Empirical Analysis Of Investor Sentiment And Its Impact On Stock Market Returns

Under the efficient market hypothesis and pure competition conditions,the transaction price of financial assets can reflect all market information.Even if there are some less rational transaction behaviors in the market,asset prices can still fluctuate near the benchmark level through competitive arbitrage.At this point,the public information in the market can be used to predict asset prices,but the reality is far from that;In the analysis of China’s stock market,we should start with behavioral finance and consider the irrational situation of investors.Based on the monthly data of csi 300,this paper constructs the investor sentiment index first,and then builds the VAR model about the dynamic relationship between investor sentiment and stock market return rate.The main research work and conclusions are as follows:(1)Construct investor sentiment index.According to the actual situation of Chinese stock market,seven sentiment indicators reflecting the characteristics of Chinese stock market are selected,and the investor sentiment index,sent,is constructed by principal component analysis.At the same time,in order to solve the possible lead and lag problems of investor sentiment,this paper adds lag value of each index as a solution;Multiple regression method is used to eliminate the influence of macro factors and remove the autocorrelation and heteroscedasticity of investor sentiment.The results of this part show that the number of IPO,IPO first-day earnings,closed-end fund discount rate,the number of newly opened accounts,monthly trading volume,monthly transaction amount and price/earnings ratio can be used as proxy indicators to effectively measure investor sentiment in China’s stock market.Principal component analysis can extract the trend of proxy variables well.(2)VAR model is constructed.Based on the stock market return rate and investor sentiment index,a vector autoregressive model is constructed,and granger test,impulse response analysis and variance decomposition analysis are carried out to test the causal relationship and dynamic relationship between investor sentiment and stock index return rate.The results of this part show that there is a Granger causality relationship between sentiment and stock index return rate,and there is a positive feedback loop between them.The increase of the return rate of stock index will make investors optimistic,while the decrease of the return rate will make investors pessimistic,and this effect is asymmetrical.Emotional volatility will also affect the change of return rate,which will gradually disappear and weaken with the increase of the number of periods across time,there is a time lag effect.