Multivariate stochastic dominance: A parametric approach

This paper proposes parameterized multivariate stochastic dominance (PMSD) rules under different distributional assumptions for a class of non-satiable risk-seeking investors. In particular, it determines the PMSD rules for both stable symmetric and Student's t distributions. Methodologically, the PMSD rules for ordering are based on comparison of i) location parameters, ii) dispersion parameters, and iii) either stability indices or degrees of freedom. In addition, it presents the main steps for evaluating such rules. This paper confirms that return tail behavior plays a crucial role in determining non-satiable investors' optimal choices.