An Axiomatic Approach to Measuring Degree of Stochastic Dominance

In recent years, a range of measures of "partial" or "degree of" stochastic dominance have been introduced. These measures attempt to determine the extent to which one distribution is dominated by another. In order to systematically assess these proposed measures and their relationship to partial stochastic dominance, we adopt an axiomatic approach. We propose axioms for measuring degree of stochastic dominance and study the relationship between them. Among other findings, we show that one and only one measure satisfies all the axioms.