On The Analytic Methods Of Stock Market Information

Stock market is a non-linear complicated system, the processes of its price evolution are decided by lots of economic individuals and economic factors that affect each other. The system is influenced by external information. Agents act on each other through exchanging information. Small variety may lead to qualitative changes through systematic affections that system making up of numerous economic individuals organizes, enhances and coordinates each other. Changes of the price are uncertain. The degree of cognizing and understanding on stock market and the ability of judging the future trend of stock market are very important to stock investors. Therefore, people have invented and created lots of theories and methods used for analyzing stock market information, and have expected to gain profit by investing in the stock market for a hundred years. Stock market information is deposited in all kinds of databases of stock exchange system by means of data and sketch. Now the stock analysis systems can help people to make use of the information on a certain degree, but they still lack the ability of withdrawing implicit, unknown information having latent value from a great deal of stock information. In this thesis, present theories and methods concerning basic analysis and technique analysis about stock market information are systematically studied, new methods used for analyzing stock market information are put forward, and merit evaluation model of listed corporation and data mining model of stock information are established. For the merit evaluation problem of listed corporation, the index system of merit evaluation of listed corporation is established and the weight of each index is decided on by using the AHP method. Then the merit evaluation model of listed corporations is established by using the Extension Method. 30 listed corporations are evaluated by the model at last. For the problem of withdrawing implicit, unknown rules and knowledge having latent value from a great deal of stock exchange information, first, the data mining processes of stock market information are designed. Secondly, the data mining model of stock information is established by using the Rough Set method. The model includes a series steps such as data collection, data pretreatment, data concentration and rules generation. Finally, the application of the model in mining stock information is expounded with a real example. In this thesis, new methods and models are provided for analyzing stock market information, and new ideas and ways are provided for further improving and perfecting the analysis system of stock exchange. It can help the section of stock supervision and management to understand quality of listed corporations, discover abnormal behaviors in stock market in time, and exercise effective supervision with regard to listed corporations and stock market. It can help listed corporations to increase comprehensive economic strength and operate achievement constantly. It can help investors to find out the future trend of stock price, and benefit them from making investment decision.