Inference for an extreme value model accounting for inter-site dependence

The spatial extreme value data observed at many sites is usually modelled by a multivariate extreme value distribution to take into account the inter-site dependence. However, the analysis involving many sites could create computational and mathematical issues because of the high dimensionality. Thi...

Full description

Saved in:
Bibliographic Details
Main Authors: Gabd, Darmesah, Tawn, Jonathan
Format: Conference Proceeding
Language:English
Subjects:
Computer simulation
Dependence
Error analysis
Extreme values
Margined securities
Normal distribution
Parameter estimation
Spatial data
Standard error
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The spatial extreme value data observed at many sites is usually modelled by a multivariate extreme value distribution to take into account the inter-site dependence. However, the analysis involving many sites could create computational and mathematical issues because of the high dimensionality. This study will illustrate an alternative method to come up with estimates for dependent extreme data. The main interest will be in the trend parameter estimate and its standard error. We will show a simulation study to illustrate the advantages of the alternative method to the multivariate extreme value distribution.
ISSN:0094-243X
1551-7616
DOI:10.1063/1.4980985