Scholarlink

<p>Environmental, social, and governance (ESG) considerations are becoming increasingly important to investors, who steer clear of businesses that engage in unethical behavior. This study aligns investors' ethical choices with portfolio optimization by including ESG evaluations. Using the Nifty 50 index, we present a framework combining financial metrics and ESG data for stock selection, return prediction, and portfolio optimization. Robust Expectation Maximization (REM) clustering segments companies based on financial and ESG indicators, while the VIKOR ("VIekriterijumsko KOmpromisno Rangiranje") MCDM (Multi-Criteria Decision Making) method identifies top-performing stocks. For return prediction, an ensemble of GRU (Gated Recurrent Unit), BiLSTM (Bidirectional Long Short-Term Memory), and XGBoost (Extreme Gradient Boosting) models is proposed. Portfolio optimization employs a hybrid Jaya-TLBO (Teaching Learning Based Optimization) metaheuristic, addressing multi-objective goals: minimizing semi-variance, maximizing the Sharpe Ratio, and promoting diversification. Empirical findings demonstrate the strategy's ability to balance risk, return, and ethical investment, offering a robust solution for socially responsible investing while achieving financial stability.</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

Leveraging ESG rating and advanced analytics for portfolio optimization

  • 期刊论文

  • 发表日期：

    2025年10月23日

查看全文
  • Veena Jain,
  • Rishi Rajan Sahay,
  • Nupur

摘要

Environmental, social, and governance (ESG) considerations are becoming increasingly important to investors, who steer clear of businesses that engage in unethical behavior. This study aligns investors' ethical choices with portfolio optimization by including ESG evaluations. Using the Nifty 50 index, we present a framework combining financial metrics and ESG data for stock selection, return prediction, and portfolio optimization. Robust Expectation Maximization (REM) clustering segments companies based on financial and ESG indicators, while the VIKOR ("VIekriterijumsko KOmpromisno Rangiranje") MCDM (Multi-Criteria Decision Making) method identifies top-performing stocks. For return prediction, an ensemble of GRU (Gated Recurrent Unit), BiLSTM (Bidirectional Long Short-Term Memory), and XGBoost (Extreme Gradient Boosting) models is proposed. Portfolio optimization employs a hybrid Jaya-TLBO (Teaching Learning Based Optimization) metaheuristic, addressing multi-objective goals: minimizing semi-variance, maximizing the Sharpe Ratio, and promoting diversification. Empirical findings demonstrate the strategy's ability to balance risk, return, and ethical investment, offering a robust solution for socially responsible investing while achieving financial stability.