This paper provides an axiomatic characterization of two rules for comparing alternative sets of objects on the basis of the diversity that they offer. The framework considered assumes a finite universe of objects and an a priori given ordinal quadernary relation that compares alternative pairs...

This paper provides an axiomatic characterization of two rules for comparing alternative sets of objects on the basis of the diversity that they offer. The framework considered assumes a finite universe of objects and an a priori given ordinal quadernary relation that compares alternative pairs...

We study the problem of estimating the parameters of a linear median regression without any assumption on the shape of the error distribution -- including no condition on the existence of moments -- allowing for heterogeneity (or heteroskedasticity) of unknown form, noncontinuous distributions,...