Uncertain random portfolio optimization via semi-variance

Semi-variance is a similar measure to variance, but it only considers values that are below the expected value. As important roles of semi-variance in finance, this paper proposes the concept of semi-variance for uncertain random variables. Also, a computational approach for semi-variance is provided via inverse uncertainty distribution. As an application in finance, portfolio selection problems of uncertain random returns are solved by minimizing semi-variance in mean-semi variance models. For better illustration, mean-semi variance model is compared with mean-variance one. Finally, for better understanding, some tables, figures and outputs are provided.