The determinants of voluntary disclosure: Integration of eXtreme gradient boost (XGBoost) and explainable artificial intelligence (XAI) techniques

Yu-Hsin Lu and Yu-Cheng Lin

International Review of Financial Analysis, 2024, vol. 96, issue PA

Abstract: Financial information transparency is vital for the various users of financial statements. This study employs the Explainable Artificial Intelligence (XAI) approach, utilizing eXtreme Gradient Boost (XGBoost) to explore management's motivations for voluntary disclosure. By transforming financial data into various plots, we introduce a voluntary disclosure model that enhances interpretability through Shapley Additive exPlanations (SHAP) techniques. These XAI methods aim to clarify different results in the voluntary disclosure literature, addressing the ongoing debate within the financial research community regarding voluntary disclosure. This research marks a significant advancement in voluntary disclosure by merging the transparency of XAI with effective voluntary disclosure prediction, offering a more comprehensive understanding of the determinants of voluntary disclosure.

Keywords: Voluntary disclosure; Unaudited earnings; eXtreme gradient boost (XGBoost); Explainable artificial intelligence (XAI) (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S105752192400509X
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:finana:v:96:y:2024:i:pa:s105752192400509x

DOI: 10.1016/j.irfa.2024.103577

Access Statistics for this article

International Review of Financial Analysis is currently edited by B.M. Lucey

More articles in International Review of Financial Analysis from Elsevier
Bibliographic data for series maintained by Catherine Liu ().