Abstract: We link public media information with private information of institutional investors and build an IPO pricing model based on Bayes’ rule under the background of Chinese IPO market. Our model gives a theoretical explanation for the micro mechanism of media information influencing institutional investors’ private information and decision-making. The results of the model show that: media information and private information of institutional investors are positively correlated, and the distribution of private information will depend on the implementation value of the media signal. Media information influences institutional investors’ decisions and their expected initial returns through two ways — incentive effect and demand effect. These two paths have negative and positive effects on the expected initial returns respectively, and the final impact of media information will depend on the relative strength of the two paths. We use 874 A-share IPO samples from 2009 to 2014 to conduct empirical tests. The empirical results respectively verify the two paths and their combined effects, which provides satisfactory support for the model conclusion.

Key words: media information, institutional investor, private information, IPO initial returns, Bayes’ rule