Two Essays on Almost Stochastic Dominance

• Main Authors: Wei-Ting Wang, 王薇婷
• Other Authors: Larry Y. Tzeng
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/42555884916042707183

博士 === 國立臺灣大學 === 財務金融學研究所 === 102 === The dissertation contains two articles to discuss how Almost Stochastic Dominance (ASD) criteria improve on risk-averse investors and non-satiation investors making decisions. Marginal Conditional Stochastic Dominance (MCSD) developed by Shalit and Yitzhaki (1994) gives the conditions under which all risk-averse individuals prefer to increase the share of one risky asset over another in a given portfolio. In this paper, we extend this concept to provide conditions under which most (and not all) risk-averse investors behave in this way. Switching from MCSD to Almost MCSD (AMCSD) helps to reconcile common practices in asset allocation and the decision rules supporting stochastic dominance relations. A financial application is further provided to demonstrate that using AMCSD can indeed improve investment efficiency. Almost first-degree stochastic dominance (AFSD) rule developed by Leshno and Levy (2002) asserts that most decision makers may prefer one uncertain prospect over another even with some violations of first-degree stochastic dominance rules. In this paper, we propose an efficiency test based on AFSD. Following Kuosmanen (2004) and Kopa and Post (2009), we respectively propose tests for portfolio admissibility and portfolio optimality under AFSD. We then show how to use linear programming to implement tests under AFSD rule and demonstrate their applications in stock markets.