The Interaction And Risk Transmission Between Investor Sentiment And Stock Price Fluctuation

Relation between investor sentiment and stock price has been a hot topic in the study of behavioral finance.Starting from the micro level of individual stocks,to explore the interaction between investor sentiment and the stock price is useful for individual investors to make more sensible investment decisions.It is also conducive to regulators to control systemic risk.The stock market is a complex system.In this system,the investor sentiment of individual stocks influences each other,which changes the trading behavior of investors and makes the stock price change.The co-movement feature of stock price makes these influences continuously spread to the whole stock market.At the same time,the stock price,as the result of investment,will give feedback to investors,and then have an impact on investment sentiment.When studying the many-to-many relationship between the sentiment of individual stock investors and the stock price,we should not only consider the influence of the sentiment on the stock price,but also consider the negative influence of the stock price on the sentiment,which makes the problem become a new dynamic complexity problem with multi-subjects and multi-relationships.Therefore,on the basis of econometric methods,this article introduces the management science and complex network modeling method,based on which the investor sentiment index is proposed,investor sentiment and share price interaction model is built,and dynamic risk spread within the investor sentiment – stock price model is traced,the main work and innovations are as follows:(1)On the basis of existing research results,this manuscript focuses on the micro level and proposes individual stocks investor sentiment capture algorithm.Through the rationality test,four indexes are selected as alternative indicators,based on which a daily investor sentiment index is finally built.Among the four indicators,the opening sentiment index fills the gap that the daily sentiment index cannot be captured during the daily trading closure.The steps are as follows: first,determine the scope of proxy variables affecting sentiment and optimize the construction process of investor sentiment indicators;secondly,take these proxy variables into rationality test;third,construct investor sentiment index by principal component analysis.Finally,the validity and superiority of the investor sentiment index constructed in this paper are verified from the rationality and robustness of the indicators by qualitative comparison and quantitative test.(2)On the basis of econometrics method,the complex system modeling method is introduced,and the construction methods of sentimental fluctuation correlation network,stock price correlation network,and emotion-stock price bilevel volatility correlation network model are proposed.This paper combines the interaction and correlation between the sentiment and stock price of Chinese financial listed companies,and makes an in-depth analysis from the perspectives of interaction identification and asymmetry.It is found that investors sentiment of financial stocks in sector is more vulnerable to internal shocks than stock price,and the correlation is stronger than stock price.According to the different sentiment and the shock bearing capacity of the stock price,the stocks are basically divided into two categories: sensitive stocks such as stock 300059.sz and stable stocks such as stock 002939.sz.In addition,through the four-mode detection method,this paper found five emotion-stock price interaction modes with the highest frequency,showing that sentiment is always the transmission source and stock price is always influenced.Finally,the asymmetric characteristics of the influence between sentiment and stock price are found.(3)The paper constructs the step-risk propagation model and the time-risk propagation model of investor sentiment-stock price,showing the mechanism of risk propagation under the impact of stock market,and verifies the effectiveness of the model.Empirical simulations are carried out under a variety of market situations.In the stock price network,when the stock is affected to the same degree,the time required for a bear market is 2 days faster than that for a bull market,and it is about 20 days faster than that for a stable market.The time difference of the three is enlarged in the emotional network,and the differences in the doublelayer emotion-stock price network are the most obvious.This shows that the extreme market is much more affected and risk spreads faster in extreme market than in the equilibrium market.The stock's risk diffusion ability is the strongest in the bear market,which takes only about 2 days for all stocks to enter the stop state,while the bull market takes 15 days,and the stable situation takes about 7-23 days.The experiment found that under stock stop mechanism,the impact of the stock sharply reduced,under the sentiment stop mechanism,the impact of the stock gets even smaller.Finally,it is suggested that investors should return to rationality as far as possible to avoid blindly following the trend.relevant departments should pay attention to the individual investment sentiment,improve the information disclosure system,timely release of authoritative information and enhance the anti-risk ability of the stock market.