Fundamental analysis of stock price by artificial neural networks model based on rough set theory ∗

It has been widely accepted that predicting stock price is not a simple task since many market factors are involved and their structural relationships are not fully known. In this study, we use both rough set theory and neural networks approach to get an effective model of stock price movement for China’s young stock market. The model is modified and tested by the most recent 6 years of data collecting from China’s stock market to make sure it is updating and withstanding in a long time. Consequently, a group of most important fundamental indicators are selected by rough set theory and these indicators are successful in detailing and predicting stock price movement in a long-term using neural networks approach. Results indicate that neural networks approach based on rough set theory is efficient in modelling and more accurate in prediction.