Improving financial distress prediction using textual sentiment of annual reports

Bo Huang, Xiao Yao (), Yinqing Luo and Jing Li
Additional contact information
Bo Huang: Renmin University of China
Xiao Yao: Central University of Finance and Economics
Yinqing Luo: Renmin University of China
Jing Li: University of International Business and Economics

Annals of Operations Research, 2023, vol. 330, issue 1, No 17, 457-484

Abstract: Abstract An accurate prediction of financial distress is beneficial to investors and allows banks and other financial institutions to build an early warning system to avoid risk contagion. This study investigated financial distress prediction using textual sentiment extracted from listed firms’ annual reports in the Chinese market. The sentiments reflected by the firms’ management discussions and analysis (MD&A) sections and audit reports were extracted separately through the application of deep learning algorithms. We found that the sentiment score extracted from MD&A sections was more optimistic compared with that extracted from audit reports. Moreover, the experimental results demonstrated that the modeling performance was significantly improved with the incorporation of textual sentiment scores, and the inclusion of sentiment from audit reports lead to a more significant incremental improvement than that from the MD&A sections. However, when both sentiment scores were included in the modeling input, the improvement in predictive accuracy was insignificant compared to the model using audit report scores only. Our study highlights the predictive power of textual information in annual reports, and shows that the textual sentiment of annual reports should be applied in distress modeling. The results provide implications for the utilization of soft information in credit risk modeling in the context of Chinese market, and such application can be further explored in other areas of operational research studies.

Keywords: Risk management; Textual sentiment; Deep learning; Financial distress prediction (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://link.springer.com/10.1007/s10479-022-04633-3 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

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:spr:annopr:v:330:y:2023:i:1:d:10.1007_s10479-022-04633-3

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10479

DOI: 10.1007/s10479-022-04633-3

Access Statistics for this article

Annals of Operations Research is currently edited by Endre Boros

More articles in Annals of Operations Research from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().