Open Journal of Mathematical Analysis

A study of the power-cosine copula

Christophe Chesneau
Université de Caen Normandie, LMNO, Campus II, Science 3, 14032, Caen, France; christophe.chesneau@gmail.com

Abstract

Copulas played a key role in numerous areas of statistics over the last few decades. In this paper, we offer a new kind of trigonometric bivariate copula based on power and cosine functions. We present it via analytical and graphical approaches. We show that it may be used to create a new bivariate normal distribution with interesting shapes. Subsequently, the simplest version of the suggested copula is highlighted. We discuss some of its relationships with the Farlie-Gumbel-Morgensten and simple polynomial-sine copulas, establish that it is a member of a well-known semi-parametric family of copulas, investigate its dependence domains, and show that it has no tail dependence.

Keywords:

Copulas; Farlie-Gumbel-Morgenstern copula; Polynomial-sine copula; Normal distribution; Tail dependence.