The Improvement Of Three - Factor Model And Its Empirical Study In Shenzhen Stock Market

In1992, Fama and French published the famous article--The cross section of expected stock returns, in which through studying the stock price movement of the U.S. stock market between1929and1963, they found out that there was no positive linear relationship between market risk β and stock return; suggesting a constant β, then the stock return was negatively related to the size of the company; suggesting a constant size, then stock return was irrelevant to its β, but being positively related to its value.The Chinese stock market is completely different from the Western stock market concerning its history of development and market environment, so could the financial theory which based on western free market be applicable to the reality of China? Chinese scholars have done some empirical research on this, but still they didn’t have a consistent answer.In recent years, Chinese capital market has made a lot of great achievements, which also profoundly affect the operation of the whole stock system. With this background, this paper studies the applicability of FF model on Shenzhen stock exchange. The empirical results show that size effects and value effects do exist, that the average yield of the stock is negatively related to the size of the company, positively related to the carrying amount of the company divided by its market value.Considering the fact that outstanding shares still co-exist with non-tradable shares in Chinese stock market, while the proportion of outstanding shares to some extent reflects the structure of property rights and economic efficiency, so the proportion of outstanding shares of listed companies may have explanatory power for stock returns.This paper altered Standard FF three-factor model by introducing a new factor, the proportion of outstanding shares. The empirical results show that the proportion of outstanding shares does have explanatory power for the returns of stocks listed in Shenzhen market, and the altered model improved the possibility of accepting null hypothesis.