The Study Of Interactive Relationship Between Online Public Opinion And Stock Market Return

With the Internet technology of China changing with each passing day, and continuous reforming and maturing in capital market, an increasing number of new-generation investors become more involved in the communication and discussion in online stock communities, online public opinion gradually become important factor in the stability of the stock market. So the research on the relation between online public opinion and the stock market in China has certain necessity and urgency. In this context, this thesis explores the interactive relationship between online public opinion and stock index returns. In order to fully understand the interaction mechanism between online public opinion and stock index return rate, after studying some articles about this subject both domestic and overseas, this thesis clarify the definition of online public opinion, the concept of abnormal stock price fluctuations and the theory of efficient market hypothesis and behavioral finance, and points out online public opinion can reflect investor sentiment and influence investor sentiment, at the same time irrational investment behaviors caused by investor sentiment can increase the volatility of the stock market. Then, this thesis quantifies the online public opinion by means of ROST, which has sentiment analysis function, makes the online public opinion become time series of sentiment index. Then we make contrastive analysis and correlation test between online public opinion and stock index return rate, concluded that there is strong correlation between them. After these, this thesis study the interactive effects by vector autoregressive model, we found that the news public opinion can influence stock index return rate more than investors' public opinion, stock index returns can in turn influence both of them significantly. However these impact and influence only exist in short term, and last for less than three days. At the fifth part, the thesis tested the asymmetric effect of interaction between online public opinion and stock index return rate by TGARCH-M model, we found that, on the one hand, good news has more influence on stock index return rate than bad news, and positive online public opinions has more influence on stock index return rate than negative online public opinions, but this kind of asymmetric effect is not wide-spread in our stock market. On the other hand, good news has more influence on online public opinion than bad news. too. It means good news tend to get more attention of investors than bad news, and further create relevant online public opinion. The influence the stock index return rate has on online public opinions, however, has no asymmetric effect.