A Novel Bi-objective Credibilistic Mean–Semivariance Portfolio Selection Problem with Coherent Triangular Fuzzy Numbers

This study uses an extension of traditional triangular fuzzy numbers, called coherent triangular fuzzy numbers, to model the portfolio selection problem. The work proposed in this paper begins by introducing an analytical expression of credibilistic semivariance of coherent triangular fuzzy numbers. This expression is then used to formulate a novel bi-objective Mean–Semivariance (M-Sv) portfolio selection model incorporating several practical constraints, such as capital budget, bounds on assets allocations, cardinality, and no-short selling constraints. The proposed bi-objective M-Sv Model is solved using an adaptation of the Multi-Objective Evolutionary Algorithm (MOEA) with a built-in repair mechanism. An empirical analysis is carried out using monthly stock return data of 18 distinct stocks selected from the Nifty 50 Index of the National Stock Exchange (NSE), Mumbai, India, to show the performance stability of the model. The obtained results are compared with the considered benchmark of the Nifty 50 Index.