Local likelihood estimation of complex tail dependence structures in high dimensions, applied to US precipitation extremes

Handle URI:
http://hdl.handle.net/10754/626512
Title:
Local likelihood estimation of complex tail dependence structures in high dimensions, applied to US precipitation extremes
Authors:
Camilo, Daniela Castro; Huser, Raphaël ( 0000-0002-1228-2071 )
Abstract:
In order to model the complex non-stationary dependence structure of precipitation extremes over the entire contiguous U.S., we propose a flexible local approach based on factor copula models. Our sub-asymptotic spatial modeling framework yields non-trivial tail dependence structures, with a weakening dependence strength as events become more extreme, a feature commonly observed with precipitation data but not accounted for in classical asymptotic extreme-value models. To estimate the local extremal behavior, we fit the proposed model in small regional neighborhoods to high threshold exceedances, under the assumption of local stationarity. This allows us to gain in flexibility, while making inference for such a large and complex dataset feasible. Adopting a local censored likelihood approach, inference is made on a fine spatial grid, and local estimation is performed taking advantage of distributed computing resources and of the embarrassingly parallel nature of this estimation procedure. The local model is efficiently fitted at all grid points, and uncertainty is measured using a block bootstrap procedure. An extensive simulation study shows that our approach is able to adequately capture complex, non-stationary dependencies, while our study of U.S. winter precipitation data reveals interesting differences in local tail structures over space, which has important implications on regional risk assessment of extreme precipitation events. A comparison between past and current data suggests that extremes in certain areas might be slightly wider in extent nowadays than during the first half of the twentieth century.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Publisher:
arXiv
Issue Date:
2-Oct-2017
ARXIV:
arXiv:1710.00875
Type:
Preprint
Additional Links:
http://arxiv.org/abs/1710.00875v1; http://arxiv.org/pdf/1710.00875v1
Appears in Collections:
Other/General Submission; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorCamilo, Daniela Castroen
dc.contributor.authorHuser, Raphaëlen
dc.date.accessioned2017-12-28T07:32:14Z-
dc.date.available2017-12-28T07:32:14Z-
dc.date.issued2017-10-02en
dc.identifier.urihttp://hdl.handle.net/10754/626512-
dc.description.abstractIn order to model the complex non-stationary dependence structure of precipitation extremes over the entire contiguous U.S., we propose a flexible local approach based on factor copula models. Our sub-asymptotic spatial modeling framework yields non-trivial tail dependence structures, with a weakening dependence strength as events become more extreme, a feature commonly observed with precipitation data but not accounted for in classical asymptotic extreme-value models. To estimate the local extremal behavior, we fit the proposed model in small regional neighborhoods to high threshold exceedances, under the assumption of local stationarity. This allows us to gain in flexibility, while making inference for such a large and complex dataset feasible. Adopting a local censored likelihood approach, inference is made on a fine spatial grid, and local estimation is performed taking advantage of distributed computing resources and of the embarrassingly parallel nature of this estimation procedure. The local model is efficiently fitted at all grid points, and uncertainty is measured using a block bootstrap procedure. An extensive simulation study shows that our approach is able to adequately capture complex, non-stationary dependencies, while our study of U.S. winter precipitation data reveals interesting differences in local tail structures over space, which has important implications on regional risk assessment of extreme precipitation events. A comparison between past and current data suggests that extremes in certain areas might be slightly wider in extent nowadays than during the first half of the twentieth century.en
dc.publisherarXiven
dc.relation.urlhttp://arxiv.org/abs/1710.00875v1en
dc.relation.urlhttp://arxiv.org/pdf/1710.00875v1en
dc.rightsArchived with thanks to arXiven
dc.titleLocal likelihood estimation of complex tail dependence structures in high dimensions, applied to US precipitation extremesen
dc.typePreprinten
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.eprint.versionPre-printen
dc.identifier.arxividarXiv:1710.00875en
kaust.authorCamilo, Daniela Castroen
kaust.authorHuser, Raphaëlen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.