A hybrid approach of linear goal programming-based best-worst and multi-attribute sorting methods to form a stock portfolio

In the last four decades, multi-attribute decision-making has always been an active field of research and application. So far, the focus has been developing or applying multi-attribute decision-making techniques to rank existing options. However, multi-attribute decision-making is not limited to this type of problem and also includes sorting problems. This study proposes a hybrid approach of linear goal programming-based best-worst and multi-attribute sorting methods to form a stock portfolio in the Tehran Stock Exchange. In this study, the weights of the attributes affecting the stock portfolio selection were determined using the linear goal programming-based best-worst method. Then, the stock portfolio was formed using the proposed multi-attribute sorting method. The proposed method can consider decision-maker preferences such as the range of stocks in the portfolio or the maximum number of stocks in each industry. The results obtained from the proposed method were compared with TOPSIS-Sort and VIKOR-Sort methods. The results showed the high accuracy of the proposed method, and the stock portfolio formed by the proposed method was significantly more profitable than other methods. The proposed approach in this research can be applied to different real-world sorting problems.