Graduation Year
2026
Date of Submission
4-2026
Document Type
Open Access Senior Thesis
Degree Name
Bachelor of Arts
Department
Economics
Reader 1
Peter Kelly
Rights Information
Yiqi Sun
Abstract
Stress is pervasive among corporate executives, yet its effect on firms' future abnormal return remains largely unexplored. This paper examines whether CEO's Linguistic stress exhibited during earnings conference calls predicts future abnormal stock returns. I provided a Large Language Model (Claude Sonnet 4.6) psychology-backed prompts to analyze CEO stress from linguistic patterns in 7491 earnings call transcripts from 470 companies in the S&P 500 during 2020-2024. I found three main results. First, the CEO linguistic stress is negatively and significantly associated with the immediate market reaction, with a one-unit increase in stress score associated with a 0.9% decrease in the two-day cumulative abnormal return around the earnings call date. Second, stress during the prepared remarks is a stronger predictor of the immediate market reaction than stress during the Q&A session, suggesting investors may view stress from scripted remarks as more informative about firms' future prospects. Third, stress scores have no predictive power for abnormal returns over 2--7 or 2--180 trading days following the call, indicating that the market fully and immediately incorporates the stress signal with no subsequent drift. These findings are consistent with the semi-strong form of market efficiency and suggest that Large Language Models can extract psychologically meaningful signals from corporate disclosures that are rapidly priced by investors.
Recommended Citation
Sun, Yiqi, "CEO Linguistic Stress and Abnormal Stock Returns Evidence from S&P 500 Earnings Conference Calls" (2026). CMC Senior Theses. 4150.
https://scholarship.claremont.edu/cmc_theses/4150
Included in
Behavioral Economics Commons, Business and Corporate Communications Commons, Corporate Finance Commons
