In this paper, the authors present a method to construct a simulation experiment system which used to prove how social network affects the stock market. The authors construct a model considered of three main stock mar...In this paper, the authors present a method to construct a simulation experiment system which used to prove how social network affects the stock market. The authors construct a model considered of three main stock market effect factors: cognitive factors, social factors and economic factors. In the simulation system, free-scale social network is generated for information relations among investors in stock market. The authors simulate the organization information propaganda processes and generate stock related news. The generated news contains economy factors and sudden events. Stockholders cooperate with the related stockholders in their social network, and make stock exchange decision based on this news and their cognitive understanding of this news. Simulation results shows that the social network news propaganda effect stock prices. If social network in stock market have circles, the news affection will be amplified.展开更多
Based on the text of posing on the network forum, the authors establish a set of keyword dictionary to measure the long and short investors' sentiment effectively, then they investigate the mutual relations between i...Based on the text of posing on the network forum, the authors establish a set of keyword dictionary to measure the long and short investors' sentiment effectively, then they investigate the mutual relations between investor sentiment and the trading market through a Multivariable BEKK-GARCH model of abnormal long and short investor sentiment, returns, and abnormal trading volume. The results show that abnormal sentiment of long investor has a negative impact on the returns and a positive impact on the abnormal trading volume; while abnormal sentiment of short investor has no impact on the return and a negative impact on the abnormal trading volume. Otherwise, there is the negative volatility effect from abnormal sentiment of long investor to returns and abnormal trading volume, the positive volatility effects from abnormal sentiment of short investor to returns, and no volatility effects from abnormal sentiment of short investor to abnormal trading volume. In addition, network forum investor sentiment is a factor affecting the trading market. The analysis of forum information plays a certain role in presenting market risk and improving efficiency in making investment decision.展开更多
Some widely-used technical indexes of stock analysis are introduced as input of BP neural networks for the prediction of ups and downs of stock market, and better accuracy of prediction is achieved. A jump training st...Some widely-used technical indexes of stock analysis are introduced as input of BP neural networks for the prediction of ups and downs of stock market, and better accuracy of prediction is achieved. A jump training strategy and three varying training ratio methods are used to accelerate the training iteration. An online prediction strategy is applied to monitor the training iteration procedure. The ratio of central distances of prediction examples is defined, in order to locate the un-stable prediction examples.展开更多
文摘In this paper, the authors present a method to construct a simulation experiment system which used to prove how social network affects the stock market. The authors construct a model considered of three main stock market effect factors: cognitive factors, social factors and economic factors. In the simulation system, free-scale social network is generated for information relations among investors in stock market. The authors simulate the organization information propaganda processes and generate stock related news. The generated news contains economy factors and sudden events. Stockholders cooperate with the related stockholders in their social network, and make stock exchange decision based on this news and their cognitive understanding of this news. Simulation results shows that the social network news propaganda effect stock prices. If social network in stock market have circles, the news affection will be amplified.
文摘Based on the text of posing on the network forum, the authors establish a set of keyword dictionary to measure the long and short investors' sentiment effectively, then they investigate the mutual relations between investor sentiment and the trading market through a Multivariable BEKK-GARCH model of abnormal long and short investor sentiment, returns, and abnormal trading volume. The results show that abnormal sentiment of long investor has a negative impact on the returns and a positive impact on the abnormal trading volume; while abnormal sentiment of short investor has no impact on the return and a negative impact on the abnormal trading volume. Otherwise, there is the negative volatility effect from abnormal sentiment of long investor to returns and abnormal trading volume, the positive volatility effects from abnormal sentiment of short investor to returns, and no volatility effects from abnormal sentiment of short investor to abnormal trading volume. In addition, network forum investor sentiment is a factor affecting the trading market. The analysis of forum information plays a certain role in presenting market risk and improving efficiency in making investment decision.
基金Supported by the National Natural Science Foundation of China(19971012)
文摘Some widely-used technical indexes of stock analysis are introduced as input of BP neural networks for the prediction of ups and downs of stock market, and better accuracy of prediction is achieved. A jump training strategy and three varying training ratio methods are used to accelerate the training iteration. An online prediction strategy is applied to monitor the training iteration procedure. The ratio of central distances of prediction examples is defined, in order to locate the un-stable prediction examples.