Answers are grounded in this paper and cite their sources.
A. The copula approach was developed for describing the connection between two or more quantitative variables.Despite the numerous existing copulas, there is still a need for new copulas or generalized versions of well-established copulas.In this study, we suggest two brand-new copulas with three parameters derived from the construction of the Farlie-Gumbel-Morgenstern copula and unique non-separable logarithmic and exponential perturbation functions.In particular, the exponential-based copula has the merit of naturally extending the well-known Celebioglu-Cuadras copula.Widely acceptable domains are established for the involved parameters.The related features and capabilities of the proposed copulas are examined in detail.It is demonstrated that they possess various kinds of shapes, are diagonally symmetric, have manageable series expansions, satisfy interesting first-order copula orders, have a versatile quadrant dependence, have no tail dependence, are not Archimedean, are typically not radially symmetric, and can model a weak or moderate dependence with the rho of Spearman as a benchmark.Numerical tables and figures are used to illustrate some findings.As a final remark, some new two-dimensional inequalities are established, and they might be interesting for reasons unrelated to those of this study.
No takes yet. Share an insight, caveat, or question.
Christophe Chesneau (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: