A Regular Bayesian Meural Network to Predict the Stock Market

In this paper, artificial neural networks, Bayesian set as a new method for forecasting the financial markets has been proposed. Financial daily market price and financial technical indicators as inputs to predict an individual's stock price closed the next day is used. Prediction of the movement of stock prices in general as an important and challenging task for the analysis of financial time series is considered. Accurate prediction of stock price movements can play an important role in helping to investors for improving of return on equity. The complexity lie in predicting the trend disorder and instability in the daily movement of the stock price. Bayesian networks regularly weighing the possible nature of a dedicated network, allow network optimization and automation of complex models that are too fine. Experimenting with stock companies, Iran Khodro and Saipa was conducted to determine the effectiveness of the model. The reason for this selection, is attractiveness of Auto industry for activists of capital market, because its return is higher than industry and market indexes.