| Traditional economics considers that, the price is determined by supply and demand, and the intersection of the two curves is a balanced state, the implementation process of the price is considered to be"Black Box". But in the stock market and derivation market with huge trade volume, economists recognize that trading rules and other factors also influence the price, therefore it results in a micro-market theory, a mixture of distributions hypothesis which is used to analyze these markets. The relationship between volume and price is one part of the study. Most research before used the absolute value of residual margin of income self-regression to measure the volatility of return. According to the study of Anderson, etc. (2003), this paper uses the way of dividing small district to measure the volatility of return and other factors of sources, at the same time, this paper compares this approach with the previous method measuring the volatility of return, and finds their differences.In this paper, we take the popularly used Bayes learning model, expectation theory, and the GARCH return volatility theory as the basis of our model, then we choose 180 stocks traded in the Shanghai Stock Exchange and use least squares analysis to analyze the factors that related the volatility of the return of these stocks, finally, we class them in to different groups by the frequency of transactions and daily average circulation market value to get a more detailed analysis.Consistent with related research before, the frequency of transactions and order imbalance can give relative strong explanation in explaining the volatility of return. We also get the conclusion that they contain different information in explaining return volatility, even if both of them contain relative stable information in explaining return volatility; average trade size is week in explaining return volatility and is instable; squares of return is better than absolute residuals in coordinate the data. |
PDF Full Text Request