Empirical Research On The Impact Of Comprehensive Sentiment Index On Stock Market Based On Investor Limited Attention

By the end of December 2016,there are more than 3000 listed companies in Shanghai Stock Exchange and Shenzhen Stock Exchange.The total market value of the two stock exchanges is about 54 trillion,in which Shanghai Stock Exchange accounts for about 32 trillion while Shenzhen Stock Exchange accounts for about 22 trillion,ranked fourth and fifth in global securities market.The scale of China's stock market plays a major role in the development of global securities market.However,as a new market,China's stock market has only established for 26 years,and has a big gap compared with the developed stock market due to the imperfect development.The unique characteristics have been established in the course of the history of China's stock market due to the policy and market economy environment.The lack of all aspects of China's stock market has caused instability and unreasonable fluctuations of market and initiated a series of malignant events,such as "8.10" incident,"327 Treasury bonds" incident,"5.30" stock disaster,fuse mechanism stop,ect.On the face of it,these events were caused by an event.However,the fundamental problem lies in the imperfect stock market system.Chinese stock market has undergone about nine fluctuations since its established,and the amplitude of stock market in 2007 and 2015 reached the maximum value causing huge losses to investors.This volatility is not been able to be explained by the standard financial theory.Because of the imperfect laws and regulations of China's stock market,too much interference of government,unprofessional quality of investors,as well as the speculative atmosphere,investor sentiment would has greater impact on the stock market.The objective of this paper is to construct a comprehensive investor sentiment index to fully reflect the mood of Chinese investors,and study the relationship between investor sentiment and cross-sectional effect of China's stock market.Firstly,this paper confirmed that taking the amount of stock posted as the indicator of investor attention can well reflect investor sentiment,verified the rationality of taking it as the indicator of emotional and added it to the proxy indicators of structuring the comprehensive emotional indicators.Secondly,this paper takes the closed fund discount rate,IPO first day rate of return,IPO circulation,turnover,the number of new accounts,investor confidence index and investor attention as the proxy variable to build the comprehensive sentiment indicators,determined seven indicators after flitting each indicator of current and lag,and constructed comprehensive emotional index by Principal Component Analysis.Finally,different portfolios and corresponding comprehensive emotional indicators were constructed according to the relevant characteristics of the stock variables,including the market value,stock price,price to earnings ratio and Price-to-Book Ratio.The A shares of the motherboard market data during two "stock disaster" period in 2007 and 2015 were selected to be compared,and the GARCH(1,1)-M model was adopted to analyze the cross-sectional effect of investor sentiment on stock returns.The results show that the portfolio of small market value stocks,low-priced stocks,high PE and high PB can get more benefits when the sentiment is high.It finds that investors did not become more rational although experied the first stock disaster by the comparison of investors' reaction when they faced the two stock disaster.Insteadly,emotional changes has more impact on the stock returns.This paper firstly discovered the various visions and irrationals in the Chinese stock market,and researched the deep reasons from the perspective of market and investors.The results show that investor sentiment has a great influence on the volatility of stock market.Then,combined the influence of Internet on investment behavior,the investor's emotional indicators were researched based on the Internet platform by programming,and the investor's emotions were classified according to the stock characteristics.Finally,the GARCH(1,1)-M model was used to analyze the cross-sectional effect of China's stock market.