Affiliation:
1. College of Finance and Statistics Hunan University Changsha China
2. Guilin Bank Guilin China
3. School of Economics and Management Guangxi Normal University Guilin China
Abstract
AbstractThis paper introduces an innovative methodology for extracting information from textual data to explain cross‐sectional stock returns, addressing limitations of conventional media tone measures. We find firms exhibiting higher media abnormal tone yield lower future returns in the Chinese market, even when controlling for common risk factors. This effect is more pronounced among firms with low investment, low profitability, and high short‐term reversal. We also find the negative premium generated by media abnormal tone results from mispricing, highlighting investor overreaction despite media's role in disseminating concurrent firm information. Furthermore, the tendency for media outlets to follow suit exacerbates investor overreaction.
Funder
National Natural Science Foundation of China