An Empirical Investigation of Chinese Stock Markets with Feed-back Models

Feed-back models in the stock markets research imply an adjustment process toward investors' expectation for current information and past experiences. The Efficient Capital Market Hypothesis, which had ignored the effect of the accumulation of information, can not explain some anomalies such as bubbles and partial predictability in the stock markets. Researchers find out that the feed-back mechanism developed by some cognitive psychologists is effective to explain some anomalies to a certain extent. In order to investigate the feed-back mechanism and to determine an effective model, we use daily data of the stock index of two Chinese stock markets. Shanghai Stock Exchange and Shenzhen Stock Exchange, from January 2007 to June 2008 with the expectational model, which is one kind of geometric lag models. Our result clearly shows the common coefficient of expectations and fourth-order autoregressive disturbance exist in the two Chinese stock markets. Furthermore, we find the scone coefficient of expectations has an autoregressive effect on disturbances in the two Chinese stock markets. Therefore the presence of such feed-back is also supported in Chinese stock markets.