Online investor sentiment in the financial futures markets

Abstract

This paper contributes to sentiment analysis in financial futures markets in two ways. First, we develop a domain-specific sentiment lexicon adapted to the language and expressions used by online futures market participants. We find that a domain-specific lexicon is essential for accurately measuring investor sentiment in futures markets. Second, we explore the relation between online investor sentiment and subsequent returns in three actively traded financial futures markets. Our results support the usefulness of the proposed sentiment index, constructed using the specific lexicon, for predicting subsequent returns over short horizons in futures markets. Sentiment timing strategies based on different lexicons were also examined, and we found that our sentiment-based lexicon market timing strategy outperforms the general financial lexicons timing strategies during the evaluation period. Overall, these findings are valuable for researchers and practitioners seeking deeper insights into investor sentiment and its impact on futures markets.