This study proposes stock portfolio optimization models for risk-averse investors under uncertain conditions. To accomplish this, the study measures risk based on the conditional drawdown-at-risk (CDaR) measure, which can prevent major declines in investment as a conservative investment strategy. Given the uncertain character of the input parameters of the problem, the study first proposes a hybrid possibilistic and flexible model by considering the CDaR measure (CDaR-HPFM) to handle uncertainty. Furthermore, to offer more robust outcomes, the study constructs a hybrid possibilistic and flexible robust model by considering the CDaR measure (CDaR-HPFRM), in the forms of a non-linear and a linear mathematicalprogramming problem. The CDaR-HPFRM can process the robustness of output decisions while dealing with uncertain parameters. The real stock exchange data of 100 companies registered on the Tehran Stock Exchange are investigated to validate the functionality of the proposed models. The results reveal that, when various penalty costs are factored in, the CDaR-HPFRM yields more efficient and more robust results than the CDaR-HPFM. Comparative results also indicate that the proposed models almost always outperform existing methods, in terms of both CDaR and rate of return measures. Proposed models can be used for all types of stock market investments (including micro-investing and investment funds) and can handle portfolio optimization and selection processes in project-based organizations.
Read full abstract