Research On The Interactive Relationship Between Financial Messages And Important Market Variables In China’s A-share Market

Stock market is an important part of capital market and it plays an important role in optimizing capital allocation, financing, preserving and appreciating asset value. As a typical representative of China’s stork market, A-share market has an important impact on promoting the development of China’s economy. The development and change of A-share market can be measured by many economic variables, among which, the market returns, trading volume and market volatility are the most typical ones. Thus, it helps a lot to judge the development of A-share market by studying the changing trend of these important variables. There are many factors that affect the important variables of A-share market and the information theory suggest that information is a very important factor driving the price of stock market. In this paper, based on the information theory, we select the characteristics of financial message in A-share market to study their same period, short-term and long-term effects on market returns, trading volume and volatility. It has an important practical significance for both theoretical research and practical guidance.First, in this paper, the relevant financial message variables and market variables in A-share market have been constructed. Based on the mixed distribution hypothesis, the reasons of selected variables are described, and the variables in this paper are constructe d by using logarithm, GARCH model and Black-Scholes model. Thus, market returns, tr ading volume and volatility have been chosen as the main research variables in A-share market, while positive message quantity, negative message quantity, positiveness and ag reement are selected as the measurement of message content and message quantitySecond, descriptive statistical analysis of the constructed variables have been done and basic statistical characteristics have been obtained. Then, the regression model have been utilized to analyze the correlation between the original variables preliminarily. The results indicate that the correlation between financial message and market variables is not significant. Also, there are some problems, such as the significant degree of some economic meaning variables is not high and fitting effect is not ideal.Third, to overcome the problems of regression model, the static interaction effects between financial message and market variables have been analyzed and the model have been improved. Stationary test of original sequence have been carried out and least square method combined with autoregressive moving model ARMA(p,q) have been used to eliminate time trend and sequence correlation, aiming to change non-stationary sequences into stationary sequences. Bidirectional regression model of financial message and market variables have been constructed based on the obtainedstationary sequences and the static interaction effects have been analyzed. The resultsshow that there are both bidirectional and unidirectional static interactive effectsbetween financial message and market variables.Finally, the dynamic interaction effects between financial message and market variables have been analyzed to improve the original regression model. The Distributed Lag Model and Vector Error Correction Model(VEC) have been utilized to analyze the short-term and long-term dynamic interactive effects of financial message and market variables in A-share market, respectively. After analysis of short-term dynamic interactive effects, the causality between financial message and market variables is examined by Granger Test. Then Unit Root Test is done on the original data and cointegration relationship between the variables is judged by E-G method, in order to construct Vector Error Correction Model in analyzing the dynamic interaction effects between financial message and market variables. The results show that there are both bidirectional and unidirectional short-term dynamic interactive effects between financial message and market variables. The unidirectional short-term dynamic interactive effects are more significant. Besides, there are only unidirectional long-term dynamic interactive effects of volatility and agreement on trading volume. The results prove that messages do affect the market and the results also demonstrate the effectiveness of information theory in analyzing the A-share market.The results of this paper help the market supervision layers, listing corporations and investors judge the current and future trend of the stock market observing financial messages. At the same time, it also provides a reference for the news media on how to forecast the trend of consensus by observing the performance of the stock market.