Enhancing Investment Decisions with Sentiment Analysis: A Probabilistic Ranking Framework

• 成果类型: Article; Early Access
• 署名作者: Ke, Zheng Tracy; Kelly, Bryan; Xiu, Dacheng
• 署名单位: Harvard University; Yale University; National Bureau of Economic Research; University of Chicago
• 刊物名称: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
• ISSN/ISSBN: 0162-1459; 1537-274X
• DOI: 10.1080/01621459.2026.2643001
• 发表日期: 2026-05-27
• 关键词: Machine learning; PENALIZED LIKELIHOOD; Return predictability; screening; sentiment analysis; Text Mining; topic modeling

摘要

We develop a probabilistic framework to extract sentiment information from text by training a model to predict and rank sentiments in newly encountered documents. Our approach imposes a joint semi-parametric model on text and ordinal response variables, addressing the challenges of sparse sentiment signals and complex response distributions. Through a word screening procedure and the use of normalized ranks, our approach achieves consistent sentiment ranking without estimating the full model. Applying our method to the Dow Jones Newswires, we demonstrate its effectiveness in extracting return-predictive signals. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.