| China's financial markets and the Internet are in the early stages of rapid development.Today,Internet platforms have become an intermediate gathering place for public opinion.The rise of Internet public opinion has had a profound impact on the capital market.Effective market theory often fails to give a reasonable explanation for the emergence of financial market anomalies.Public sentiment expressed on social media and investor attention reflected on search engines may become new measures of stock markets in emerging countries and provide insights into investors' psychology.In 2018,the Changsheng biological problem vaccine event began to be exposed.Internet public opinion has pushed the problem enterprises to the forefront of the wave,and investors' abnormal attention to the pharmaceutical industry is increasing,which may further affect China's vaccine listed companies and even other pharmaceutical industry companies.This thesis aims to expand the perspective of information and the mechanism of public opinion dissemination,and explores the dynamic impact of online public opinion and investor attention on China's pharmaceutical stock market based on the faulty vaccine event.In this thesis,we extract public sentiment information from the public opinion of the faulty vaccine event in China.Based on the basic sentiment dictionary,a sentimentoriented vaccine topic-oriented sentiment dictionary was constructed using the sentiment point mutual information algorithm,and then sentiment scoring was implemented on the topic Weibo according to the sentiment fusion algorithm.Then we use the identification of Weibo opinion leaders to divide the two-level flow of communication,and analyze the public sentiment changes in the two-level communication process of the faulty vaccine event.Using the public sentiment in Sina Weibo and the investors' attention of Baidu search,as well as the daily and weekly stock returns of 95 pharmaceutical companies,we establish a three-variable panel vector autoregressive model to discuss dynamic effects of public sentiment and investor attention on the stock market from the perspective of public opinion supply and investor information demand.Using the public opinion mutation point test of time series,it is found that the biological vaccine industry and low market value Pharmaceutical Stock Index earnings are more vulnerable to the impact of negative public opinion.The results of Granger causality test show that public sentiment can provide useful information for predicting the stock return.Public sentiment and investor concern are leading factors for the changes in stock price of high-market-value pharmaceutical company.In the process of two-step flow of communication,the empirical evidences from generalized impulse response show that the impact of public sentiment in the second-step stage on the stock return is significantly less than that in the first-step stage.More interestingly,in the second stage of public opinion dissemination,the impact of public sentiment on low-market-value pharmaceutical companies is greater than that on high-market-value companies.The results of the cumulative impulse response further show that the public is even more distrustful of low-market-value pharmaceutical companies,and the public sentiment expressed on social media platforms cannot be underestimated.Finally,using the variance decomposition technology,it was found that public sentiment contributed a lot to the company's stock return.The public sentiment had little difference in the strength of contribution to pharmaceutical companies regardless of market value in the early stage,but in the later stage,companies with high market value received investment more attention.This study has valuable managerial implications and can provide new insight into the investment decision-making.Broadening the dynamic research of information perspective is conducive to strengthening the management of public opinion at the industry level and enterprise level of capital market. |
PDF Full Text Request