Interpretable ESG–sentiment hybrid deep learning for asset return forecasting with quantified interactions and latency-aware deployment

In this study, we developed a new artificial intelligence (AI) model that combines ESG data with financial news sentiment to forecast asset returns.

Published in Economics

Like

Share this post

Choose a social network to share with, or copy the URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

💡 What is the study about?
Financial markets are influenced not only by numbers and economic data, but also by sustainability information (ESG) and investor sentiment from news. However, it is still not well understood how these two forces interact. In this study, we developed a new artificial intelligence (AI) model that combines ESG data with financial news sentiment to forecast asset returns.

📊 Key findings
• The model achieves 94.5% directional accuracy in predicting market movements.
• It performs better than several advanced machine learning and large language model benchmarks.
• We find clear interactions between ESG signals and investor sentiment.
• Investor sentiment becomes more important during market turbulence, while ESG factors matter more in stable periods.
• The approach also produces better risk-adjusted investment performance and lower drawdowns.

🌍 Overall, the study shows that sustainability information and investor psychology together shape financial markets, and that interpretable AI can help investors make better decisions.

Reference (Open Access) :
Mishra, S., Mayaluri, Z. L., Liew, C. Y., Sahoo, P. K., & Samantaray, A. K. (2026). Interpretable ESG–sentiment hybrid deep learning for asset return forecasting with quantified interactions and latency-aware deployment. Scientific Reports. doi:10.1038/s41598-026-41985-3

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in