research

published and accepted papers

1. publication

   The News in Earnings Announcement Disclosures: Capturing Word Context Using LLM Methods

   F., Siano

   2025, Management Science, Forthcoming

   Abs

   This study examines the information content of textual disclosures in firms’ earnings announcements. Using a large language model (LLM) to capture information in both words and word context, I show that the news in earnings press releases (i) explains three times more variation in short-window stock returns than a host of textual measures based on dictionary and non-LLM machine learning methods; (ii) doubles the R2 of an array of financial statement surprises, modeled with conventional regression or machine learning approaches; and (iii) accounts for a large fraction of immediate price revisions within just 5 minutes of release. LLM-modeled conference calls further enhance R2 by one-fourth compared to press releases and financial surprises. Textual disclosures are more informative when earnings are less persistent and during periods of aggregate uncertainty. Most news arises from text describing numbers, at the beginning of the disclosure, and including novel contents. These findings highlight the role of firms’ textual disclosures in moving stock prices and advance our understanding of how investors utilize corporate disclosures.

2. publication

   Changes in Risk Factor Disclosures and the Variance Risk Premium

   Lyle, M., Riedl, E., and F., Siano

   2023, The Accounting Review 98 (6): 1-26

   Abs HTML PDF

   This paper examines how changes in risk disclosures affect uncertainty about risk. We measure changes in risk disclosures using the addition and removal of individual risk factors to firms’ 10-K filings, identified via textual analysis of the risk factors section. Our market outcome is the variance risk premium (VRP), which captures the market’s pricing of uncertainty about firm risk. Following recent theoretical predictions, we predict and empirically document that newly disclosed signals of risk factor exposure—reflected in added and removed individual risk factors—decrease the uncertainty surrounding firm risk, as proxied via the VRP. We further confirm that individual risk factors offer incremental insights compared with alternative textual risk measures. Collectively, our findings suggest that textually evaluating individual risk factors reveals information about the uncertainty regarding firm risk.

3. publication

   Transfer Learning and Textual Analysis of Accounting Disclosures: Applying Big Data Methods to Small(er) Datasets

   Siano, F., and P.D., Wysocki

   2021, Accounting Horizons 35 (3): 217-244

   Abs HTML PDF

   We introduce and apply machine transfer learning methods to analyze accounting disclosures. We use the examples of the new BERT language model and sentiment analysis of quarterly earnings disclosures to demonstrate the key transfer learning concepts of: (i) pre-training on generic "Big Data", (ii) fine-tuning on small accounting datasets, and (iii) using a language model that captures context rather than stand-alone words. Overall, we show that this new approach is easy to implement, uses widely-available and low-cost computing resources, and has superior performance relative to existing textual analysis tools in accounting. We conclude with suggestions for opportunities to apply transfer learning to address important accounting research questions.

selected working papers

1. working paper

   Re-Adjusted Financial Statement Data: Challenges in Replicating Research

   Lyle, M., Siano, F., and T., Yohn
2. working paper

   When Micro Firms Speak Macro: Evidence on Firms’ Macroeconomic Disclosures

   Holstead, C., Kalay, A., Sadka, G., and F., Siano
3. working paper

   Beyond Words: The Challenge of Measuring Financial Text Across Borders

   Pinto, J., and F., Siano
4. working paper

   The Predictive Ability of ESG Risk Disclosures

   Korganbenkova, A., Riedl, E., Siano, F., and Y., Sun