Momentum, Contrarian, And Patterns Of Information Response In China Stock Market

Momentum and contrarian arc the strongest and most puzzling asset-pricing anomalies in stock market. This study mainly investigates the investment performance of momentum strategies and contrarian strategies and the mode of information response in China stock market. There are four purposes in this paper. First, we examine whether the momentum strategies and contrarian strategies can create significant profits under different formulation horizons and holding horizons, whether past factors (market return, characteristic of individual stock) can provide an important implication about the profits of momentum and contrarian strategies. Second, we discuss the reasons for the significant profits of momentum or contrarian strategies, including seasonality, cross-sectional risk factors, time-varying risk premium, industry momentum, and stock underreaction, overreaction, and random walk. Third, we discuss the link of time series predictability of stock returns and momentum profits, including stock underreation, overreaction, delayed reaction, and time-varying risk premium. Finally, we study the special mode of information response in China stock market. Based this, we interpret short-term underreaction and overreaction, and analyze the patterns and mechanism of lead-lag relationship of equity returns.The main conclusions and suggestions are as follows:First, we find the momentum strategies under different formation horizons and holding horizons (1-48 weeks) are more successful in China stock market as a whole. Concretely, medium and short term strategies(over 4-24 weeks holding period) generate significantly positive returns. In addition, super-short term strategies also earn significantly positive returns. Nevertheless the contrarian strategies under long formation horizons and holding horizons ( beyond 36 weeks) may yield significantly positive returns. We also find, considering markert factor (market return, cross-sectional variance of equity returns) and characteristic of individual stock (trading volume, firm size, price, price/earnings, BE/ME), there exist clearer and more significant patterns of profitability in momentum or contrarian strategies.Second, for the reasons of the profits of the momentum or contrarian strategies, we find: (1) Momentum or contrarian strategies exhibit an interesting pattern of seasonality in China stock market different from January effect found in mature market, which seems to exist a link with the cycle of disclosure of accounting information for fiscal year. (2) For cross-sectional risk factors, both CAPM and Fama&French three-factor model cannot account for profitability of medium and short term momentum strategies, but for long-term strategies, may explain a little. (3) For time-varying risk premium and industrial momentum, serial autocorrelation of factor returns including size, BE/ME, and industry contributes mostly to momentum profits, especially industry momentum. (4) For the debate between behavioral models and random walk, the reasons of profitability of strategies with different formation period differ widely. Medium and short term momentum is driven primarily by underreaction. Super-short term momentum strategies perform worse, which attribute to short-term overreaction resulting by noise trading, In addition, for long-term strategies, there exist significant mean reversion in return and overreaction phenomenon.Third, for time series characteristic of momentum profits, described by behavioral models, because of inherent biases in the way that investors interpret information, a variety of irrational patterns of information response result in individual stocks underreacting to information or covarying more strongly, and generate time series predictability of equity return. Thus momentum and contrarian profits may arise in a variety of time series ways. We investigate time series characteristic of momentum profits in china stock market, empirical results show momentum profits arise in a variety of complex ways. For medium and short termstrategies, time series...