Presenting a model for stock portfolio optimization based on a combination of GARCH-copula models in Tehran Stock Exchange

Therefore, in the present study, a model for stock portfolio optimization based on a combination of GARCH-copula models in the Tehran Stock Exchange was presented. The present study is in the group of descriptive-correlational researches in terms of practical purpose and data collection method. Also, the statistical sample of the study includes 50 more active companies in the fourth quarter of 1398. For this purpose, the monthly stock return information of these companies was studied over a period of 10 years between 2011 to 2020, and therefore the number 120 rows of observations for each company are the basis of the analysis. The findings of this study show that that the Garch-Copola EVT method has the necessary efficiency to form a portfolio. In terms of risk criteria, it can be seen that this method has presented the lowest risk among the existing methods, and these results confirm the relationship between risk and return in investment activities. Although in this method, a smaller return is obtained than other methods, but the risk will be lower for the investor. Therefore, it can be accepted that this method has been effective in order to optimize the stock portfolio. Comparing the performance of this method with the uniform weights method, it can be seen that the Sharpe ratio in the portfolio with uniform weights was significantly larger than this ratio in the Garch-Copola portfolio. Therefore, it seems that in terms of Sharpe's criterion, the uniform weights method performed better than the proposed method and this method did not have an acceptable efficiency in improving the performance of the portfolio compared to the uniform weights method. Although based on the Sharpe criterion, this method has shown the worst performance among the portfolio formation methods, but in terms of the risk criterion, it can be seen that the risk of this portfolio is significantly lower compared to other methods. Therefore, it can be accepted that the formation of the portfolio using the Garch-Copola EVT method has been able to reduce the portfolio risk compared to other methods.