Momentum On Chinese Stock Market: Empirical Analysis And Source From Size

Momentum is one of the strongest and most puzzling asset pricing anomalies. Jegadeesh and Titman(1993) shows that past winners continue to outperform past losers over horizons of 3-12 months. Domestic research also shows that there exists such a phenomenon in Chinese stock market. This paper first examines whether momentum exists for all the stocks in A-share market, including Shanghai stock market and Shenzhen stock market, from January,1994 to December, 2003. It is concluded that in the short time momentum is significant in the market, but in the long run the profitability of contrarian strategy is observed. In addition, this paper examines the subsample from January 1996 to May 2001 as the robust test in order to determine whether such a phenomenon is consistent in the market. Based on the past conclusion that the profitability of momentum is closely connected with the synchronic risk faced by the individual stocks such as size, analyst coverage, and B/M ratio, this paper separately tests the profitability of momentum strategy for big-size companies and small-size companies. The result shows that strong momentum is observed for big companies not only in the short run but also in the long horizon. On the contrary, the small companies are enjoying the profitability from momentum strategy in the short run but from contrarian strategy in the long horizon. Such a pattern for the small companies is obviously keeping with the pace of that of the whole A-share market. In order to determine the source of the momentum and contrarian profits, this paper tests the autocorrelation and cross-section correlation between the big-size companies and small-size companies. And bootstrap method is adopted to verify the significance. It is found that there exist obvious autocorrelation for different-sized companies. Moreover, obvious cross-section correlation and cross-predictability are observed between different sized companies. The results indicate that there exists predictability not only for the same quintile of companies but also for different quintiles of companies.Based on the research on the behavior patterns of domestic investors, this paper concluded that the above results can be explained by the overreaction theory(DHS model) and information-diffusion model.