An Augmented Fama and French Three-Factor Model Using Social Interaction

We propose that social interaction is a potential risk factor in the asset pricing model of the stock market. "Panorama net" is a website for listed companies and investors to interact in China, which the data extracted from are used. Attention, talkativeness and warmness are constructed from them as a measurement of social interaction. We test whether the social interaction variables provide additional information in the asset pricing model proposed by Fama and French. Empirical results suggest that the social interaction is capable to influence stock return, but it performs differently across industries. In most industries which people are familiar with, such as health care and machinery, social interaction provides additional information; while in most stocks in industries far from daily life, it doesn't help to explain stock return. The results reveals that the social interaction is a potential risk factor in asset pricing model. It also helps listed companies and regulators to utilize interaction website in a better way.