Research And Application Of Data Mining Technology In Analyzing Securities

The securities market is the essential constituent of the market economy, it helps the company collecting capital. The stability of stock market is affecting the development of China economy, its function in market economy can not be neglect. But, the securities market is a complex, dynamic economical system, it is affected by many factors, which made the securities investment have no rule to follow.Along with the China economic development, the securities investment has play an important role in people’s life. However the stock market is unpredictable, only few investors can obtain the abnormal return, therefore how to forecast the trend of stock, how to obtain the long-term gains, is the concern to each investor.Data mining takes the massive data as the analysis object, which searches the useful information from a massive data. It has been widely used in bank, telecommunication, retail sale, and insurance and so on. The application of stock market mainly concentrates on the customer relationship management, the market forecast and risk management and so on. Which use a variety method, such as the cluster analysis, the association analysis, the decision tree, the neural network, and the time series analysis.The article uses the data mining technology to analyze the stock market. we use the decision tree to carry on the company fundamental analysis, we selects all companies of security market in 2008 as the objects, use net assets per share, cash flow per share, return on equity, revenue growth, net profit growth rate and earnings per share to carry on analysis, uses earnings per share to classify the target company, and using the decision tree to forecast company earnings.In the article we use constituent stocks of the Shanghai and Shenzhen 300 index, through the association rule analysis, excavates the strong relevance between the stocks. Companies belong to the same region, manage same or close product often have same tendency. More importantly, the association analysis can discover strong relevance between stocks witch belong to the different profession, the different region. Through data mining reveals the rules like "in the same day, the stock A rises, stock B also rises", "in the same day, stock A rises, stock B rises, then stock C also rises".We use the association analysis forecast the stock. We take the constituent stocks of the Shanghai and Shenzhen 300 index as object of research, through the association analysis, we find the rules like "Stock A and stock B rise, stock C rises in the next day", such rules can be used to forecast the stock.