Forecasting carbon market with large language models

Bangzhu Zhu et al.

Managerial Finance2026https://doi.org/10.1108/mf-08-2025-0632article
AJG 1ABDC B
Weight
0.50

What the paper says

Purpose This paper aims to develop an accurate carbon price forecasting framework using large language models (LLMs) to overcome limitations of conventional structured data methods. Design/methodology/approach This paper details the comprehensive LLM framework that integrates historical prices with market drivers and news texts through advanced prompt engineering, utilizing zero-shot inference with the Chinese carbon market as the case study. Findings This paper highlights the superior performance achievements, including enhanced accuracy in price and directional predictions, improved forecasting capabilities through news integration and increased market transparency. Originality/value This paper emphasizes the novel contributions, specifically the first comprehensive LLM framework for carbon market forecasting that combines structured and unstructured data and the innovative zero-shot inference methodology that eliminates the need for prior training examples.

Open paper page →

Cite this paper

https://doi.org/https://doi.org/10.1108/mf-08-2025-0632

Or copy a formatted citation

@article{bangzhu2026,
  title        = {{Forecasting carbon market with large language models}},
  author       = {Bangzhu Zhu et al.},
  journal      = {Managerial Finance},
  year         = {2026},
  doi          = {https://doi.org/https://doi.org/10.1108/mf-08-2025-0632},
}

Paste directly into BibTeX, Zotero, or your reference manager.

Flag this paper

Forecasting carbon market with large language models

Flags are reviewed by the Arbiter methodology team within 5 business days.


Evidence weight

0.50

Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40

F · citation impact0.50 × 0.4 = 0.20
M · momentum0.50 × 0.15 = 0.07
V · venue signal0.50 × 0.05 = 0.03
R · text relevance †0.50 × 0.4 = 0.20

† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.