開示トーンと機械学習を統合した財務業績予測
Integrating Disclosure Tone and Machine Learning for Financial Performance Prediction (原題)
Imad Ud Din Durrani, Hassan Raza
🤖 gxceed AI 要約
日本語
本研究は、パキスタンの非金融企業を対象に、開示トーン（ポジティブ、ネガティブ、不確実性）を機械学習モデルに統合し、財務業績予測を向上させる。年次報告書からテキストデータを抽出し、財務トーン辞書で分析。SVM、ANN、RF、LSTMを比較し、RFが最高精度（EPS:0.82, ROA:0.82, ROE:0.85）を達成。定性情報と定量データの統合が予測精度を高めることを示した。
English
This study integrates disclosure tone (positivity, negativity, uncertainty) into machine learning models to enhance financial performance prediction for non-financial firms in Pakistan. Textual data from annual reports were analyzed using a financial tone lexicon, and combined with financial metrics to train SVM, ANN, RF, and LSTM models. Random Forest achieved the highest accuracy (EPS: 0.82, ROA: 0.82, ROE: 0.85), demonstrating that integrating qualitative tone with quantitative data improves forecasting. The hybrid approach offers insights for investors, analysts, and policymakers in emerging markets.
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開示が進む中、定性情報の定量化は有報や統合報告書の分析に応用可能。AIを活用した開示分析は投資家対応やESG評価に有用で、日本企業の開示品質向上に示唆を与える。
In the global GX context
Globally, this research aligns with the growing trend of using AI to analyze corporate disclosures for ESG and financial performance. It contributes to the literature on disclosure tone and machine learning, offering a methodology that can be applied to TCFD/ISSB-aligned reports to extract insights for investors and regulators.
👥 読者別の含意
🔬研究者:Provides a novel hybrid approach combining tone analysis with ML for financial prediction, useful for further research in disclosure analytics.
🏢実務担当者:Offers a practical method for integrating qualitative disclosure tone into financial forecasting, potentially enhancing investment analysis and risk assessment.
🏛政策担当者:Highlights the value of qualitative disclosure in emerging markets, suggesting that policies encouraging transparent reporting could improve market efficiency.
📄 Abstract（原文）
This study aims to enhance financial performance forecasting by integrating disclosure tone variables—positivity, negativity, and uncertainty—into machine learning models for non-financial firms in Pakistan. Existing financial forecasting models often rely solely on quantitative metrics, neglecting the qualitative aspects of corporate disclosures. This study addresses this gap by incorporating tone variables alongside key financial indicators such as ROA, ROE, and EPS. Textual data were extracted from annual reports and analyzed using a financial tone lexicon. After pre-processing, tone scores were combined with financial metrics to train predictive models using Support Vector Machine (SVM), Artificial Neural Networks (ANN), Random Forest (RF), and Long Short-Term Memory (LSTM). Random Forest consistently outperformed other models, achieving the highest accuracy (EPS: 0.82, ROA: 0.82, ROE: 0.85), while SVM and ANN showed moderate performance, and LSTM performed comparatively lower (EPS: 0.64). This research introduces a novel hybrid forecasting approach that integrates qualitative tone analysis with quantitative data, addressing limitations in traditional models. It offers practical insights for investors, analysts, and policymakers operating in emerging markets marked by information asymmetry, and contributes to advancements in financial technology and corporate disclosure practices.
🔗 Provenance — このレコードを発見したソース
- semanticscholar https://nijbm.numl.edu.pk/index.php/BM/article/download/255/124first seen 2026-08-27 05:49:00 · last seen 2026-09-21 05:21:16
🔔 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。