Empirical Research On Herding Effect Of A

Herd effect is an important research scope in behavioral finance, which affects the stability, effectiveness and regularity of financial market to some extent. Also it is feasible to measure the above market situations from the point of herd effect, providing a theoretical basis for the level of market maturity and efficiency.This article describes the research backgrounds of herd effect. Related theories are introduced to clarify the definition, classification and characteristics of herd effect, models and methods are discussed on generating proxy variables and testing herd effect, so empirical methods built on different herd effect measures are explained in detail.The author uses CSAD method here to measure the herd effect. By using empirical model on CSI 300 constituent stocks’ price data from 1999 to 2004, excluding the market fluctuation period within 2007 and 2008, I divide the research period into two separate parts, from 1999 to 2006 and from 2009 to 2014. The author then find an obvious herd effect in China stock market, no matter whether the market goes up or falls down, whether runs smoothly or fluctuates sharply, and the herd effect becomes more significant when the price rises. That means investors in A-share intend to "buy" rather than "sell". Compared to 1999-2006, 2008-2014 seems to have more smooth market volatility, as the herd effect becomes significantly weaken. In the meantime, I also compare the Ashare to H-share, the US stock market and other developed capital markets, then find no obvious existence of herd behavior in such mature markets. In summary, compared to the developed capital markets, China’s A-share market is still having obvious herd behavior, with much room for improvement.