Pages: 85-97

Abstract

Most real-world optimization has several objectives that generally conflict with each other. Investors in the capital markets pursue several objectives to optimize stock portfolio as well. The present study aimed to optimize multi-objective portfolio using imperialist competitive algorithm in Tehran Stock Exchange. The statistical sample included the top 30 companies listed in Tehran Stock Exchange (TSE) from 2005 to 2015. For this aim, we first used an autoregressive integrated moving average (ARIMA) model to model the return series. Then, portfolio risk was firstly calculated based on generalized autoregressive conditional heteroscedasticity (GARCH) variance models in compliance with the Markowitz approach. Moreover, the results obtained from the present study showed that the imperialist competitive algorithm works well in developing stock portfolios. According to the findings, it is confirmed and recommended to apply the imperialist competitive algorithm for selecting and optimizing stock portfolios. The successful performance of this algorithm is due to the continuous superiority over the certificate market portfolio to the claim of their compatibility with the problem, which is not negligible and deniable.