Research On The Influencing Factor And Formation Mechanism Of Excess Volatility In Chinese Stock Market

Chinese stock market was founded in the economic reforms, and is special in the market structureand operation mechanism with the existence of price volatility of high frequency and largefluctuations. It has being a long-term and arduous task to find characteristics of excessive volatilityand the way to suppress it, so it still remains a problem.This paper begins with drawing on the latest research results at home and abroad, redefines themeaning of excessive volatility as a mixture of cross-section and underlying-derivatives abnormality,based on two-dimensional asset pricing plan, then indicates excessive volatility exists in the stockmarket because of behavior factors, after isolating the normal fluctuations caused by fundamentals.At the same time, the formation mechanism of excessive volatility is analyzed through three smallsections, completing the vision path of individual investors-groups of investors-market analysis.First, a survey obtained with the non-rational investors showed the existence of non-rationalfeatures of cognitive bias and deviant behavior in Chinese individual investors; and the causalrelationship between cognitive bias and deviant behavior is verified through structural equation model,in which, overconfidence impacts most among the cognitive bias and risk appetite in deviant behavior.Then, the formation and steady-state of groups’ non-rational behavior were analyzed with imitatebehavior model, which shows market sentiment will be formed by individual sentiment; meanwhile,according to the dispersion index, the validation gives out that group behavior is not obvious when thestock market increases but obvious when it goes down.Finally, the use of two non-symmetrical impact models, TARCH and EGARCH, verifies thenon-symmetry characterization of non-rational cognitive and behavioral phenomena in the stockmarket volatility. The results show that, the non-rational level is higher when market volatility isdown than upward fluctuations, and more possible to cause excessive volatility, which means theearly negative information will obviously lead to excessive volatility.