Outliers Mining And Impact Analysis Based On The High Frequency Data In Stock Market

The study of the relationship between quantity and price of the stock market has always been the basis of research on market behavior, for the relationship between quantity and price theory, both at domestic and abroad after years of research, but, in the abnormal of a relationship have to determine the quantity and price is on the study of the relationship between abnormal cases almost none.Along with the rapid development of China’s financial markets since 2005, the investment amount is gradually enlarged.Especially in recent years, the stock market has turned for the better, more money into the market, led to the stock market movements more blind and extreme.In this new situation, regulating the behavior of investors, the stock market monitoring and management is particularly important. In this case, the relationship between quantity and price anomalies of digging and makes more sense.In this paper,based on the false under the premise of the determination of quantitative price there is A certain relationship, with China’s Shanghai pudong development bank (600000) in the stock market as the research object, the amount of stock price movements on the relationship between quantity and price point of outliers mining and research.On outliers mining, respectively using the method based on distance, density and time series model, to ensure the accuracy and completeness of outliers mining.In the influence of abnormal value analysis, respectively on the impact of outliers on temporal order number and the serial correlation after the appearance of the abnormal value analysis, to find the general rules that affect the outliers. Finally gives the relationship between quantity and price and type of outliers, and the value of practical application is given, and the shortage of the article.