Some properties of multivariate measures of concordance

M. D. Taylor
DOI: https://doi.org/10.48550/arXiv.0808.3105
2008-08-23
Abstract:We explore the consequences of a set of axioms which extend Scarsini's axioms for bivariate measures of concordance to the multivariate case and exhibit the following results: (1) A method of extending measures of concordance from the bivariate case to arbitrarily high dimensions. (2) A formula expressing the measure of concordance of the random vectors $(\pm X_1,...,\pm X_n)$ in terms of the measures of concordance of the "marginal" random vectors $(X_{i_1},...,X_{i_k})$. (3) A method of expressing the measure of concordance of an odd-dimensional copula in terms of the measures of concordance of its even-dimensional marginals. (4) A family of relations which exist between the measures of concordance of the marginals of a given copula.
Probability,Statistics Theory
What problem does this paper attempt to address?