| At present, micro-blog, WeChat, Post Bar and other social media is full of the Internet. In the whole social media, personal information is released is information transfer and receiver. The Internet from a simple information publishing platform evolved into social media platform. Especially on China's stock market, although since the establishment of Shanghai and Shenzhen two city has maintained rapid development, stock trading technology has reached the level of foreign countries, but the environment in the market transactions, trading mechanism and regulatory policy still needs to be improved, market is irrational, "policy market", "fair" filled the entire stock market. The stock market information about dragons and fishes jumbled together in social media spread a lot of false negative information, together, become a powerful trend and "herd" effect, the healthy development of troubled China's stock market. In 2015 twelfth The third session of the Tenth National People's Congress, Premier Li Keqiang proposed the new concept of "Internet", to promote mobile Internet, cloud computing, data, combined with the networking and the modern manufacturing industry, the importance of Internet technology improved to a new strategic. Therefore, how to accurate identification of network information influence on stock returns, to clarify the relationship between the Internet and the stock market's information, provide the basis for policy, it is of great significance.Based on the above research background, we will quantify the Internet data as the data source of investor sentiment, to study its effect on stock returns. This article as applied economics papers, follow the general questions, analyze issues and solve problems. Based on the existing literature review, found that there are two major problems:on the one hand the theory is insufficient, the two is in the quantitative investor sentiment does not make full use of information. To solve the first problem, this paper found that the traditional finance effect on stock market return factor analysis is more mature, but the subjective factors is seldom considered, and the behavioral finance theory can make up for its shortcomings, therefore this paper introduced three factors the model make up the basis of existing theory. For the second problem, this paper crawls through the stock forum data to quantify investor sentiment, the introduction of model points Analysis of its impact on the stock market returns.In the part of empirical analysis, the first is the study of investor sentiment index and Shanghai Composite Index closing price, trading volume, the correlation between yields, according to two two different variables for time synchronization, leading correlation lag.The second part is the analysis of investor sentiment and stock market liquidity.In the last part, based on the three factor model, to split SSE 50 index on the object of study, whether the research will introduce three factors in the model of investor sentiment on stock returns after the model explanatory power significantly improved.The innovation in this paper are the following two points:Firstly, as the starting point of the problem, establish investor sentiment mining based on Internet data, analysis of the impact of network public opinion on the stock market, to provide new ideas for related researchSecondly, technical, abandon the dependence on the emotional analysis software, based on the preparation of R web crawler program, the formation of its own database, for the next step in the analysis of the text data accumulated experience.This paper still needs to be further improved:Firstly, improve their data crawling the ability and efficiency of written crawler software capable of automatic data capture, and now on the market of the emotion software quality problem, developed specifically for the analysis of stock market sentiment word dictionary based on. This will improve segmentation and analysis of the emotional quality and effect.Secondly,In the empirical analysis, this paper is still based in traditional FF model analysis of the influence of investor sentiment on stock market returns, this analysis method can promote the further research on the theory, but for guiding the investors to the specific investment unrewarding. In the next study can corresponding stock selection model for the development of different sectors of the stock market and found stock market ups and downs of the rules, a higher yield than the market portfolio, guide retail investors are rational investment, timely to avoid the crash, ensure the expected benefits... |
PDF Full Text Request