typically not feasible for practical large-scale problems. Graphical models, on the other hand, are capable of handling sizable number of variables, but have yet to be explored in the realm of extreme-value analysis. To bridge the gap, an extreme-value graphical model is introduced in this paper, i.e., an ensemble-of-trees of pairwise copulas (ETPC). In the proposed graphical model, extreme-value marginal distributions are stitched together by means of a pairwise copulas, which in turn are the building blocks of the ensemble of trees. Novel linear-complexity stochastic gradient-based algorithms are then developed for learning the ETPC model and inferring missing data. As a result, the ETPC model is applicable to extreme-value problems with thousands of variables. It can be proven that, under mild conditions, the ETPC model exhibits the favorable property of tail-dependence between an arbitrary pair of sites (variables); consequently, the model is able to reliably capture statistical dependence between extreme values at different sites. Experimental results for both synthetic and real data demonstrate the advantages of the ETPC model in modeling fitting, imputation, and computational efficiency. |
| first_indexed | 2024-10-01T07:27:14Z |
| format | Journal Article |
| id | ntu-10356/84387 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T07:27:14Z |
| publishDate | 2017 |
| record_format | dspace |
| spelling | ntu-10356/843872020-03-07T13:57:28Z Modeling Spatial Extremes via Ensemble-of-Trees of Pairwise Copulas Yu, Hang Uy, Wayne Isaac T. Dauwels, Justin School of Electrical and Electronic Engineering Graphical Models Computational Modeling Assessing the risk of extreme events in a spatial domain, such as hurricanes, floods, and droughts, presents a unique significance in practice. Unfortunately, the existing extreme-value statistical models are typically not feasible for practical large-scale problems. Graphical models, on the other hand, are capable of handling sizable number of variables, but have yet to be explored in the realm of extreme-value analysis. To bridge the gap, an extreme-value graphical model is introduced in this paper, i.e., an ensemble-of-trees of pairwise copulas (ETPC). In the proposed graphical model, extreme-value marginal distributions are stitched together by means of a pairwise copulas, which in turn are the building blocks of the ensemble of trees. Novel linear-complexity stochastic gradient-based algorithms are then developed for learning the ETPC model and inferring missing data. As a result, the ETPC model is applicable to extreme-value problems with thousands of variables. It can be proven that, under mild conditions, the ETPC model exhibits the favorable property of tail-dependence between an arbitrary pair of sites (variables); consequently, the model is able to reliably capture statistical dependence between extreme values at different sites. Experimental results for both synthetic and real data demonstrate the advantages of the ETPC model in modeling fitting, imputation, and computational efficiency. MOE (Min. of Education, S’pore) Accepted version 2017-08-14T08:28:42Z 2019-12-06T15:44:07Z 2017-08-14T08:28:42Z 2019-12-06T15:44:07Z 2017 Journal Article Yu, H., Uy, W. I. T., & Dauwels, J. (2017). Modeling Spatial Extremes via Ensemble-of-Trees of Pairwise Copulas. IEEE Transactions on Signal Processing, 65(3), 571-586. 1053-587X https://hdl.handle.net/10356/84387 http://hdl.handle.net/10220/43582 10.1109/TSP.2016.2614485 en IEEE Transactions on Signal Processing © 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The published version is available at: [http://dx.doi.org/10.1109/TSP.2016.2614485]. 16 p. application/pdf |
| spellingShingle | Graphical Models Computational Modeling Yu, Hang Uy, Wayne Isaac T. Dauwels, Justin Modeling Spatial Extremes via Ensemble-of-Trees of Pairwise Copulas |
| title | Modeling Spatial Extremes via Ensemble-of-Trees of Pairwise Copulas |
| title_full | Modeling Spatial Extremes via Ensemble-of-Trees of Pairwise Copulas |
| title_fullStr | Modeling Spatial Extremes via Ensemble-of-Trees of Pairwise Copulas |
| title_full_unstemmed | Modeling Spatial Extremes via Ensemble-of-Trees of Pairwise Copulas |
| title_short | Modeling Spatial Extremes via Ensemble-of-Trees of Pairwise Copulas |
| title_sort | modeling spatial extremes via ensemble of trees of pairwise copulas |
| topic | Graphical Models Computational Modeling |
| url | https://hdl.handle.net/10356/84387 http://hdl.handle.net/10220/43582 |
| work_keys_str_mv | AT yuhang modelingspatialextremesviaensembleoftreesofpairwisecopulas AT uywayneisaact modelingspatialextremesviaensembleoftreesofpairwisecopulas AT dauwelsjustin modelingspatialextremesviaensembleoftreesofpairwisecopulas |