Inequality Measures for Bivariate Distribution with Load Sharing Dependence

Abstract

Abtsrcat

In the univariate case Lorenz order and related summary measures of inequality such as the Gini Index seem to be quite generally accepted as appropriate indicators of inequality. Several extensions of the Lorenz order and Gini index to the multivariate setup has been proposed by various authors. The choice of an apt measure depends on the dependence enjoyed by the original bivariate distribution. When the distribution has a load sharing dependence it is more adequate to study the univariate conditional distributions than the joint distributions. In this paper we study the inequality measures specifically, Lorenz function and Gini index of these distributions and characterize the original bivariate distribution using these measures.