Is Positive Sentiment in Corporate Annual Reports Informative? Evidence from Deep Learning

Abstract

We use a novel text classification approach from deep learning to more accurately measure sentiment in a large sample of 10-Ks. In contrast to most prior literature, we find that positive, and negative, sentiment predicts abnormal return and abnormal trading volume around 10-K filing date and future firm fundamentals and policies. Our results suggest that the qualitative information contained in corporate annual reports is richer than previously found. Both positive and negative sentiments are informative when measured accurately, but they do not have symmetric implications, suggesting that a net sentiment measure advocated by prior studies would be less informative.

Keywords: Corporate annual reports, 10-K filings, Textual analysis, Textual sentiment classification, Deep learning

JEL Classification: C81, D83, G10, G14, G30, M41