Stock Price Forecasting Based On Rough Set Theory

Data mining aims to get previously unknown and potentially useful knowledge from a large amount of data to offer decision support. Rough set theory is a mathematical tool for use in circumstances that are characterized by vagueness and uncertainty. It has been proven to be very useful in the field of data mining. With the development of the stock market, lots of history exchange data has been stored in database. It attracts more and more attention that how to use these history exchange data to discover the rules of the stock market. Based on binary discernible matrix, it is studied of the binary discernible matrix reducible algorithm, Attribute reduction algorithm and values extraction algorithm. After considered the information of row and column provided by the binary discernible matrix, a more efficient attribute reduction algorithm is designed. Then, it is pointed out the defect of the current method that used to construct stock decision table. A new method is designed after considered the four apply aspects of the technical indicators, i.e. extreme value, deviation, cross, and invalidation to construct a more reasonable decision table. The percent rise or fall of three days after is placed in decision attribute in order to forecast the stock price. A new discrete method based on cloud transform has been researched, the characteristics of technical indicator are also considered in practice. A software is developed to realize the present knowledge discovery system which include four modules: user interface, database, pretreatment of data, and rule extraction. A practice indicates that the present method has a fine capacity to forecast the trend of the stock price. At last, it is discussed of the valuable research fields in the future research.