The Chinese News Sentiment around Earnings Announcements

We examine the effect of Chinese news on announcement drift and investigate its application to portfolio management, applying a linguistic analysis to extract various dimensions of the information content. Our empirical results reveal a positive (negative) relationship between news sentiment and cumulative abnormal returns in the pre- (post-) earnings announcement period, thereby confirming that the market response takes into consideration all relevant information on the related firm. The application of public news sentiment to portfolio management indicates that long (short) stocks with low (high) news sentiment and high public news surprises will earn positive excess returns. We suggest that the relevant news of individual stocks could be applied to the prediction of abnormal returns and portfolio management.