Textual Analysis

Extracting the Structure of Press Releases for Predicting Earnings Announcement Returns ArXiv ID: 2509.24254 “View on arXiv” Authors: Yuntao Wu, Ege Mert Akin, Charles Martineau, Vincent Grégoire, Andreas Veneris Abstract We examine how textual features in earnings press releases predict stock returns on earnings announcement days. Using over 138,000 press releases from 2005 to 2023, we compare traditional bag-of-words and BERT-based embeddings. We find that press release content (soft information) is as informative as earnings surprise (hard information), with FinBERT yielding the highest predictive power. Combining models enhances explanatory strength and interpretability of the content of press releases. Stock prices fully reflect the content of press releases at market open. If press releases are leaked, it offers predictive advantage. Topic analysis reveals self-serving bias in managerial narratives. Our framework supports real-time return prediction through the integration of online learning, provides interpretability and reveals the nuanced role of language in price formation. ...

Can We Reliably Predict the Fed’s Next Move? A Multi-Modal Approach to U.S. Monetary Policy Forecasting ArXiv ID: 2506.22763 “View on arXiv” Authors: Fiona Xiao Jingyi, Lili Liu Abstract Forecasting central bank policy decisions remains a persistent challenge for investors, financial institutions, and policymakers due to the wide-reaching impact of monetary actions. In particular, anticipating shifts in the U.S. federal funds rate is vital for risk management and trading strategies. Traditional methods relying only on structured macroeconomic indicators often fall short in capturing the forward-looking cues embedded in central bank communications. This study examines whether predictive accuracy can be enhanced by integrating structured data with unstructured textual signals from Federal Reserve communications. We adopt a multi-modal framework, comparing traditional machine learning models, transformer-based language models, and deep learning architectures in both unimodal and hybrid settings. Our results show that hybrid models consistently outperform unimodal baselines. The best performance is achieved by combining TF-IDF features of FOMC texts with economic indicators in an XGBoost classifier, reaching a test AUC of 0.83. FinBERT-based sentiment features marginally improve ranking but perform worse in classification, especially under class imbalance. SHAP analysis reveals that sparse, interpretable features align more closely with policy-relevant signals. These findings underscore the importance of integrating textual and structured signals transparently. For monetary policy forecasting, simpler hybrid models can offer both accuracy and interpretability, delivering actionable insights for researchers and decision-makers. ...

ChatGPT and Corporate Policies ArXiv ID: 2409.17933 “View on arXiv” Authors: Unknown Abstract We create a firm-level ChatGPT investment score, based on conference calls, that measures managers’ anticipated changes in capital expenditures. We validate the score with interpretable textual content and its strong correlation with CFO survey responses. The investment score predicts future capital expenditure for up to nine quarters, controlling for Tobin’s $q$ and other determinants, implying the investment score provides incremental information about firms’ future investment opportunities. The investment score also separately forecasts future total, intangible, and R&D investments. Consistent with theoretical predictions, high-investment-score firms experience significant positive short-term returns upon disclosure, and negative long-run future abnormal returns. We demonstrate ChatGPT’s applicability to measure other policies, such as dividends and employment. ...

Corporate Climate Risk: Measurements and Responses ArXiv ID: ssrn-3508497 “View on arXiv” Authors: Unknown Abstract This paper conducts a textual analysis of earnings call transcripts to quantify climate risk exposure at the firm level. We construct dictionaries that measure Keywords: Climate Risk, Textual Analysis, Earnings Calls, Environmental Exposure, Corporate Equities Complexity vs Empirical Score Math Complexity: 4.0/10 Empirical Rigor: 6.0/10 Quadrant: Street Traders Why: The research focuses on textual analysis and dictionary construction with relatively basic statistical measures, placing it in low-to-moderate math complexity. However, the use of earnings call transcripts, firm-level quantification, and likely implementation of text mining tools suggests a data-heavy, backtest-ready approach suited for practical trading or risk management. flowchart TD A["Research Goal<br>Quantify firm-level climate risk"] --> B["Data Source<br>Earnings Call Transcripts"] B --> C["Methodology<br>Textual Analysis & Dictionary Construction"] C --> D["Computational Process<br>Measure Risk Exposure Scores"] D --> E{"Key Outcomes"} E --> F["Climate Risk Quantified<br>at Firm Level"] E --> G["Discriminates between<br>Physical & Transition Risks"]

Textual Analysis in Accounting and Finance: A Survey ArXiv ID: ssrn-2959518 “View on arXiv” Authors: Unknown Abstract Relative to quantitative methods traditionally used in accounting and finance, textual analysis is substantially less precise. Thus, understanding the art is of Keywords: Textual Analysis, Accounting Research, Finance Research, Natural Language Processing, General (Accounting & Finance) Complexity vs Empirical Score Math Complexity: 1.0/10 Empirical Rigor: 2.0/10 Quadrant: Philosophers Why: The paper is a survey of textual analysis methods, which are conceptually oriented and less mathematically dense, and while it discusses empirical applications, it lacks the specific implementation details, code, or backtests required for high empirical rigor. flowchart TD A["Research Goal:<br>Textual Analysis in Accounting & Finance"] --> B["Data Collection"] B --> C["Preprocessing & Normalization"] C --> D["Textual Analysis Methodology"] D --> E["Statistical & Computational Processing"] E --> F["Key Findings/Outcomes"] subgraph B ["Data/Inputs"] B1["Financial Statements"] B2["Regulatory Filings"] B3["Earnings Calls"] B4["News & Social Media"] end subgraph C ["Preprocessing"] C1["Tokenization"] C2["Stopword Removal"] C3["Stemming/Lemmatization"] end subgraph D ["Methodology"] D1["Linguistic Metrics"] D2["Sentiment Analysis"] D3["Topic Modeling"] D4["Machine Learning"] end subgraph E ["Computational Processes"] E1["Feature Extraction"] E2["Statistical Inference"] E3["Model Validation"] end subgraph F ["Outcomes"] F1["Financial Prediction"] F2["Risk Assessment"] F3["Market Efficiency Insights"] end

Textual Analysis in Accounting andFinance: A Survey ArXiv ID: ssrn-2504147 “View on arXiv” Authors: Unknown Abstract Relative to quantitative methods traditionally used in accounting and finance, textual analysis is substantially less precise. Thus, understanding the art is of Keywords: Textual Analysis, Accounting Research, Finance Research, Natural Language Processing, General (Accounting & Finance) Complexity vs Empirical Score Math Complexity: 3.0/10 Empirical Rigor: 2.0/10 Quadrant: Philosophers Why: The paper is a survey of textual analysis methods, focusing on conceptual frameworks and methodological ’tripwires’ rather than advanced mathematical derivations or empirical backtesting; it emphasizes understanding the art and science of text processing without presenting new quantitative models or implementation-heavy data. flowchart TD A["Research Goal: Quantify Text in Financial Contexts"] --> B["Data Sources<br>10-Ks, Earnings Calls, News"] B --> C["Methodology<br>Preprocessing &amp; Dictionaries"] C --> D["Computational Process<br>Sentiment/Readability Scoring"] D --> E{"Outcome"} E --> F["Findings: Sentiment predicts returns/volatility"] E --> G["Findings: Readability impacts cost of capital"]