Affiliation:
1. School of Economics and Management Southwest Jiaotong University Chengdu China
2. Service Science and Innovation Key Laboratory of Sichuan Province Chengdu China
Abstract
AbstractLeveraging deep learning to analyse COVID‐19 image sentiment, this study reveals its significant impact on stock market dynamics. It highlights how vivid imagery prompts marked emotional responses, altering market performance and how news sentiment can modulate this effect. Further, it underscores the pivotal role of forum‐based investor sentiment, particularly affecting small‐minus‐big stocks during downturns and trading week commencements. This research not only advances behavioural finance understanding but also informs management and regulatory strategies.
Funder
National Natural Science Foundation of China