A Study On Investment Behavior Mode Of Security Market

Traditional theory based on individual investors is considered that investors should be rational to maximize its effectiveness. But recently as the development of behavioral finance, financial academia start to pay attention to experimental research and empirical analysis on the investors the specific investment decision-making process and conduct. They found the emotional, subjective feeling of personality and psychological factors play an essential role in the financial investment. Investors do not always make decisions in a rational manner, not only by their own intrinsic cognitive errors, also environmental interference by outsiders.The period that Chinese scholars study on behavioral finance theory is still short. In the area of financial investment and regulatory practice is rare. Indeed, phenomena that behavioral finance researched have existed in Chinese stock market, such as the over confident, herding, momentum and reversal effects, etc. This paper tries to act finance investor psychology, behavior characteristics and be briefed on the basis of common cognitive errors. In Chinese stock market, investors focused on individual acts of some empirical research on behavioral finance investment strategy, Behavioral Finance will be promoting the popularization and development in China.This paper guided by behavioral finance, studied on Chinese stock trading trend, in a bid to stock investors in the secondary market through the transaction data mining. To find the existing trade patterns. Studying on Chinese stock investors in transactions, This paper has be divided into individual accounts of investors opened accounts with institutional investors and individual investors. SAS process advocated by reference SEMMA (sample, explore, modify, model, and assess). In light of the purpose of analysis——find the mode of the transaction by data mining. The main results were divided into the following five parts:The first part recall the development of behavioral finance in China's stock market,analysis of the unique characteristics and summarize the investor's misbehave. China's stock market in the development of the market, policies on the regulation,listed companies and investors constituted. In such areas as foreign investors constitute a mature market there are substantial differences. Equity financing as a listed company often be considered"free loans", the lack of regulatory policy. Institutional investors can have control of small stock prices, which eventually lead to the poor awareness of investment in stocks of individual investors. Strong sense of speculation. The Chinese stock market investors constituted mainly by individual investors, individual investors are showing speculative behavior, the entire stock market volatility will be inevitable, the mood around the individual, there will be some deviation behavior.The second part will be an introduction to the methods of Data Mining. Different goals use different methods of data mining. All statistical data mining from the theoretical foundation, the mechanism relies on the database management system. Generally speaking, the basis theory for Data Mining sources from statistics. Implementation mechanism depends on the database management system. However, statistical data mining prototype model with a considerable gap, which is facing a massive data because data mining. Considering the efficiency of data analysis and statistical analysis of the various methods in mathematics much perfect, if used for the analysis of such data would be very inefficient. Explore the use of data mining is the main mode of decision tree model.The third part is the Data Clean. Data mining is the secondary data mining, data collection has been identified in analytical purposes, that may not exist for the collection of data mining directly useful information, the need for data processing matrix. The first is the data preparation process, in addition to the data format conversion and data integration focus on the data interpolation. Through self-interpolation method, and make full use of the most primitive data. The incomplete information and digital information interpolate, excepting for a small part of the data is not completed 99.9% of the foreign data interpolation good for interpolation to lay a solid foundation for the follow-up analysis.The forth part is data exploring. By scan the data find transaction background information redundancy and unusual transaction data. Take some step to correct it. The last part is modeling the data ,and analysis results. Firstly, the existence of the Chinese stock market disposition effect. Secondly, each of the buying and selling decisions, breakeven analysis. Personal transactions with hidden bodies difference in the investment transactions were used to the three-part decision tree analysis model discovered more attention to a low of 20 trading days; There is a different gender impact of the decision on the sale; The period of less than 50 days under the circumstances, the possibility of losses, This conclusion of the transactions were not completed notable; The ability to profit showing a significant correlation with the overall, the overall decline in the process, greatly increasing the possibility of buying losses. This forecast is due to the limited capacity of the individual may not be able to detect when in the end; institutional investors in the analysis demonstrated a good deal of profitability, and the relatively low frequency of transactions.