THE NONLINEAR MECHANISM OF INVESTORS’ BEHAVIOR AFFECTING THE MULTIFRACTALITY OF STOCK MARKET: EVIDENCE FROM CHINA

This study provides evidence of the role of investor herd behavior as a mechanism by which capital market openness affects the multifractal characteristics. A threshold network is constructed among the shares cross-listed in A-shares in mainland China and H-shares in Hong Kong. Then extract the largest sub-network and compare the herding effect and multifractal features corresponding to different network structures. The results show that (i) the tightness of network increases significantly after the implementation of the SHSCP, and time-varying indicator show that the short-term effect is greater than the longer one; (ii) the herd effect is not evident in the selected sample prior to SHSCP, however, it undergoes a substantial increase following its introduction, particularly during periods of extreme negative market volatility; (iii) the return series constructed from the set of nodes constituting the largest sub-network exhibits both the herding effect and the multifractal intensity more significantly than the return series constructed from a randomly selected set of nodes, and this effect intensifies significantly under extreme negative volatility condition.