Test And Selection Of Capital Asset Pricing Model In China Stock Market

Asset pricing is one of the most fundamental issues in financial economics. It has undoubtedly played an important role in security returns forecasting and investment decision making. This thesis asks the question what kind of asset pricing model works in emerging markets. In particular, this thesis will study the case of China.Based on the analysis, in the period of July 1995 to December 2004, the Chinese stock markets demonstrate (a) size effect, (b) book-to-market equity effect, (c) earnings-price ratio effect, (d) dividend yield effect, and (e) past stock price effect. Then, this thesis testifies that the Fama-French three-factor model can capture the cross-sectional variation in returns of portfolios formed on those five variables mentioned above for the 1995-2004 period, while the Carhart four-factor model does not. Finally, I construct a new four-factor model by substituting one of the existing factors with a new risk factor. This new model works well and even better than the Fama-French three factor model does for certain portfolios.Therefore, the new four-factor model can be used by investors who are interested in Chinese stock markets to select asset portfolios and evaluate their performance.