Mean-variance model

In this paper, we use fuzzy numbers and possibility theory to model possibility. The purpose of this work is to determine the optimal investment model based on the neural network method for fuzzy LR, trapezoidal and triangular numbers in an optimal portfolio. It is listed on the Tehran Stock Exchange to maximize "returns" and reduce "risk" to find the optimal portfolio. Therefore, to achieve this goal, the problem of multi-objective nonlinear programming is addressed. Also, by substituting the mean-variance model and the standard mean deviation instead of the Markowitz mean-variance model, the selection of the optimal portfolio in the possible space is examined. Finally, after calculating the model of the possibility of fuzzy numbers, we reach the optimal stock portfolio, which can be used to set the stock portfolio that has the highest returns and the lowest risk.

This study was conducted to determine the stock portfolio with the best return and low-risk investments using momentum-contrarian strategies (MCSs). The momentum-Contrarian strategy is one of the well-known models to construct the portfolio which suggests buying the stocks with the best performance (the winner stocks) and selling the stocks with the worst performance (the loser stocks). The optimal values of the portfolio's objective function and the weight of all assets in the portfolio that are not necessarily the same are calculated by defining a nonlinear multivariate optimization model combined with momentum-contrarian strategies (MCSs). The return information of companies listed on the Tehran Stock Exchange from 2014 to 2019 was used to select the best optimal portfolio. The results confirmed the stability in the profitability of the contrarian optimal portfolio with minimum risk compared to other optimal portfolios. Furthermore, through MATLAB software the optimal weight of assets in the optimal portfolio is calculated based on statistical data.