The Role of Machine Learning in Predictive Trading : A Comparative Analysis of AI Trading Bots in the Context of Environmental Responsibility

This study explores the integration of environmental metrics within AI-driven trading strategies, focusing on their impact on long short-term memory (LSTM) neural networks. Recent advancements in artificial intelligence (AI) have revolutionised various sectors, including finance, where AI algorithms significantly enhance predictive trading. Despite the increasing emphasis on sustainable investment practices, the effectiveness of incorporating ESG scores into AI trading models remains largely underexplored. The primary objective of this research is to determine whether the inclusion of environmental scores can enhance the performance of LSTM predictive trading models. The study employs a comparative analysis using historical data from ten large US companies across diverse industries, integrating ESG scores and CO2 reduction goals into the LSTM algorithm. The results indicate that contrary to expectations, environmental scores do not positively impact the performance of LSTM training. In some instances, including these factors even hindered the model's ability to predict market trends accurately.