pith. sign in

arxiv: 2512.02510 · v3 · submitted 2025-12-02 · 💰 econ.GN · q-fin.EC

Forecasting financial distress in dynamic environments AI adoption signals and temporally pruned training windows

Frederik Rech (1) , Hussam Musa (2) , Martin \v{S}ebe\v{n}a (3) , Siele Jean Tuo (4) ((1) School of Economics , Beijing Institute of Technology , Beijing , China (1) Faculty of Economics , Shenzhen MSU-BIT University
show 12 more authors
Shenzhen China (2) Faculty of Economics Matej Bel University Bansk\'a Bystrica Slovakia (3) Faculty of Arts Social Sciences Hong Kong Baptist University Hong Kong China (4) Business School Liaoning University Shenyang China)
This is my paper

Pith reviewed 2026-05-17 02:58 UTC · model grok-4.3

classification 💰 econ.GN q-fin.EC
keywords financial distress forecastingAI adoptionmachine learning classifierstemporal distribution shiftsChinese A-share firmspatent datatextual disclosurestree-based ensembles
0
0 comments X

The pith

Firm-level AI adoption proxies improve machine learning forecasts of corporate financial distress when training windows are temporally pruned to recent data.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper investigates whether signals of artificial intelligence adoption at the firm level can enhance predictions of financial distress beyond traditional accounting measures. It uses data from Chinese listed firms between 2007 and 2023 and tests six machine learning models in a setup where the test period is fixed but the training data is progressively limited to more recent years. The results show consistent gains in predictive accuracy and fewer missed distress cases, especially with tree-based methods. This approach highlights the need for models that adapt to changing technological landscapes rather than relying on long historical datasets.

Core claim

AI proxies derived from textual disclosures and patent filings provide additional forecasting power for corporate financial distress. In out-of-sample tests with a fixed final test year, models incorporating these proxies show better discrimination and lower Type II errors compared to baselines using only fundamentals. The gains are largest in tree-based ensemble classifiers. Model performance varies non-monotonically with the length of the training window, with recent data yielding superior results over full historical spans, while single-year windows are unstable.

What carries the argument

AI adoption indicators constructed from firm textual disclosures and patent data, integrated into machine learning classifiers trained under chronologically pruned windows to handle temporal distribution shifts.

If this is right

  • AI proxies consistently improve out-of-sample discrimination and reduce Type II errors.
  • Tree-based ensembles show the strongest performance gains from including AI signals.
  • Predictive accuracy is non-monotonic in training window length, favoring recent data over complete histories.
  • Single-year training windows prove unreliable for robust forecasts.
  • Financial ratios remain the primary drivers, but AI adoption adds incremental content whose risk interpretation shifts with training regimes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Distress forecasting systems in tech-intensive industries may require regular retraining on recent observations to maintain accuracy.
  • The approach could extend to other domains where rapid technological change alters firm risk profiles, such as credit risk or supply chain stability.
  • Future work might test whether direct measures of AI implementation, like investment in AI tools, yield even stronger signals than disclosure-based proxies.

Load-bearing premise

The constructed AI adoption proxies from disclosures and patents accurately reflect actual firm-level adoption of transformative technologies without substantial measurement error or confounding factors.

What would settle it

A replication using direct measures of AI technology usage or implementation data that finds no improvement in forecast performance when added to accounting fundamentals would falsify the claim.

read the original abstract

Forecasting corporate financial distress increasingly requires capturing firms' adoption of transformative technologies such as artificial intelligence, yet model performance remains vulnerable to temporal distribution shifts as these technologies diffuse. This study investigates whether firm-level artificial intelligence (AI) adoption proxies improve forecasting performance beyond standard accounting fundamentals. Using a panel of Chinese A-share non-financial firms from 2007 to 2023, we construct AI indicators from textual disclosures and patent data. We benchmark six machine learning classifiers under a strictly chronological design that fixes the final test year and progressively prunes the training history to capture temporal change. Results indicate that AI proxies consistently improve out-of-sample discrimination and reduce Type II errors, with the strongest gains in tree-based ensembles. Predictive performance is non-monotonic in training window length; models trained on recent data outperform those using full history, while single-year training proves unreliable. Explainability analyses reveal financial ratios as primary drivers, with AI adoption signals adding incremental forecasting content whose interpretation as a risk factor varies across training regimes. Our findings establish AI proxies as valuable predictors for distress screening and demonstrate that adaptive, temporally pruned forecasting windows are essential for robust early warning models in rapidly evolving technological and economic environments.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript investigates whether firm-level AI adoption proxies constructed from textual disclosures and patent data improve machine learning forecasts of financial distress for Chinese A-share non-financial firms (2007–2023). Using six classifiers under a strictly chronological out-of-sample design with fixed final test year and progressively pruned training windows, it reports that the proxies enhance discrimination, reduce Type II errors (especially in tree-based ensembles), and that performance is non-monotonic in training-window length with recent data outperforming full history.

Significance. If the results hold after addressing methodological gaps, the work would contribute to the financial-distress and credit-risk literature by showing incremental predictive content from technology-adoption signals and by demonstrating the practical value of temporally adaptive training windows in the presence of distribution shifts. The chronological evaluation design and emphasis on Type II error reduction are positive features that could inform regulatory early-warning systems.

major comments (3)
  1. [Abstract / Methods] Abstract and proxy-construction description: the central claim that AI proxies improve out-of-sample performance rests on the unverified assumption that textual and patent-based indicators measure actual transformative AI adoption rather than disclosure volume or strategic reporting. No validation (e.g., correlation with external adoption surveys or falsification tests on non-AI patents) is supplied, which is load-bearing given varying disclosure incentives in the Chinese A-share panel over 2007–2023.
  2. [Results] Results and evaluation sections: performance gains are reported without details on hyperparameter tuning, cross-validation procedures, or statistical significance tests for improvements in discrimination metrics or Type II error rates. This omission prevents assessment of whether the reported gains are robust or could arise from tuning choices.
  3. [Robustness checks / Temporal design] Robustness and temporal design: while the paper emphasizes temporally pruned windows, it does not report sensitivity checks to alternative chronological splits, different pruning thresholds, or redefinitions of the AI proxies. Such checks are necessary to support the claim that gains are consistent and that recent-data windows are reliably superior.
minor comments (2)
  1. [Data and variables] Clarify the precise construction rules (keywords, NLP pipeline, patent lag handling) for the AI indicators in the main text or an appendix table so that the proxies can be replicated.
  2. [Tables and figures] Label training-window lengths and performance metrics consistently across tables and figures to improve readability of the non-monotonicity result.

Simulated Author's Rebuttal

3 responses · 1 unresolved

We thank the referee for the constructive and detailed comments. We address each major point below and indicate the revisions made to the manuscript.

read point-by-point responses

  1. Referee: [Abstract / Methods] Abstract and proxy-construction description: the central claim that AI proxies improve out-of-sample performance rests on the unverified assumption that textual and patent-based indicators measure actual transformative AI adoption rather than disclosure volume or strategic reporting. No validation (e.g., correlation with external adoption surveys or falsification tests on non-AI patents) is supplied, which is load-bearing given varying disclosure incentives in the Chinese A-share panel over 2007–2023.

    Authors: We acknowledge that our AI proxies are indirect signals and could partly capture disclosure volume or strategic reporting rather than transformative adoption. In the revised manuscript we have expanded the proxy-construction subsection to discuss these limitations explicitly, including the keyword and IPC selection criteria used to focus on substantive AI content. We have also added a falsification exercise using non-AI patent classes that shows no comparable predictive gains, supporting specificity. Comprehensive external adoption surveys for the full 2007–2023 Chinese A-share panel do not exist, so direct correlation validation remains infeasible. revision: partial

  2. Referee: [Results] Results and evaluation sections: performance gains are reported without details on hyperparameter tuning, cross-validation procedures, or statistical significance tests for improvements in discrimination metrics or Type II error rates. This omission prevents assessment of whether the reported gains are robust or could arise from tuning choices.

    Authors: We agree that these details are essential. The revised manuscript now includes a new subsection describing the hyperparameter grid search and rolling time-series cross-validation performed within each chronologically pruned training window. We additionally report McNemar tests on classification outcomes and Diebold-Mariano tests on AUC differences, confirming that the gains from adding AI proxies are statistically significant. These procedures and test results appear in the main text and a new appendix table. revision: yes

  3. Referee: [Robustness checks / Temporal design] Robustness and temporal design: while the paper emphasizes temporally pruned windows, it does not report sensitivity checks to alternative chronological splits, different pruning thresholds, or redefinitions of the AI proxies. Such checks are necessary to support the claim that gains are consistent and that recent-data windows are reliably superior.

    Authors: We have added a dedicated robustness section containing three sets of checks: (i) shifting the fixed test year to 2022 and 2021, (ii) additional pruning thresholds (3-, 7-, and 10-year windows), and (iii) alternative proxy definitions (text-only and patent-only). The new table shows that AI-proxy gains and the advantage of recent windows remain consistent across specifications, although effect sizes vary modestly with window length. revision: yes

standing simulated objections not resolved
  • Direct correlation of the AI proxies with external firm-level AI adoption surveys for Chinese A-share firms over the entire 2007–2023 period, as no such comprehensive survey data are available.

Circularity Check

0 steps flagged

No significant circularity in empirical forecasting setup

full rationale

The paper is an empirical ML study that constructs AI proxies from textual and patent data as inputs, trains classifiers on chronologically pruned windows, and evaluates discrimination on held-out future periods against actual distress labels. No equations or claims reduce predictions to fitted parameters by construction, no self-definitional loops, and no load-bearing self-citations that substitute for external validation. Results are tested against independent out-of-sample benchmarks, making the chain self-contained and falsifiable.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the validity of textual and patent-based proxies as measures of AI adoption and on standard machine-learning assumptions for non-stationary financial panel data.

axioms (1)
  • domain assumption AI indicators from textual disclosures and patent data serve as valid proxies for firm-level AI adoption
    Invoked when constructing the AI adoption variables that are then added to the feature set.

pith-pipeline@v0.9.0 · 5615 in / 1139 out tokens · 45721 ms · 2026-05-17T02:58:28.037143+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

49 extracted references · 49 canonical work pages

  1. [1]

    Artificial intelligence and jobs: Evidence from online vacancies

    Acemoglu, D., Autor, D., Hazell, J., Restrepo, P., 2022. Artificial intelligence and jobs: Evidence from online vacancies. J. Labor Econ. 40(S1). https://doi.org/10.1086/718327

    work page doi:10.1086/718327 2022

  2. [2]

    Predicting corporate financial failure using macroeconomic variables and accounting data

    Acosta-González, E., Fernández-Rodríguez, F., Ganga, H., 2019. Predicting corporate financial failure using macroeconomic variables and accounting data. Comput. Econ. 53(1), 227–257. https://doi.org/10.1007/s10614-017-9737-x

    work page doi:10.1007/s10614-017-9737-x 2019

  3. [3]

    Financial ratios, discriminant analysis and the prediction of corporate bankruptcy

    Altman, E.I., 1968. Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. J. Finance 23(4), 589–609. https://doi.org/10.2307/2978933

    work page doi:10.2307/2978933 1968

  4. [4]

    Corporate Financial Distress: A Complete Guide to Predicting, Avoiding, and Dealing With Bankruptcy, 2nd ed

    Altman, E.I., 1983. Corporate Financial Distress: A Complete Guide to Predicting, Avoiding, and Dealing With Bankruptcy, 2nd ed. Wiley Interscience, John Wiley and Sons

    work page 1983

  5. [5]

    Financial distress prediction in an international context: A review and empirical analysis of Altman’s Z-score model

    Altman, E.I., Iwanicz-Drozdowska, M., Laitinen, E.K., Suvas, A., 2017. Financial distress prediction in an international context: A review and empirical analysis of Altman’s Z-score model. J. Int. Financ. Manag. Account. 28(2), 131–171. https://doi.org/10.1111/JIFM.12053

    work page doi:10.1111/jifm.12053 2017

  6. [6]

    Artificial intelligence, firm growth, and product innovation

    Babina, T., Fedyk, A., He, A., Hodson, J., 2024. Artificial intelligence, firm growth, and product innovation. J. Financ. Econ. 151, 103745. https://doi.org/10.1016/j.jfineco.2023.103745

    work page doi:10.1016/j.jfineco.2023.103745 2024

  7. [7]

    Innovate or die: Corporate innovation and bankruptcy forecasts

    Bai, Q., Tian, S., 2020. Innovate or die: Corporate innovation and bankruptcy forecasts. J. Empir. Financ. 59, 88–108

    work page 2020

  8. [8]

    Financial ratios as predictors of failure

    Beaver, W.H., 1966. Financial ratios as predictors of failure. J. Account. Res. 4, 71–111. https://doi.org/10.2307/2490171

    work page doi:10.2307/2490171 1966

  9. [9]

    The role of data for AI startup growth

    Bessen, J., Impink, S.M., Reichensperger, L., Seamans, R., 2022. The role of data for AI startup growth. Res. Policy 51(5), 104513. https://doi.org/10.1016/j.respol.2022.104513

    work page doi:10.1016/j.respol.2022.104513 2022

  10. [10]

    Normalizing pandemic data for credit scoring

    Breeden, J.L., 2025. Normalizing pandemic data for credit scoring. J. Risk Financ. Manag. 18(11), 657. https://doi.org/10.3390/jrfm18110657

    work page doi:10.3390/jrfm18110657 2025

  11. [11]

    URL https://www.aeaweb.org/articles? id=10.1257/mac.20180386

    Brynjolfsson, E., Rock, D., Syverson, C., 2021. The productivity J-curve: How intangibles complement general purpose technologies. Am. Econ. J. Macroecon. 13(1), 333–372. https://doi.org/10.1257/mac.20180386

    work page doi:10.1257/mac.20180386 2021

  12. [12]

    In search of distress risk

    Campbell, J.Y., Hilscher, J., Szilagyi, J., 2008. In search of distress risk. J. Finance 63(6). https://doi.org/10.1111/j.1540-6261.2008.01416.x

    work page doi:10.1111/j.1540-6261.2008.01416.x 2008

  13. [13]

    Research on the impact of artificial intelligence technology on urban public health resilience

    Chen, E., Zhang, H., 2025. Research on the impact of artificial intelligence technology on urban public health resilience. Front. Public Health 12. https://doi.org/10.3389/fpubh.2024.1506930

    work page doi:10.3389/fpubh.2024.1506930 2025

  14. [14]

    Artificial intelligence and corporate risk-taking: Evidence from China

    Chen, H., Zhang, M., Zeng, J., Wang, W., 2024. Artificial intelligence and corporate risk-taking: Evidence from China. China J. Account. Res. 17(3), 100372. https://doi.org/10.1016/j.cjar.2024.100372

    work page doi:10.1016/j.cjar.2024.100372 2024

  15. [15]

    Patent statistics: A good indicator for innovation in China? Patent subsidy program impacts on patent quality

    Dang, J., Motohashi, K., 2015. Patent statistics: A good indicator for innovation in China? Patent subsidy program impacts on patent quality. China Econ. Rev. 35, 137–155. https://doi.org/10.1016/j.chieco.2015.03.012

    work page doi:10.1016/j.chieco.2015.03.012 2015

  16. [16]

    Impact of artificial intelligence on enterprise risk management

    Dvorsky, J., 2025. Impact of artificial intelligence on enterprise risk management. A case study from the Slovak SME segment. J. Bus. Sect. 3(1), 96–103. https://doi.org/10.62222/CAJA0666

    work page doi:10.62222/caja0666 2025

  17. [17]

    Artificial intelligence (AI): Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and policy

    Dwivedi, Y.K., Hughes, L., Ismagilova, E., Aarts, G., Coombs, C., Crick, T., et al., 2021. Artificial intelligence (AI): Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and policy. Int. J. Inf. Manag. 57, 101994. https://doi.org/10.1016/j.ijinfomgt.2019.08.00

    work page doi:10.1016/j.ijinfomgt.2019.08.00 2021

  18. [18]

    European country heterogeneity in financial distress prediction: An empirical analysis with macroeconomic and regulatory factors

    Fernández-Gámez, M.Á., Soria, J.A.C., Santos, J.A.C., Alaminos, D., 2020. European country heterogeneity in financial distress prediction: An empirical analysis with macroeconomic and regulatory factors. Econ. Model. 88, 398–407. https://doi.org/10.1016/j.econmod.2019.09.050

    work page doi:10.1016/j.econmod.2019.09.050 2020

  19. [19]

    Prediction of financial distress: An empirical study of listed Chinese companies using data mining

    Geng, R., Bose, I., Chen, X., 2015. Prediction of financial distress: An empirical study of listed Chinese companies using data mining. Eur. J. Oper. Res. 241(1), 236–247. https://doi.org/10.1016/j.ejor.2014.08.016

    work page doi:10.1016/j.ejor.2014.08.016 2015

  20. [20]

    Research Policy , author =

    Goldfarb, A., Taska, B., Teodoridis, F., 2023. Could machine learning be a general purpose technology? A comparison of emerging technologies using data from online job postings. Res. Policy 52(1), 104653. https://doi.org/10.1016/j.respol.2022.104653

    work page doi:10.1016/j.respol.2022.104653 2023

  21. [21]

    Artificial intelligence trust, risk and security management (AI TRiSM): Frameworks, applications, challenges and future research directions

    Habbal, A., Ali, M.K., Abuzaraida, M.A., 2024. Artificial intelligence trust, risk and security management (AI TRiSM): Frameworks, applications, challenges and future research directions. Expert Syst. Appl. 240, 122442. https://doi.org/10.1016/j.eswa.2023.122442

    work page doi:10.1016/j.eswa.2023.122442 2024

  22. [22]

    Artificial intelligence and firm resilience: Empirical evidence from natural disaster shocks

    Han, M., Shen, H., Wu, J., Zhang, X.M., 2025. Artificial intelligence and firm resilience: Empirical evidence from natural disaster shocks. Inf. Syst. Res. https://doi.org/10.1287/isre.2022.0440

    work page doi:10.1287/isre.2022.0440 2025

  23. [23]

    Does AI contribute to systemic risk reduction in non-financial corporations? Q

    Han, W.-Z., Meng, W., 2025. Does AI contribute to systemic risk reduction in non-financial corporations? Q. Rev. Econ. Financ. 100, 101973. https://doi.org/10.1016/j.qref.2025.101973

    work page doi:10.1016/j.qref.2025.101973 2025

  24. [24]

    Diagnosis with incomplete multi-view data: A variational deep financial distress prediction method

    Huang, Y., Wang, Z., Jiang, C., 2024. Diagnosis with incomplete multi-view data: A variational deep financial distress prediction method. Technol. Forecast. Soc. Chang. 201, 123269. https://doi.org/10.1016/j.techfore.2024.123269

    work page doi:10.1016/j.techfore.2024.123269 2024

  25. [25]

    Forecasting the macrolevel determinants of entrepreneurial opportunities using artificial intelligence models

    Jabeur, S.B., Ballouk, H., Mefteh-Wali, S., Omri, A., 2022. Forecasting the macrolevel determinants of entrepreneurial opportunities using artificial intelligence models. Technol. Forecast. Soc. Chang. 175, 121353. https://doi.org/10.1016/j.techfore.2021.121353

    work page doi:10.1016/j.techfore.2021.121353 2022

  26. [26]

    Mining semantic features in patent text for financial distress prediction

    Jiang, C., Zhou, Y., Chen, B., 2023. Mining semantic features in patent text for financial distress prediction. Technol. Forecast. Soc. Chang. 190, 122450. https://doi.org/10.1016/j.techfore.2023.122450

    work page doi:10.1016/j.techfore.2023.122450 2023

  27. [27]

    Development of new generation of artificial intelligence in China: When Beijing’s global ambitions meet local realities

    Khanal, S., Zhang, H., Taeihagh, A., 2024. Development of new generation of artificial intelligence in China: When Beijing’s global ambitions meet local realities. J. Contemp. China 34(151), 19–42. https://doi.org/10.1080/10670564.2024.2333492

    work page doi:10.1080/10670564.2024.2333492 2024

  28. [28]

    Intertwining artificial intelligence and efficiency: An empirical analysis of AI focus and operational efficacy in Chinese listed firms

    Li, Q., Zhang, Y., Um, G., 2025. Intertwining artificial intelligence and efficiency: An empirical analysis of AI focus and operational efficacy in Chinese listed firms. Financ. Res. Lett. 80, 107451. https://doi.org/10.1016/j.frl.2025.107451

    work page doi:10.1016/j.frl.2025.107451 2025

  29. [29]

    Does artificial intelligence facilitate the balancing of short-term returns and long-term growth in firms? Evidence from China

    Lin, L., Sun, R., 2026. Does artificial intelligence facilitate the balancing of short-term returns and long-term growth in firms? Evidence from China. Technol. Forecast. Soc. Chang. 223, 124460. https://doi.org/10.1016/j.techfore.2025.124460

    work page doi:10.1016/j.techfore.2025.124460 2026

  30. [30]

    Artificial intelligence adoption and corporate financial risk

    Liu, S., Gao, L., Chen, M., 2025. Artificial intelligence adoption and corporate financial risk. Financ. Res. Lett. 85, 107938. https://doi.org/10.1016/j.frl.2025.107938

    work page doi:10.1016/j.frl.2025.107938 2025

  31. [31]

    Credit default prediction of Chinese real estate listed companies based on explainable machine learning

    Ma, Y., Zhang, P., Duan, S., Zhang, T., 2023. Credit default prediction of Chinese real estate listed companies based on explainable machine learning. Financ. Res. Lett. 58, 104305. https://doi.org/10.1016/j.frl.2023.104305

    work page doi:10.1016/j.frl.2023.104305 2023

  32. [32]

    Corporate governance and financial distress in China: A multi-dimensional nonlinear study based on machine learning

    Meng, Q., Zheng, X., Wang, S., 2024. Corporate governance and financial distress in China: A multi-dimensional nonlinear study based on machine learning. Pac.-Basin Financ. J. 88, 102549. https://doi.org/10.1016/j.pacfin.2024.102549

    work page doi:10.1016/j.pacfin.2024.102549 2024

  33. [33]

    Bankruptcy prediction using machine learning and Shapley additive explanations

    Nguyen, H.H., Viviani, J.L., Ben Jabeur, S., 2023. Bankruptcy prediction using machine learning and Shapley additive explanations. Rev. Quant. Financ. Account. 65, 107–148. https://doi.org/10.1007/s11156-023-01192-x

    work page doi:10.1007/s11156-023-01192-x 2023

  34. [34]

    Financial ratios and the probabilistic prediction of bankruptcy

    Ohlson, J.A., 1980. Financial ratios and the probabilistic prediction of bankruptcy. J. Account. Res. 18(1), 109–131. https://doi.org/10.2307/2490395

    work page doi:10.2307/2490395 1980

  35. [35]

    Effects of early patent publication on knowledge dissemination: Evidence from U.S

    Okada, Y., Nagaoka, S., 2020. Effects of early patent publication on knowledge dissemination: Evidence from U.S. patent law reform. Inf. Econ. Policy 51, 100852. https://doi.org/10.1016/j.infoecopol.2020.100852

    work page doi:10.1016/j.infoecopol.2020.100852 2020

  36. [36]

    The possibilities of using AutoML in bankruptcy prediction: Case of Slovakia

    Papík, M., Papíková, L., 2025. The possibilities of using AutoML in bankruptcy prediction: Case of Slovakia. Technol. Forecast. Soc. Chang. 215, 124098. https://doi.org/10.1016/j.techfore.2025.124098

    work page doi:10.1016/j.techfore.2025.124098 2025

  37. [37]

    Application of intellectual capital in SME bankruptcy

    Papíková, L., Papík, M., 2023. Application of intellectual capital in SME bankruptcy. Appl. Econ. 56(55), 7317–7338. https://doi.org/10.1080/00036846.2023.2281291

    work page doi:10.1080/00036846.2023.2281291 2023

  38. [38]

    Surviving the pandemic: Financial distress prediction for Slovak SME manufacturers

    Rech, F., Isaboke, C., Xu, H., 2025. Surviving the pandemic: Financial distress prediction for Slovak SME manufacturers. J. Bus. Sect. 3(1), 41–51. https://doi.org/10.62222/SNRN2189

    work page doi:10.62222/snrn2189 2025

  39. [39]

    The Chinese approach to artificial intelligence: An analysis of policy, ethics, and regulation

    Roberts, H., Cowls, J., Morley, J., Taddeo, M., Wang, V., Floridi, L., 2021. The Chinese approach to artificial intelligence: An analysis of policy, ethics, and regulation. AI Soc. 36(1), 59–77. https://doi.org/10.1007/s00146-020-00992-2

    work page doi:10.1007/s00146-020-00992-2 2021

  40. [40]

    Forecasting bankruptcy more accurately: A simple hazard model

    Shumway, T., 2001. Forecasting bankruptcy more accurately: A simple hazard model. J. Bus. 74(1), 101–124. https://doi.org/10.1086/209665

    work page doi:10.1086/209665 2001

  41. [41]

    AI is helping companies redefine, not just improve, performance

    Schrage, M., Kiron, D., Candelon, F., Khodabandeh, S., Chu, M., 2023. AI is helping companies redefine, not just improve, performance. MIT Sloan Manag. Rev. 64(3)

    work page 2023

  42. [42]

    Machine learning for corporate default risk: Multi-period prediction, frailty correlation, loan portfolios, and tail probabilities

    Sigrist, F., Leuenberger, N., 2023. Machine learning for corporate default risk: Multi-period prediction, frailty correlation, loan portfolios, and tail probabilities. Eur. J. Oper. Res. 305(3), 1390–1406. https://doi.org/10.1016/j.ejor.2022.06.035

    work page doi:10.1016/j.ejor.2022.06.035 2023

  43. [43]

    ESG performance and financial distress prediction of energy enterprises

    Song, Y., Li, R., Zhang, Z., Sahut, J.-M., 2024. ESG performance and financial distress prediction of energy enterprises. Financ. Res. Lett. 65, 105546. https://doi.org/10.1016/j.frl.2024.105546

    work page doi:10.1016/j.frl.2024.105546 2024

  44. [44]

    Impact of macroeconomic indicators on bankruptcy prediction models: Case of the Portuguese construction sector

    Sousa, A., Braga, A., Cunha, J., 2022. Impact of macroeconomic indicators on bankruptcy prediction models: Case of the Portuguese construction sector. Quant. Financ. Econ. 6(3), 405–432. https://doi.org/10.3934/QFE.2022018

    work page doi:10.3934/qfe.2022018 2022

  45. [45]

    Dynamic financial distress prediction using instance selection for the disposal of concept drift

    Sun, J., Li, H., 2011. Dynamic financial distress prediction using instance selection for the disposal of concept drift. Expert Syst. Appl. 38(3), 2566–2576. https://doi.org/10.1016/j.eswa.2010.08.046

    work page doi:10.1016/j.eswa.2010.08.046 2011

  46. [46]

    Firm default prediction: A Bayesian model-averaging approach

    Traczynski, J., 2017. Firm default prediction: A Bayesian model-averaging approach. J. Financ. Quant. Anal. 52(3). https://doi.org/10.1017/S002210901700031X

    work page doi:10.1017/s002210901700031x 2017

  47. [47]

    AI adoption and ESG performance: Evidence from China

    Yang, G., Yang, X., 2025. AI adoption and ESG performance: Evidence from China. Int. Rev. Econ. Financ. 104, 104659. https://doi.org/10.1016/j.iref.2025.104659

    work page doi:10.1016/j.iref.2025.104659 2025

  48. [48]

    Predicting financial distress of Chinese listed companies using machine learning: To what extent does textual disclosure matter? Int

    Zhao, Q., Xu, W., Ji, Y., 2023. Predicting financial distress of Chinese listed companies using machine learning: To what extent does textual disclosure matter? Int. Rev. Financ. Anal. 89, 102770. https://doi.org/10.1016/j.irfa.2023.102770

    work page doi:10.1016/j.irfa.2023.102770 2023

  49. [49]

    Artificial intelligence techniques for financial distress prediction

    Zhong, J., Wang, Z., 2022. Artificial intelligence techniques for financial distress prediction. AIMS Math. 7(12). https://doi.org/10.3934/math.20221145

    work page doi:10.3934/math.20221145 2022