Robust portfolio selection using interval random programming

This paper addresses the portfolio selection problem in a robust manner. In practice, it is difficult to collect all information to determine the precise bounds of the box uncertainty set used in robust portfolio optimization. To solve this problem, we introduce a novel uncertainty set: interval random uncertainty. We apply our interval random chance-constrained programming to robust semi-absolute deviation portfolio selection under interval random uncertainty in the element of mean vector. The method for generating the uncertainty set from historical data is discussed. An hybrid-intelligent algorithm is applied to solve the robust portfolio model. Finally, we compare the potentially significant economic benefits of investing in portfolios computed using classical model and the model introduced here. And the robustness is achieved at relatively high performance and low cost.