<p>This study examines how environment, social, and governance (ESG) related news and investor sentiment (IS) affect cryptocurrency prices, with a focus on major digital assets like Bitcoin. Using machine learning techniques, this study analyzes daily data between January 2021 and February 2024 to understand the role of sentiment in this highly speculative market. The findings show that both ESG news and investor sentiment significantly influence cryptocurrency returns. Furthermore, the data revealed that negative sentiments have a stronger effect on cryptocurrency prices, suggesting a loss aversion tendency among investors. While investor sentiment has a greater impact on cryptocurrency returns than equity returns, ESG news is more influential in equity markets. To test predictive performance, a Long Short-Term Memory (LSTM) model was used and found to outperform regression based statistical model. This research offers new insights into how non-financial information shapes digital asset pricing. The results have practical implications for traders, portfolio managers, and policymakers.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Analyzing the Impact of ESG News and Investor Sentiment on Cryptocurrency Markets Using Machine Learning

  • Sougata Banerjee

摘要

This study examines how environment, social, and governance (ESG) related news and investor sentiment (IS) affect cryptocurrency prices, with a focus on major digital assets like Bitcoin. Using machine learning techniques, this study analyzes daily data between January 2021 and February 2024 to understand the role of sentiment in this highly speculative market. The findings show that both ESG news and investor sentiment significantly influence cryptocurrency returns. Furthermore, the data revealed that negative sentiments have a stronger effect on cryptocurrency prices, suggesting a loss aversion tendency among investors. While investor sentiment has a greater impact on cryptocurrency returns than equity returns, ESG news is more influential in equity markets. To test predictive performance, a Long Short-Term Memory (LSTM) model was used and found to outperform regression based statistical model. This research offers new insights into how non-financial information shapes digital asset pricing. The results have practical implications for traders, portfolio managers, and policymakers.