Answers are grounded in this paper and cite their sources.
Faced with the multi-objective conflict problem of risk minimisation and environmental benefit maximisation in green asset allocation, the traditional single-objective optimisation method has obvious limitations. This study innovatively proposes a hybrid multi-objective optimisation algorithm (HMOA). The algorithm combines the Pareto optimisation ability of the non-dominated sorting genetic algorithm 2 (NSGA-II), the dynamic convergence of particle swarm optimisation, and the subjective viewpoint quantification mechanism of the Black–Litterman model; it introduces a displacement fluctuation strategy to enhance diversity and uses a long short-term memory (LSTM) network to dynamically generate prior information. The test results show that HMOA is significantly better than the baseline model in terms of hypervolume (HV) = 0.85, inversion generation distance = 0.12, and other indicators (such as 8.97% higher than the HV of NSGA-II). In practical verification, its Sharpe ratio reached 1.25, and its carbon dioxide (CO2) emission reduction reached 1.256 million tons of carbon dioxide. In addition, the robustness test shows that the HV change rate of the algorithm is only 5.2% under noise, which proves that the algorithm can effectively balance multi-objective conflicts. Therefore, this study provides a new method with complete theory and strong practicability for green asset allocation, and promotes the innovative application of multi-objective optimisation theory in sustainable finance.
No takes yet. Share an insight, caveat, or question.
Bei Liu (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: