Stock Price Prediction Modeling Using Artificial Neural Network Approach and Imperialist Competitive Algorithm Based On Chaos Theory

Stock market is one of the options available to invest in liquidity. Investors in this area used a variety of approaches to predict stock prices. But due to the nonlinear relationship between variables affecting stock prices, Artificial Neural Networks are one of the most suitable approaches for this work. These networks, through different search optimization algorithms, try to identify the relationships between these variables. The higher the algorithms used, the higher the efficiency of the algorithms, the more accurate the identification of the relationships between the variables. In this paper, an attempt has been made to combine chaotic maps and colonial competition algorithms with the reform movement angle to the colonial colonies so that we can deal with the possibility of being trapped in local optimum to reduce as much as possible. Therefore, using this approach, it is tried to predict the stock price of Iran Khodro Company. To evaluate the performance of the proposed approach to other conventional approaches of neural network education, three perspectives: the degree of accuracy of prediction, the amount of memory used and the time of execution were used. The results show that the proposed approach has a better performance than other approaches.