An Empirical Study of Generalized Almost Stochastic Dominance in Investment Decisions

This project empirically applies generalized almost stochastic dominance (GASD) rule to examine whether stock portfolios are preferred to bond portfolios by most investors. Unlike previous papers which examined the investment decision under almost first-degree and second-degree stochastic dominance (AFSD and ASSD) proposed by Leshno and Levy (2002), this project examines the investment decision under generalized almost second-degree stochastic dominance (GASSD) rule proposed by Tsetlin et al. (2015). With real data, I will first calculate violation ratios against stochastic dominance (SD) rules across different investment horizons for stock portfolios dominating the bond portfolios via three types of GASSD: (ε1, 0)-GASSD, (0, ε2 )-GASSD, and (ε1, ε2 )-GASSD, respectively. Then based on the values of the parameter ε1 and ε2 estimated by the literature, I will demonstrate whether and when the stock portfolios dominate the bond portfolios. Finally, to compare my results under GASSD rules with the results of previous papers, I will also calculate the violation ratios under AFSD and ASSD rules.