A new portfolio selection model with interval-typed random variables and the empirical analysis.

This paper proposes a new portfolio selection model, where the goal is to maximize the expected portfolio return and meanwhile minimize the risks of all the assets. The average return of every asset is considered as an interval number, and the risk of every asset is treated by probabilistic measure. An algorithm for solving the portfolio selection problem is given. Then a Pareto-maximal solution could be obtained under order relations between interval numbers. Finally, the empirical analysis is presented to show the feasibility and robustness of the model. [ABSTRACT FROM AUTHOR]