Forecasting Stock Prices In Tehran Stock Exchange Using Recurrent Neural Network Optimized by Artificial Bee Colony Algorithm

This Research in a hybrid approach, using Recurrent Neural Networks (RNN) based on Artificial Bee Colony algorithm (ABC), is going to provide the optimal model to predict the stock prices in Tehran Stock Exchange. For this purpose, using the stock datafor companies listed in the first market of the Tehran Stock Exchange, traded between 2011 and the end of 2015,after the definition of different technical and fundamental variables, using step by step regression-correlation, factors affecting stock prices in Tehran Stock Exchange were selected as model input is defined. Then the artificial bee colony algorithm in an atmosphere of parametric design, to optimize the weights and biases of recurrent neural network is used. To evaluate the model, several criteria for a given stock of listed companies in Tehran Stock Exchange are used. The results show that the neural network optimized with artificial bee colony algorithm has considerably accuracy, compared to other forecasting methods.