Multivariate discrete copulas, with applications in probabilistic weather forecasting
Methodology
2015-12-18 v1
Abstract
In probability and statistics, copulas play important roles theoretically as well as to address a wide range of problems in various application areas. We introduce the concept of multivariate discrete copulas, discuss their equivalence to stochastic arrays, and prove a multivariate discrete version of Sklar's theorem. These results provide the theoretical frame for multivariate statistical methods to postprocess weather forecasts made by ensemble systems, including the ensemble copula coupling approach and the Schaake shuffle.
Cite
@article{arxiv.1512.05629,
title = {Multivariate discrete copulas, with applications in probabilistic weather forecasting},
author = {Roman Schefzik},
journal= {arXiv preprint arXiv:1512.05629},
year = {2015}
}
Comments
arXiv admin note: text overlap with arXiv:1305.3445