Research On The Impact Of Investor Sentiment On Stock Market Return Volatility Based On Noise Trading

Investors are the cornerstone of the entire capital market ecology.In recent years,with the in-depth study of behavioural finance in the stock market,there is increasing evidence that investors’ cognitive biases and emotional biases are important factors contributing to stock market volatility.For an emerging stock market such as the Chinese stock market,the application of investor sentiment theory can explore in depth the irrational causes of stock market returns and volatility in addition to fundamental factors,thus providing an empirical basis for the prevention and mitigation of financial system risks.Based on existing academic research,this paper uses theoretical analysis and empirical testing methods to investigate the relationship between investor sentiment and the impact of stock market return volatility.Given the difference in trading strategies between sentimentdriven individual investors and institutional investors,this paper examines individual and institutional investor sentiment separately.The research includes,firstly,expanding the DSSW(De Long-Shleifer-Summers-Waldmann)noise trading model,analyzing the impact of individual and institutional noise trader sentiment on stock market return volatility through mathematical modeling theory,and then two hypotheses were put forward.Secondly,on the basis of market proxy variables reflecting investor sentiment,text analysis methods were combined to mine investor sentiment information in online social networking platforms,and partial least squares regression was used to measure the investor sentiment index.Then,based on the results of the directed acyclic graph analysis,a fixed sample period and recursive forecast variance decomposition are used to investigate the contemporaneous relationship between stock market volatility and investor sentiment and its dynamic impact.Finally,a threshold regression model was developed to analyze the differential impact of investor sentiment on stock market returns within different zone regimes.The findings of this paper are shown below.Firstly,the results of the study based on the analysis of directed acyclic graphs suggest that stock market volatility and changes in individual investor sentiment caused changes in institutional investor sentiment in terms of contemporaneous effects.The results of the fixed sample period based prediction error variance decomposition and the recursive sample period based prediction error variance decomposition analysis further indicate that individual and institutional investor sentiment has some influence on stock market volatility,but this influence is generally diminishing;individual investor sentiment has a ’self-fulfilling’mechanism and is increasing,individual investor sentiment has an increasing influence on institutional investor sentiment.Secondly,the results of the threshold test indicate that the current stock market return variable is the threshold variable and that the threshold variable classifies stock market returns into three intervals: low,medium and high.Further analysis shows that within the low return interval,i.e.when stock market returns are negative,there is no significant linear relationship between sentiment and stock market returns for either institutional or individual investors.In the middle interval,both individual and institutional investors’ sentiment had a significant negative impact on stock market returns.In the high return interval,i.e.when stock market returns are positive,individual investor sentiment is positively correlated with stock market returns,implying that high individual investor sentiment brings about further increases in stock prices.However,there is no significant correlation between institutional investor sentiment and stock market returns in the high-return range.