The Particularity Of The Momentum Effect Empirical Studies In The Chinese Stock Market

The purpose of this paper is to explore the effect of particularity of the Chinese stock market to the results of momentum strategy through testing and analyzing the validation of momentum effect among the sample stocks of the CSI300Index.First, the article makes an instruction on the traditional financial economic theory, including the efficient market hypothesis (EMH), the traditional CAPM model and its extensions. Consistent with the method described in Jegadeesh and Titman (1993), we construct investment strategies J/K/Type, with J and K represent "the Sorting Period" and "Holding Period" respectively. Type is designed in order to analyze effects of cash dividend on strategy J/K, which is to distinguish the strategies built on different kinds of returns, with possible value1or0. We use Monthly returns and close of Chinese stock market database of CAMAR between May2005and December2011and change values of J and K to test the momentum effect. Combining the information of the CSI300Index and the reality of Chinese A-share market, we analyze effects of J/K/Type strategies resulting from cash bonus, equity weight and the β coefficient of stocks with different values of J and K.We can’t see obvious evidence for existence of momentum effect merely based on average monthly returns of portfolios. Conversely, changes of J and K do not change the conclusion that losers beat winners among CSI300Index components.Besides, we find that, generally, the shorter of "Sorting Period" and longer of "Holding Period" indicates higher possibility of more returns. With account of particularity of Chinese stock market, such as scheme of dividend payoff and existence of heavyweight stocks those make China different from the characteristics of other developed countries, we question the argument that there is no momentum effect in Chinese stock markets, holding the opinion that in order to validate the effects of momentum, factors like cash dividends, equity weight and sample range will play their roles. For example, both the relationship between stocks’market weight and β and effects of heavyweight stock on other stocks’ β are analyzed, empirically and theoretically. We find stock’s β coefficient is likely to increase first and then decrease, with their weight increasing. Besides, when stock’s weight is high enough, β is more likely to be less than1.