Smart Forecasting of Carbon Prices Using Machine Learning and Neural Networks: When ARIMA Meets XGBoost and LSTM
機械学習とニューラルネットワークを用いた炭素価格のスマート予測：ARIMAとXGBoost、LSTMの比較 (AI 翻訳)
Giorgos Kotsompolis, Panagiotis Cheilas, Konstantinos N. Konstantakis, Evangelos Sfakianakis, Stephane Goutte, Panayotis G. Michaelides
🤖 gxceed AI 要約
日本語
本研究は、炭素価格予測において、従来のARIMAモデルと機械学習（XGBoost）・深層学習（LSTM）の性能を比較。2010年12月から2025年1月までの日次データを用いた結果、XGBoostとLSTMはARIMAを上回り、両者のハイブリッドモデルが最も高い予測精度を示した。炭素市場参加者や政策立案者にとって有用な知見を提供する。
English
This study compares ARIMA, XGBoost, LSTM, and a hybrid model for carbon price forecasting using daily data from December 2010 to January 2025. Results show that XGBoost and LSTM outperform ARIMA, and the hybrid model achieves the highest accuracy. The findings support machine learning and neural network approaches as effective tools for carbon pricing prediction, relevant for ETS participants and policymakers.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
日本では2023年度から東証カーボン・クレジット市場が始動し、炭素価格予測の需要が高まっている。本論文の手法は、日本の排出量取引制度やGXリーグ参加企業の価格リスク管理に応用可能。ただし、データがEU ETSベースであれば、日本の市場特性に合わせた調整が必要。
In the global GX context
Accurate carbon price forecasting is critical for global ETS design and transition finance. This paper demonstrates that machine learning and neural network models can significantly improve prediction accuracy over traditional econometric models, offering practical tools for investors and regulators in carbon markets worldwide. The hybrid approach is particularly relevant for jurisdictions like the EU, China, and emerging markets.
👥 読者別の含意
🔬研究者:Provides a benchmark comparison of ARIMA, XGBoost, LSTM, and hybrid models for carbon price forecasting, with implications for model selection and ensemble methods.
🏢実務担当者:Offers actionable forecasting tools for carbon traders and corporate sustainability teams to manage price risk and optimize compliance strategies.
🏛政策担当者:Highlights the potential of advanced forecasting to improve market oversight and policy evaluation in emissions trading schemes.
📄 Abstract（原文）
ABSTRACT Accurate prediction of carbon prices is crucial for policymakers, investors, and other participants in emissions trading schemes (ETS) and during regulatory transitions. In this work, carbon price movements are forecasted using a nonlinear ARIMA model as the baseline, alongside XGBoost and LSTM as competing models. The widely adopted XGBoost model is a machine learning (ML) technique, while the LSTM model belongs to the class of Recurrent Neural Network (RNN) models. To harness the predictive strengths of both approaches, we also employ a hybrid model that averages forecasts from the LSTM and XGBoost models. The dataset used in this study is in daily format, ranging from December 1, 2010, to January 10, 2025. The results show that both XGBoost and LSTM outperform the baseline ARIMA model. Furthermore, the hybrid model demonstrates statistically significant improvements in forecasting accuracy compared to the baseline model. These findings suggest that ML‐ and RNN‐based approaches can serve as effective alternatives to traditional statistical and econometric models in carbon pricing forecasting.
🔗 Provenance — このレコードを発見したソース
- openaire https://doi.org/10.1002/for.70025first seen 2026-05-05 19:06:49
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。