An Empirical Study On The Influence Of Investor Sentiment On Stock Returns

Behavioral finance theory believes that irrational investors will not only exist for a long time,but will increase their impact on the market as the number increases.Scholars have pointed out that investor sentiment is an important factor for irrational traders to cause market fluctuations.According to data from the Wind database,as of the beginning of 2021,China's stock market has 4186 listed companies with a market value of 80.16 trillion yuan and 170 million investors.However,it must be seen dialectically that the overall quality of investors in my country's stock market is not high,and market supervision needs to be improved,which can easily lead to more speculative transactions.There is a close connection between investor sentiment and speculative trading.So what kind of "role" does investor sentiment play in my country's stock market? First,we must know how to describe investor sentiment in my country's stock market,and secondly,we need to empirically study the impact of investor sentiment on my country's stock market.The data studied in this paper are mainly taken from the CSMAR databases and Wind databases.Due to the lack of data,the time span of the data is from June 2012 to December 2020,and the data frequency is all monthly.The data needed to measure investor sentiment includes sentiment proxy indicators and control variables.Sentiment proxy indicators include closed-end fund discount rate,monthly IPO volume,Hu Shen300 ETF trading volume,market turnover rate,investor confidence index,and the number of new accounts opened each month,The control variables include the Shanghai Stock Exchange Composite Index's rate of return,CPI,PPI,IAV and MBCI.Investor sentiment's impact on stock returns needs to be selected from the stock's characteristic and Fama-French three factors.The characteristic include stock liquid market value,price-to-book ratio,and historical volatility,etc.The three factors are taken from CAMAR,calculated according to the market value of circulation.There are two methods to measure investor sentiment,namely principal component analysis(PCA)and partial least squares(PLS).Through qualitative testing,inspection and quantitative "twin share" return to judge the merits test measurement methods;In the part of studying the influence of investor sentiment and stock returns.Firstly,perform ADF test,Granger causality test and ARCH effect test on the variables.Secondly,use the GARCH model to test the impact of investor sentiment on stock returns.Finally,Use nonparametric analysis and multiple regression analysis to verify the sensitivity differences of stocks with different eigenvalues to investor sentiment.The article is divided into six chapters.The chapter one is the introduction.Firstly,explain the research background,problems and significance of this article;secondly,explain the research methods and research framework;finally,summarize the research innovations and research deficiencies.The chapter two is literature review.Mainly focusing on investor sentiment theory and investor sentiment Measurement methods,the impact of investor sentiment on stock returns,and related empirical research are carried out in four aspects.The Chapter three is theoretical analysis and research hypotheses.Specifically introducing the DSSW model and investor sentiment seesaw theory,providing the main support for the research hypotheses.The chapter four is the measurement of investor sentiment,including the selection of sentiment proxy indicators and the comparison of measurement methods.The chapter five is the impact of investor sentiment on stock returns,first part is a univariate test;In second part,use GARCH model to verify investor sentiment and stocks;In the ending part,non-parametric methods and multiple regression are used to test the relationship between stocks with different eigenvalues and investor sentiment.The chapter six is summary.Summarizes the research results,and puts forward relevant suggestions and future prospects.There are two main innovations in this article.First,the Hu Shen300 ETF trading volume and investor confidence index are added to the selection of investor sentiment proxy indicators.The former can reflect the speculative nature of the market,and the latter can reflect the expectations of investors participating in stock transactions on the market,and exclude Excluding the first-day yield of IPO,because After the pilot registration system for the Science and Technology Innovation Board and the Chi Next Board,the IPO yield after 2019 will far exceed44%,affecting the accuracy of investor sentiment measurement;second,According to the investor sentiment seesaw theory,increase the eigenvalue standard for stock classification;Thoroughly study the sensitivity coefficients of stocks and investor sentiment under different eigenvalues,and propose different research results.