J Syst Sci Complex (2012) 25: 529–548 DEPENDENCE BETWEEN STOCK RETURNS AND INVESTOR SENTIMENT IN CHINESE MARKETS: A COPULA APPROACH ∗ Xunfa LU · Kin Keung LAI · Liang LIANG DOI: 10.1007/s11424-012-9332-0 Received: 19 November 2009 / Revised: 5 January 2011 c The Editorial Office of JSSC & Springer-Verlag Berlin Heidelberg 2012 Abstract Using data of newly opened stock trading accounts in China as a proxy of investor sentiment index, the authors employ the time-varying copula-GARCH model with Hansen’s skewed Student-t innovations to investigate the dynamic dependence between investor sentiment and stock returns. The empirical findings show that shifts in investor sentiment are asymptotically positively correlated to stock returns in extreme value situations in both A shares market and B shares market in China, that is to say, stock prices will increase (decrease) more when investors become more bullish (bearish). Also, results show that the dependence between investor sentiment and stock returns is time-varying, which means that the traditional Pearson’s correlation based on normal distribution is not enough to describe the relationship between stock market behavior and investor behavior. Key words Behavioral finance, copula, GARCH, investor sentiment, newly opened stock trading accounts. 1 Introduction More and more attention is being paid to the investor sentiment theory by researchers working on behaviors of market participants in the financial industry. The classical finance theory supposes that investors are completely rational, resulting in researchers ignoring the ex- istence of noise traders and their role in determining market movement. Based on this theory, many elegant financial models have been constructed, such as the capital asset pricing model (CAPM) [1-4] , the arbitrage pricing theory (APT) [5] , the efficient-market hypothesis (EMH) [6] , and so on. However, the increase in complexity of behavior of financial markets from the 1980s onwards has challenged the classical finance theory in several ways. Consequently, the behav- ioral finance theory has emerged, at least in part, in response to distortions in the traditional Xunfa LU Department of Management Sciences, City University of Hong Kong, Hong Kong, China; School of Business, University of Science and Technology of China, Hefei 230026, China. Email: xflu@mail.ustc.edu.cn. Kin Keung LAI Department of Management Sciences, City University of Hong Kong, Hong Kong, China. Email: mskklai@cityu.edu.hk. Liang LIANG School of Business, University of Science and Technology of China, Hefei 230026, China. Email: lliang@ustc.edu.cn. ∗ This research is supported by the National Natural Science Foundation of China under Grant No. 70821001. ⋄ This paper was recommended for publication by Editor Shouyang WANG.