摘要：The mean-variance (MV) model is commonly used in portfolio optimization for comparing uncertain prospects. This model however relies strictly on the assumptions that the returns of assets follow normal distribution or the investor's utility function is quadratic. In reality these two conditions do not hold. The mean-Gini (MG) model has been proposed to overcome the limitations of the mean-variance model. In this paper, portfolio compositions and performances employing the MV and MG models utilizing data from the Malaysian share market are compared. Results of this study show that the MG portfolio outperforms the MV portfolio. MG model is not restricted to normal distributions and quadratic utility function enabling investors to construct second degree stochastic dominance (SSD) efficient portfolios thus MG model is a better alternative model for risk-averse investors
