abstract 92

Bulletin of Computational Applied Mathematics (Bull CompAMa) 

92

Solutions to two open problems on bivariate variable-power copulas (Research Paper)

Christophe Chesneau


This article presents two innovative bivariate copulas: one is a single-parameter variable-power copula, while the other is a single-parameter variable-power exponential copula. Both have the merit of answering certain open questions and unifying the independence and referenced copulas of the literature. Taking advantage of the flexibility of the variable-power transformation, they provide a versatile framework for modeling complex bivariate dependence structures. We describe their construction methodology and study their mathematical properties, including expressions for copula densities and survival copulas, asymmetric features, order relations, various dependence measures, and pairs of value generation. A focus is put on their negative dependence. Our results thus enrich the literature by offering original statistical dependence modeling based on the modern concept of variable-power copula.

Keywords: Variable-power copulas; dependence model; correlation measures.


Cite this paper:

Chesneau C.

Solutions to two open problems on bivariate variable-power copulas

Bull. Comput. Appl. Math. (Bull CompAMa)

Vol. 12, No.1 pp.93-118, 2024.