Investor sentiment and its nonlinear effect on stock returns¬タヤNew evidence from the Chinese stock market based on panel quantile regression model

a r t i c l e i n f o Keywords: Investor sentiment Stock returns Chinese A-share stock market Firm characteristics Penalized panel quantile regression model This paper employs the panel quantile regression model to study the nonlinear effect of investor sentiment on monthly stock returns in the Chinese A-share stock market. The findings show that the influence of investor sentiment is significant from 1 month to 24 months. Its effect is asymmetric and reversal, that is, it is positive and large for stocks with high returns in the short term while negative and small in the long term. This reversal effect verifies the existence of a strong overreaction in the Chinese stock market. We also find that Chinese investors have notable cognitive bias and speculation tendency. The past twenty years have witnessed several huge crashes in global stock markets. The crash of Hang Seng Index in 1997 led to 55.55% loss in the East Asian Financial Crisis. The crash of the Taiwan stock market resulted in 64.53% loss in 2001 of its market value from previous year. The crash of the Nikkei index in 2003 lost 78.9% of the market value in the 1989. The Internet Bubble crash in the 2000 resulted in a loss of about 70.25% of its value in 2002. The Shanghai Composite Index also suffered an 80.99% loss in 2008. These huge crashes in the global stock markets can be summarized in two features. First, unexpected slumps in the stock market can not be explained by public market information and macroeconomic conditions. Second, fluctuation of the stock market is unbalanced and asymmetric since market crash could happen in a short time while the rise to original stock price needs to take a long time. We think that these abnormal phenomena in the stock market are very valuable and meaningful to research since they can not be explained by mainstream classical financial theories, including the efficient-market hypothesis (EMH) (Fama, 1970), and asset pricing models, such as the CAPM model (Sharpe, 1964), the macroeconomic factor model (Chen et al., 1986) and the three-factor model (Fama and French, 1993), etc. On the other hand, some scholars think that investor sentiment could explain the phenomena mentioned above, such as They think that investor sentiment is crucial to affect stock returns because of limits of arbitrage, unexpected demand shock, unbalanced specific reference and utility functions in gain and loss. …