The Effect Of Institutional Investors Sentiment Under Different States Of The Finance Market

In recent years,my country’s financial market has often experienced financial anomalies of "swelling and plummeting".Traditional financial theories are difficult to explain,but behavioral finance can be explained from the perspective of investor sentiment.However,the current research on investor sentiment mainly focuses on the irrational behavior of individual investors,while ignoring the behavior of institutional investors themselves.Therefore,this paper conducts a reasonable study of investor sentiment based on the state of different stock markets.It is mainly divided into four parts: indicator construction,investor sentiment fitting,market division,and the relationship between investor sentiment and RME and RMV.First,based on the CICSI Sentiment Index and BW Index,combined with the principles of accuracy,systematicness,scientificity and operability of index selection,comprehensively select the Shanghai Composite Index related trading volume,new account openings,market optimism index,buy Eight indicators,including index,confidence index,price-earnings ratio-Shenwan 300,investor sentiment index,and Shanghai Stock Exchange base index trading volume,build an investor sentiment indicator system.Secondly,based on the principal component analysis and partial least squares method,the investor sentiment is respectively fitted.The principal component model selects the first three principal components as comprehensive variables,and the cumulative contribution rate reaches82.82%.The principal component score data of each sample is used as the Changes in investor sentiment.At the same time,the principal component score and partial least squares fitting value are compared with the actual closing price of the Shanghai Composite Index.The result shows that the fitting effect of the partial least squares method is better than that of the principal component analysis method.Then,based on the market judgment standard of Pagan et al.,the BB rule is used to divide the stock market state.This paper adds the concept of mid-market,and divides the monthly data from January 2010 to January 2020 into three markets: bull market,bear market and mid-market,with a total of 11 time periods,in which the number of time periods included in each market is 3,5,3.Finally,the relationship between investor sentiment and equal-weight(RME)and weighted(RMV)daily returns is studied based on the GARCH family of models.The logarithmic first-order difference is performed on the original data to obtain stationarity,and according to the heteroscedasticity of the residual sequence,GARCH,T-GARCH,and E-GARCH models are selected for research,combined with AIC and SBC criteria,and suitable lag terms are used to fit Variance,and finally select GARCH(1,1)model,TGARCH(1,1)model and EGARCH(1,1).The model results show that institutional investor sentiment has not changed the relationship between asset returns and risk;in the stable period of the bull market and the optimistic period of the medium market,institutional investors are optimistic about the reversal effect of small and medium-cap stocks;in the pessimistic period of the bear market,institutional investors are optimistic about the large market Momentum effect of stocks.