An Investigation Of The Market Reaction To The News Of "Put In ST" And "Cancel ST "

Special Treatment (ST) has been put in practice in China stock market since 1998. Aiming at setting up the out of market mechanism, increasing the transparency of market and revealing risk to the investor, it is a particular institution in China stock market, which can not be found in any other countries. Investigating how the stock market would react to the news of "put in special treatment" and "cancel special treatment ", this paper test the semi-strong efficiency of China stock market, summarize the market reaction pattern of such kind of news, and provide some suggestions to the investor and supervisor. This paper uses the method of event study, which has been widely used in literatures to test the semi-strong efficiency of financial market. Our results show that in the aspect of return-generating model, compared to the traditional market model, market model modified by GARCH is more convictive since it takes volatility cluster into consideration. By doing parametric test of the distribution of abnormal return, the paper demonstrate that the semi-strong form of the EMH does not hold in China stock market. Meanwhile, both similarities and differences are found in the reaction pattern of the news of "put in special treatment" and "cancel special treatment ". There is delay reaction and converse reaction to good news and overreaction to bad news. These phenomena show the existence of leverage effect in China stock market. The innovations of this paper are as follows. First, the samples of our study are ST stocks that experienced both "put in special treatment" and "cancel special treatment ". By putting both the "put in special treatment effect" and "cancel special treatment effect" together, our investigation is more complete than other literatures before. Second, in the aspect of return-generating model, in order to take volatility cluster into consideration, the paper introduces the market model modified by GARCH. Third, the most traditional event study method, in which the estimation window is completely separated of the event window, is not used in this paper. Instead, the author introduces an estimation window with a constant length, which is updating continually with the date. This kind of window setting fits the behavior habit and psychology of the investor better.