Multi-period fuzzy portfolio selection model with preference-regret criterion

In this paper, we explore a preference-regret decision problem for multi-period portfolio selection in a fuzzy environment. To do this, we first define a possibilistic preference-regret measure to quantify an investor’s regret for not holding the ex-post best-performing strategy. Then, we present a possibilistic preference-regret model with real investment constraints for multi-period fuzzy portfolio selection. In our proposed model, we assume that the investor aims to identify an optimal investment strategy that maximizes the cumulative possibilistic preference-regret concerning portfolio returns over the entire investment horizon. To enhance the practicality of our model, we incorporate several real-world investment constraints, including transaction costs, cardinality constraint, dependency constraint, and budget constraint. Furthermore, we develop an improved co-evolutionary particle swarm optimization algorithm to solve our proposed model. Finally, we conduct a comparative analysis using a numerical example to demonstrate the practicality of our model and validate the effectiveness of the solution algorithm.