Powerful Linear Tests for DARA Stochastic Dominance
We develop and implement linear formulations of stochastic dominance decision rules based on Decreasing Absolute Risk Aversion (DARA) for discrete probability distributions.Our approach to DARA SD (DSD) is based on a piecewise-linear representation of log marginal utility and a local linear approximation to its exponentiation.This approach can compare a given prospect with an alternative prospect, a discrete set of multiple alternatives,or a polyhedral set of linear combinations of prospects.Our tests can be implemented by solving a relatively small system of linear inequalities.An empirical application to historical stock market data suggests that the passive stock market index is DSD inefficient relative to actively managed, concentrated portfolios of small-cap stocks.The mean-variance rule and N-th order stochastic dominance rules substantially underestimate the degree of market portfolio inefficiency, because they do not penalize the unfavorable skewness of diversified portfolios, in violation of DARA.
Stochastic dominance utility theory decreasing absolute risk aversion linear programming bootstrapping mean-variance analysis market portfolio efficiency pricing kemel skewness
Yi Fang Thierry Post
Economics at the Center for Quantitative Economics and the Business School of Jilin University;2699 Qianjin Street, 130012, Changchun City, China
国内会议
长春
英文
303-323
2013-07-20（万方平台首次上网日期，不代表论文的发表时间）