Study On Herding Behavior In Shanghai Stock Market Based On Asymmetric Regime-Switching Model

According to traditional theory, the price of financial asset can fully reflect all available information of the asset, and according to the information obtained, investors can make rational choice which is independent and is not influenced by other investors’ behavior. However, in a real stock market, investors may ignore their own understanding of the information in decision making and imitate other investors’ behavior in the market, which is called herd behavior or herding. Herd behavior causes financial asset prices deviate from their true value and produces price bubbles. Moreover, herd behavior and financial crisis are closely related, therefore, it has aroused widespread concern in the academic circle and supervision departments. Especially in China mainlang stock market, due to the lack of operation experience, market information asymmetry, frequent policy intervention and shortage of investors’ skills, herd behavior exists significantly in China mainland stock market as opposed to mature stock markets in developed country or region such as US and Hong Kong.This paper introduces asymmetric terms into Markov regime-switching model and further establishes an asymmetric Markov regime-switching model based on aforementioned researches, and selects data of the stock daily return in Shanghai A-shares and B-shares, Hong Kong stock market and New York stock exchange as empirical samples, and compares the result with mature stock markets in US and Hong Kong.Through the dynamic study on the impact of "the good news" and "the bad news" on herd behavior in China mainland stock market, and estimates the two states of mainland stock market endogenously by introducing Markov regime-switching into CCK model and innovatively studies the difference of herding under different states in China mainland stock market. Moreover, this paper studies A-shares and B-shares respectively and compares the results with those of the mature stock markets in US and Hong Kong, so as to get more intuitive understanding of the characteristics of the herding in mainland stock market and the difference between developing stock market and mature stock market.The empirical results show that there is significant herd behavior in Shanghai stock market as a whole, herding in B-shares is stronger than that in A-shares, while no evidence of herding is shown in US and Hong Kong markets. At the same time, the herd behavior is asymmetric in A-shares:there is herd behavior in high-volatility state periods, but during low-volatility state periods, the herding only exists when the market is going up. Moreover, herd behavior in high-volatility state is stronger than that in low-volatility state and herding in up market is stronger than in a down one, investors have a significant trend to chase after go up.