BL-LSTM approach for sustainable portfolio optimization: A water market case study.

Purpose: This study aims to introduce an innovative investment methodology that synergistically combines the Black-Litterman (BL) approach with Long Short-Term Memory Neural Networks (LSTM) and ARIMA, and applies it to the water market. With a specific focus on contributing to Sustainable Development Goal (SDG) number 6, 'Ensure access to water and sanitation for all,' our objective is to enhance decision-making in the water market by integrating responsible water management principles. Methodology: The methodology involves integrating LSTM and ARIMA predictions as views in the BL model to construct portfolios for the water market. This unique combination leverages historical data and advanced predictive techniques to enhance investor decision-making. Results: The resulting portfolio outperforms traditional mean-variance models and water ETF benchmarks. The combined use of LSTM and ARIMA with the BL framework effectively generates returns that surpass industry standards. Practical implications: Investors engaging in the water market using this methodology can make informed decisions, contributing to sustainable water resource management. The approach aligns with SDG number 6, promoting responsible and sustainable investment practices. Originality: This study contributes to the field by introducing a novel approach that synthesizes the BL framework with LSTM and ARIMA, offering investors a powerful tool for more insightful decision-making in the water market. [ABSTRACT FROM AUTHOR]