Recognizing the Patten of Beta Based on Rough Sets and Support Vector Machine

Beta is calculated by linear analysis between the closing prices of stocks and the security index of stock market. However, many studies have showed there are strong relationships between beta and financial information. Since the traditional statistical techniques have many limitations in disposing deficient and high noisy data, the past studies rested on proving the relationships between financial information and systematic risk. In this study, the hybrid system of rough sets and support vector machine (SVM) was employed to dispose the problem of pattern recognizing, in which rough sets were used for accelerating or simplifying the process of training SVM by eliminating the redundant data from database. Therefore, this paper used the hybrid system to recognize the clusters of beta with financial information. At last the effectiveness of our approach was verified by testing the hybrid system with the companies which listed on Shenzhen stock market.