关键词: MRS-SJC-Copula模型, 修正的ICSS算法, 结构突变, 动态联动性

Abstract: This paper studies the linkage between different stock markets, which considers the structure changes of mean and variance due to the phenomenon of high and low volatility of stock markets in different economic growth cycles. Firstly, based on the dynamic SJC-Copula with Markov state transition, this paper measures the linkage between the mainland stock market and Hong Kong stock market. Then, using the modified ICSS algorithm tests the structural change points of the upper and lower tails. The empirical results show that there is a non-linear asymmetric time-varying dependence between domestic and Hong Kong stock markets and a probability transition between high and low states. The stock index is affected by the negative news more than the positive news because of the dynamic linkage, and the upper and lower tails are affected by the last information continually. “Shanghai-Hong Kong-Tong”, “Shenzhen-Hong Kong-Tong”, Sino-US trade war and other factors make the upper and lower tails have structural changes. The linkage between the mainland and Hong Kong stock markets has increased and the volatility of the markets has become stronger.

Key words: MRS-SJC-Copula model, modified ICSS algorithm, structural change, dynamic linkage