Forecasting Financial Distress for Shaping Public Policy: An Empirical Investigation

Soumya Ranjan Sethi and Dushyant Mahadik

Acta Informatica Pragensia, 2025, vol. 2025, issue 1, 63-87

Abstract: Background: Prediction of financial distress has been made more accurate and reliable through machine learning methods. Financial stress affects the business corporate entity, society and the general economy. Analysing such nonlinear events is essential for preventing the dangers and supporting a favourable economic climate. Objective: This paper seeks to develop a robust predictive model for identifying firms in the Indian context other than the financial service sector that may face financial distress and also to check the impact of one essential predictor, i.e., future cash flow, on financial distress prediction. Besides, the study also aims at making research that can inform public policy and provide recommendations. Methods: The study employs financial information from the Prowess Database but is confined to non-financial service sector firms in India. Logistic regression, linear discriminant analysis (LDA), and artificial neural networks (ANNs) are applied to predict financial distress and their ability to foretell future cash flows. Other methods adopted in evaluating the models include accuracy, sensitivity, and specificity. Results: ANNs outperform the other models based on accuracy and predictability, which are higher than the rates given by the other two models, namely logistic regression and LDA. The ANN model performs well in identifying financially distressed firms; thus, it is informative in evaluating their financial position. Also, results suggest that future cash flow substantially affects financial distress prediction, an essential new variable that needs to be considered in future research. Conclusion: This predictive model of financial distress further gives a sound platform for the corresponding sector in India. In general, ANNs offer profound opportunities for managers, investors, policymakers, regulators and shareholders as an effective tool for preventive decision-making to reinforce the corporate world. This research demonstrates that high-level machine-learning approaches are still crucial in financial analysis and policymaking.

Keywords: Forecasting; Public policy; Sustainability; ANN; Future cash flow; FCF; Logistic regression; LR; Linear discriminant analysis; LDA (search for similar items in EconPapers)
Date: 2025
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
http://aip.vse.cz/doi/10.18267/j.aip.253.html (text/html)
http://aip.vse.cz/doi/10.18267/j.aip.253.pdf (application/pdf)
free of charge

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:prg:jnlaip:v:2025:y:2025:i:1:id:253:p:63-87

Ordering information: This journal article can be ordered from
Redakce Acta Informatica Pragensia, Katedra systémové analýzy, Vysoká škola ekonomická v Praze, nám. W. Churchilla 4, 130 67 Praha 3
http://aip.vse.cz

DOI: 10.18267/j.aip.253

Access Statistics for this article

Acta Informatica Pragensia is currently edited by Editorial Office

More articles in Acta Informatica Pragensia from Prague University of Economics and Business Contact information at EDIRC.
Bibliographic data for series maintained by Stanislav Vojir ().