Impact of International Environmental Emergency on Stock Market Dynamics: A RoBERTa Model and Pearson Correlation Approach

Understanding market reactions to environmental pollution crises is essential for assessing investor sentiment and financial stability. While prior studies have explored general market responses, the role of investor psychology during such crises remains underexplored. To bridge this gap, we propose...

Full description

Saved in:
Bibliographic Details
Main Authors: Wenxin Tang, Qingyang Liu, Yanrong Hu, Hongjiu Liu
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11048581/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Understanding market reactions to environmental pollution crises is essential for assessing investor sentiment and financial stability. While prior studies have explored general market responses, the role of investor psychology during such crises remains underexplored. To bridge this gap, we propose a sentiment analysis framework based on the RoBERTa-WWM-ext-large model combined with Pearson correlation analysis. This approach investigates the relationship between environmental incidents, investor sentiment, and stock price dynamics. Experiments on real-world stock data reveal that sudden environmental events intensify the correlation between sentiment shifts and market volatility, often with a one-day temporal lag. Further analysis shows that different sentiment categories display varying levels of association with trading metrics—for instance, fear sentiment exhibits a strong positive Pearson correlation (up to 0.66) with transaction volume in one stock after the event. Moreover, industries such as aquaculture and salt demonstrate notably higher sensitivity to sentiment fluctuations. These findings offer practical insights for financial practitioners and policymakers seeking to understand sentiment-driven behavior in environmentally induced market disruptions.
ISSN:2169-3536