Spatial composite likelihood inference using local C-vines

We present a vine copula based composite likelihood approach to model spatial dependencies, which allows to perform prediction at arbitrary locations. It combines established methods to model (spatial) dependencies. On the one hand spatial differences between the variable locations are utilized to m...

Full description

Saved in:
Bibliographic Details
Published in:Journal of multivariate analysis 2015-06, Vol.138, p.74-88
Main Authors: Erhardt, Tobias Michael, Czado, Claudia, Schepsmeier, Ulf
Format: Article
Language:English
Subjects:
Comparative analysis
Daily mean temperature
Local dependency modeling
Mathematical models
Non-Gaussian dependencies
Normal distribution
Parameter estimation
Predictions
Spatial R-vine model
Spatial statistics
Validation studies
Vine copulas
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present a vine copula based composite likelihood approach to model spatial dependencies, which allows to perform prediction at arbitrary locations. It combines established methods to model (spatial) dependencies. On the one hand spatial differences between the variable locations are utilized to model the degree of spatial dependence. On the other hand the flexible class of C-vine copulas are used to model the spatial dependency structure locally. These local C-vine copulas are parametrized jointly, exploiting a relationship between the copula parameters and the corresponding spatial distances and elevation differences, and are combined in a composite likelihood approach. This spatial local C-vine composite likelihood (S-LCVCL) method benefits from the fact that it is able to capture non-Gaussian dependency structures. The development and validation of the new methodology is illustrated using a data set of daily mean temperatures observed at 73 observation stations spread over Germany. For validation continuous ranked probability scores are utilized. Comparison with another vine copula based approach and a Gaussian approach for spatial dependency modeling shows a preference for vine copula based (spatial) dependency structures.
ISSN:0047-259X
1095-7243
DOI:10.1016/j.jmva.2015.01.021