Reference-Based Almost Stochastic Dominance Rules with Application in Risk-Averse Optimization

TL;DR

This paper introduces reference-based almost stochastic dominance (RSD) rules that balance general risk aversion with individual preferences, providing a robust framework for decision-making under uncertainty, especially in portfolio optimization.

Contribution

The paper develops RSD rules that incorporate individual utility functions into stochastic dominance, along with an optimization model and approximation algorithm for practical applications.

Findings

RSD rules effectively capture individual risk preferences.

The proposed algorithm efficiently solves portfolio optimization problems.

Application demonstrates improved decision robustness.

Abstract

Stochastic dominance is a preference relation of uncertain prospect defined over a class of utility functions. While this utility class represents basic properties of risk aversion, it includes some extreme utility functions rarely characterizing a rational decision maker's preference. In this paper we introduce reference-based almost stochastic dominance (RSD) rules which well balance the general representation of risk aversion and the individualization of the decision maker's risk preference. The key idea is that, in the general utility class, we construct a neighborhood of the decision maker's individual utility function, and represent a preference relation over this neighborhood. The RSD rules reveal the maximum dominance level quantifying the decision maker's robust preference between alternative choices. We also propose RSD constrained stochastic optimization model and develop an approximation algorithm based on Bernstein polynomials. This model is illustrated on a portfolio optimization problem.