| Under the rationality Capital Asset Pricing Model CAPM (Capital Asset Pricing Model) theory’s support, by detecting the linear relationship between the CSAD (Cross-sectional Absolute Deviation) and portfolio returns,the CCK Model (Chang, Cheng and Khorana (2000)) tests whether the herd effect exists in financial market. Based on this model, many scholars have studied the stock market, and found a different degree of herd behavior, particularly domestic herding effect is more obvious. At present, the research on the futures market herd effect is relatively lacking, only some of the literature, the study of individual varieties. This paper studies the overall perspective of herd behavior, the commodity futures market in China futures market to make use of CCK model first least-squares regression, it concluded that China’s futures market as a whole no herd behavior. Since traditional least-squares regression only consider conditions mean regression, cannot detect the tail information, in order to do more comprehensive research herding effect influence on the futures market, further model to make use of the quantile regression, we still did not found herd behavior. In reality, the herd behavior is occurred in partial, so this paper further divided the whole market into a number of plates, the least squares regression’s result is that metal the plate and oil plate has a faint herd behavior, but quantile regression concluded that a low quantile degree of CSAD have relatively severe herd behavior. Under the least-squares regression’s results without herding plate, but through the quantile regression, the site also found a weak herd behavior in the low tail. It embodies the superiority of quantile, and it can be more comprehensive detection of local information, the conclusion of the differences between different quantile interval, is helpful for investors to choose a better portfolio, to increase the efficiency of the futures market. |
PDF Full Text Request