The Impact Of Industry Factor On Characteristics Of Stock Returns And Its Asset Pricing Research

Firstly, this paper investigates the impact of industry classification on characteristics of stock returns. We test whether size, BM, and momentum characteristics exist the industry effect and whether the impact of industry classification on characteristics of stock returns exist monthly effect by using the panel regression method and theoretical model in A share market from January 2008 to December 2011. The finding show industry classification affects BM effect in March, except which, industry classification has no effect on size, BM and momentum effects; Besides, the impact of industry classification on BM effect exists “March effect”, however, when we see all the months as a whole, industry classification dose not yet affect BM effect.So, we can preliminarily conclude that size, BM and momentum effects have nothing to do with industry classification, characteristics of stock returns do not exist industry effect, that is, industry classification do not affect size, BM and momentum effects. The research results provide preliminary premise for studying the ability of industry factors to involving in pricing.Secondly, in order to further analyze the relations between industry and the characteristics of stock returns, the paper considers whether the premiums exhibit any asymmetric patterns for firms that rank above and below their industry averages on various firm characteristics by using the panel regression method in A share market from January 2008 to December 2011. The results show that size and BM premiums are symmetric for firms that rank above and below their industry averages on size and BM characteristics, while momentum premium exhibits asymmetric for firms that rank above and below their industry averages on momentum characteristic in China's stock market.Finally, we analyze the influence of industry factors on stock returns and investigate the ability of industry factors to involving in pricing by using the panel regression method and principal component analysis in A share market from January 2007 to December 2011, which respectively based on Carhart four-factor model and Fama-French three-factor model. The results show that industry factors affect stock returns, the extracted common factors based on Carhart four-factor model do not exhibit significant risk premiums, while the extracted industry factors based on Fama-French three-factor model exist significant risk premiums. So, we finally establish multi-factor asset pricing model based on Fama-French three-factor model that contains industry common factors.