The Empirical Study Of The Relevance Between Sentiment And Earnings Of Investors In The Stock Market

In recent years,with the research of anomalies in the financial market,behavioral finance challenges classic financial theories at some level.In China,the stock market starts late,though it develops quickly,the fluctuation is really violent.Furthermore,the investors are irrational and immature in the actual operation of trading stocks.Therefore,in consideration of the current situation of Chinese market,this paper choose to do some dynamic analysis between the investor sentiment and stock returns.This article decided to define the investor sentiment as the deviation of investor expectation of the future price of the financial product.Secondly,the sentiment indexes are classified into three groups according to the access of the data,the obvious norms,the recessive norms and the sentiment proxy variables.Finally,the monthly investor confidence indexes which are issued by the Chinese Securities Investor Protection Fund Co.,Ltd are selected as the study indicators.After the analysis,it can be found that the variation of investor confidence index coordinates with the Shanghai index returns and the growth enterprise index returns.In the meantime,the degree of the coordination between the variation of the investor confidence index and the Shanghai index returns are much higher than the growth enterprise index returns.And then,the 1 lag period vector autoregressive model(VAR)is constructed to analyze the dynamic relationship between the investor sentiment and stock returns.Next,the impulse response analysis and variance decomposition are used to analyze the correlation and the investor confidence index and the Shanghai index returns,the growth enterprise index returns are reciprocal causation which are tested by the Granger Causality test.Next,the two steps method which are raised by Engle and Granger is used to establish the error correction model,which verifies the conclusion of the VAR model.In consideration of the fact that there exists correlation between investor sentiment and market trend,this article choose GARCH and TARCH models to test the asymmetric effect in Chinese stock market,and it has been proved that the bad news would have greater volatility than the good news.At last,in consideration of the asymmetric effect,the TARCH,EGARCH and CARCH models are selected to research the impact of good news and bad news in different stages of the market.In a word,the results are obtained by the time series model.However,due to the changes in the structure of investors and the limitations of the measurement indicators,the conclusions may not be able to fully reflect the current state of stock market.In the future,the individual differences and market trend should be considered as the main factors to do the research,and a comprehensive measurement index established by some multi dimensional factors should be taken into consideration,which would represent the investor sentiment well.The most important is that,it would provide a reference for the securities regulatory agencies and market participants to defend the high volatility and the risk.