A risk averse robust portfolio optimization under severe uncertainties by using IGDT approach: Iran Stock Exchange

Portfolio optimization in finance and economy is more than a mathematical model for improving performance under uncertainty constraints. Practically all organizations seek to create value by selecting the best portfolios that consume the least resources and obtaining high expected portfolio returns and controlling risk. In the context of the portfolio selection problem, severe uncertainties would significantly affect the technical and financial aspects. This paper presents a bi-level information gap decision theory (IGDT) risk averse decision-making tool for robust portfolio optimization problems to help organizations or investors for managing their portfolios and finding the best transactions with severe uncertainty variables (price and return) to process the forecast data generated by the learning prediction method in order to construct the optimal stock portfolios that a target profit is guaranteed. The heuristic solution approach is constructed and the augmented ε-constraint method is used to solve the proposed bi-level IGDT robust optimization problem. The effectiveness and efficiency of the proposed model are evaluated on the Iranian Stock Market. The results show the efficiency and effectiveness of the proposed model for selecting the best stocks. The Mont Carlo simulation method is applied for the validation of results.