The pair-copula construction for spatial data: a new approach to model spatial dependency
Gräler Benedikt, Pebesma Edzer
Research article (journal)Abstract
Copulas are a flexible tool to model dependence of random variables. They cover the range from perfect negative topositive dependence, include the independent case and incorporate asymmetric dependence as well as the widely usedGaussian dependence structure. The pair-copula construction for multivariate copulas exploits the ease of bivariatecopulas and suggests a decomposition of a multivariate copula into a set of bivariate ones. We successfully adaptedthis approach for spatial data and developed a powerful spatial pair-copula based interpolation method.
Details about the publication
Journal: Procedia Environmental Sciences
Volume: 7
Issue: 0
Page range: 206-211
Status: Published
Release year: 2011
Language in which the publication is written: English
Link to the full text: http://www.sciencedirect.com/science/article/pii/S1878029611001630
Keywords: vine copulas; pair-copula; interpolation; spatial dependence
Authors from the University of Münster
| Gräler, Benedikt | Professur für Geoinformatik (Prof. Pebesma) |
| Pebesma, Edzer | Professur für Geoinformatik (Prof. Pebesma) |