Interpretable predictive model for listed companies ESG greenwashing based on XGBoost and SHAP
XGBoostとSHAPに基づく上場企業のESGグリーンウォッシングの解釈可能な予測モデル (AI 翻訳)
Jianfeng Zhang, Tiantian Qi
🤖 gxceed AI 要約
日本語
本研究は、中国上場企業の2009～2022年のデータを用い、XGBoostとSHAPを組み合わせたESGグリーンウォッシング予測モデルを提案。16指標（企業特性・外部圧力）を入力とし、5-fold交差検証とグリッドサーチで最適化。1期ラグモデルで86.82%の予測精度を達成し、財務特性がガバナンス特性より重要。非重汚染産業や国有企業で性能が高い。
English
This study proposes an interpretable predictive model for ESG greenwashing in listed companies using XGBoost and SHAP, based on Chinese listed companies data from 2009-2022 with 16 input variables. The one-period-lagged model achieves 86.82% accuracy, revealing that corporate financial characteristics have greater impact than governance. The model performs better in non-heavy-pollution industries and state-owned enterprises.
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SSBJ開示基準の導入に伴い、グリーンウォッシング監視の重要性が増している。本モデルは日本企業の開示データに応用可能で、投資家保護や規制当局の早期警戒に資する可能性がある。
In the global GX context
Globally, greenwashing detection is critical for trust in ESG disclosures. This AI-driven model offers a transparent, interpretable approach that could be adapted for regulatory frameworks like the EU's CSRD or SEC climate rules, enhancing ex-ante risk management.
👥 読者別の含意
🔬研究者:Provides a robust, interpretable ML framework for greenwashing prediction with feature importance analysis.
🏢実務担当者:Offers a tool to assess greenwashing risk in portfolio companies or internal disclosures.
🏛政策担当者:Demonstrates feasibility of automated greenwashing detection for regulatory oversight.
📄 Abstract（原文）
The prediction of ESG greenwashing among listed companies is crucial for ex-ante control of deceptive ESG information disclosures by companies and for early warning of ESG investment risks. Machine learning techniques are commonly used to predict corporate behavior, but their application in ESG greenwashing is limited and lacks interpretability. This study proposes a predictive model for ESG greenwashing in listed companies using the enhanced XGBoost algorithm and SHAP interpretation method. Using a dataset of Chinese listed companies from 2009 to 2022, this study selects 16 indicators of corporate and external pressure characteristics as the model’s input variables, combines five-fold cross-validation and grid search parameter tuning, and constructs two company ESG greenwashing prediction models. The prediction performance is compared with Random Forest, SVM, LightGBM, BP neural network, and three XGBoost methods. The SHAP method is used to explain the contribution of the main indicators in predicting the company’s ESG greenwashing. The results show that the one-period-lagged corporate characteristics model achieves 86.82% prediction accuracy, and corporate financial characteristics have a greater impact on prediction results than corporate governance characteristics. Further analysis shows that the one-period-lagged corporate characteristics model performs better in non-heavy-pollution industries and state-owned enterprises. Supplementary Information The online version contains supplementary material available at 10.1038/s41598-026-42004-1.
🔗 Provenance — このレコードを発見したソース
- semanticscholar https://doi.org/10.1038/s41598-026-42004-1first seen 2026-07-18 08:23:22
🔔 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。