Article ID Journal Published Year Pages File Type
1152429 Statistics & Probability Letters 2011 12 Pages PDF
Abstract

In this work, we compare conditional distributions derived from bivariate archimedean copulas in terms of their respective variabilities using the dispersive stochastic order. Specifically, we fix the underlying copula and we consider the effect of increasing the second component on the variability of the conditional distribution of the first component. Characterizations are provided in terms of the generator and of the marginal distributions. Several examples involving standard parametric copulas such as Clayton and Frank ones are discussed.

Related Topics
Physical Sciences and Engineering Mathematics Statistics and Probability
Authors
, ,