Font Size: a A A

Network Affect Investor Sentiment On The Stock Price

Posted on:2017-02-03Degree:MasterType:Thesis
Country:ChinaCandidate:X H DingFull Text:PDF
GTID:2359330512463154Subject:Investment economics
Abstract/Summary:PDF Full Text Request
With the popularity of the Internet and the maturity of the social media network,people increasingly rely on the Internet to obtain all the necessary information,network information is becoming more and more important in people's daily life.With the increasing number of investors in China's stock market,in order to reduce the cost of information research,individual investors,which are in the disadvantage of information,are expected to choose the valuable stock investment information from the massive network information as the basis of decision.However,the authenticity of the network information can not be verified,and they are more likely to focus on those with a strong emotional color of false information,once these information staggered,the truth of the investors are likely to be illusory confusion,to take concerted action,and even the resulting "herding effect",resulting in asset price volatility.Therefore,it becomes very meaningful to more effectively identify and judge the impact of network information on stock price under the imperfect system of network information supervision,to explore the mechanism of network information's effect on stock price,the decision-making of investor's behavior,the implementation of relevant policies to provide a reference for reference.Based on the theory of behavioral finance,this paper,based on the framework of the noise trading theory,takes the posting and posting of stocks in the social media forum as the substitution variables of network investors,this paper uses text data mining and text categorization techniques to study the mechanism of investor sentiment on stock prices in the short term.First of all,this paper reviews and analyzes the theoretical basis and literature of investor sentiment,and on the basis of predecessors' research,it determines the research methods and ideas,and combines internet data with the background of internet data.Stock price fluctuation as the research focus of this paper;In the selection of investor network emotional variables,it is divided into two types: the network posting quantity numerical index and the investor sentiment index textual index.The investor sentiment index is the text data mining,sorting,aggregation and emotion classification.Then the composite index of emotional index construction,investor sentiment index is divided into investor sentiment tracking index and investor sentiment index.Secondly,in the study of the impact of investor sentiment on stock price volatility,the correlation between investor sentiment and stock price characteristics is analyzed.Then the panel regression model is established to compare the similarities and differences between investor network emotional numerical index and textual emotional index on stock price volatility.In order to complete the prediction of the investor sentiment,this paper studies the influence mechanism of investor sentiment on stock price volatility and the inverse effect of stock price volatility on investor sentiment.On the basis of the above research results,the article starts from the stock market cycle,respectively,to study the impact of investor sentiment on the fluctuation of stock price during the rise and fall of the market,to further verify the previous conclusions.Finally,the article concluded that,whether in the whole sample period,or in the stock market cycle,investor network and stock price fluctuations in the relationship between at least the following: The first point is the amount of web postings and stock returns stock price volatility,but the effect is less than the emotional tracking index,emotional tracking index can better explain and predict stock price volatility,investor sentiment is higher,the higher the yield,while the investor sentiment,the lower the yield low.The second point is the emotional convergence index can significantly affect the stock price volatility and turnover rate,emotional convergence index is lower,the greater the differences between investor sentiment,the greater the stock price volatility,stock turnover is also the higher.The third point is the amount of network postings can better reflect the actual participation of the stock market,posting the greater the higher the degree of market participation,and vice versa is.The fourth point is to verify the short-term,the stock market trends on investor sentiment has a significant network of influence,resulting in a two-way relationship between the two.
Keywords/Search Tags:Stock market, text mining, Web posting volume, sentiment index tracking, convergence sentiment index
PDF Full Text Request
Related items