AI-Driven Framework for Sustainable Investments in Indian Stock Market Offers Superior Performance

Strategic Insights -

A recent study on sustainability research has yielded promising results, offering a three-stage AI-driven framework to effectively manage Environmental, Social, and Governance (ESG) portfolios while maximizing predictive accuracy and investment efficiency. The research aims to address the limitations of existing frameworks in integrating artificial intelligence (AI) to optimize ESG portfolios, particularly in the Indian stock market. The study proposes a novel approach using advanced machine learning and optimization techniques to enhance return prediction accuracy, optimize portfolio allocation, and perform dynamic portfolio rebalancing.

Key Takeaways:

  • The research proposes a three-stage AI-driven framework to manage ESG portfolios, comprising Multivariate Bidirectional Long Short-Term Memory (MBi-LSTM) network for return prediction, Non-dominated Sorting Genetic (NSG) algorithm for portfolio allocation, and Actor-Critic Reinforcement Learning (RL) algorithm for dynamic portfolio rebalancing.
  • The framework demonstrated superior performance in predictive modeling, portfolio allocation, and portfolio rebalancing, outperforming traditional methods in terms of return, risk, and ESG-Sortino Ratio.
  • The research showed that the MBi-LSTM network achieved the lowest errors in all parameters, while the NSG-enabled portfolio allocation method achieved a 68.58% higher similarity to actual returns than the Mean-Variance Markowitz (MVM) model.
  • The framework outperformed traditional agent-based methods in portfolio rebalancing, with returns higher by 120%, 235%, and 101% in average, maximum, and minimum returns, respectively.
  • The study highlights the effectiveness of the demonstrated Multi-Objective AI-driven framework in validating against conventional evaluation metrics, showcasing its superior performance and practical viability for sustainable and Socially Responsible Investments (SRIs) in the Indian stock market.
  • The research is led by Apurv Gaurav, Mechanical Engineering Department, Indian Institute of Technology Kharagpur, with additional authors Kripamay Baishnab and Piyush Kumar Singh.

Statistics:

  • 68.58% higher similarity to actual returns achieved by the NSG-enabled portfolio allocation method compared to the Mean-Variance Markowitz (MVM) model.
  • 120%, 235%, and 101% higher returns achieved by the Actor-Critic Reinforcement Learning (RL) algorithm in average, maximum, and minimum returns, respectively, compared to traditional agent-based methods.

Sources:

  • Gaurav, A., Baishnab, K., & Singh, P. K. (2025). Intelligent ESG portfolio optimization: A multi-objective AI-driven framework for sustainable investments in the Indian stock market. Sustainable Futures, 9, 100832.
  • NewsRx. Research from Indian Institute of Technology Kharagpur Reveals New Findings on Sustainability Research (Intelligent ESG portfolio optimization: A multi-objective AI-driven framework for sustainable investments in the Indian stock market). Ecology, Environment & Conservation. July 4, 2025; p 682.