Forecasting Bank Efficiency Using Data Envelopment Analysis with Directional Distance Functions and Machine Learning: Time-Series Validation and Shapley Value Interpretation

Chau Dinh Linh
Pages 1120-1135
Abstract Views 501
PDF Downloads 380
Crossref Citations -
DOI https://doi.org/10.47738/jads.v7i2.1244

Abstract


This study develops a structured framework to forecast the operational efficiency of commercial banks in Vietnam. The analysis is based on a balanced panel of 27 banks over the period 2016–2024. Bank efficiency is first measured using a directional distance function within a data envelopment analysis framework (DEA – DDF). This approach incorporates both desirable outputs and undesirable outputs related to credit risk. The estimated efficiency scores are then used as prediction targets in several machine learning models. Model performance is evaluated under both conventional test settings and time-series cross-validation, and predictions are interpreted using Shapley value–based analysis (SHAP). Under a conventional test set, the gradient boosting model (XGBoost) shows the best performance, with a root mean squared error of 0.060 and a coefficient of determination (R²) of 0.353. However, when time-series cross-validation is applied to reflect realistic forecasting conditions, predictive accuracy declines sharply. The average coefficient of determination falls to approximately 0.005. This suggests that static validation can overstate performance and that forecasting efficiency in a changing financial environment remains difficult. The interpretation results provide additional insights. Net interest margin has a positive effect on predicted efficiency, although the effect weakens at very high levels. The cost-to-income ratio shows a threshold around 0.55, beyond which efficiency declines more strongly. Bank size has a largely neutral impact. The interaction between capital adequacy and profitability shows a conditionally negative pattern. Prediction errors are larger in the most recent year and among banks with very high efficiency scores. In summary, the results highlight both the potential and the limitations of machine learning in forecasting efficiency and emphasize the importance of time-aware validation.


Keywords


Bank Efficiency; DEA – DDF; Machine Learning; Time-Series Validation; SHAP

Full Text:

PDF

References


Ö. O. Akdeniz, H. A. Abdou, A. Hayek, J. C. Nwachukwu, A. A. Elamer, and C. Pyke, “Technical efficiency in banks: a review of methods, recent innovations and future research agenda,” Review of Managerial Science, vol. 18, no. 11. Springer Science+Business Media, p. 3395, Dec. 21, 2023. doi: 10.1007/s11846-023-00707-z.

K. D. Dutta and M. Saha, “Do competition and efficiency lead to bank stability? Evidence from Bangladesh,” Future Business Journal, vol. 7, no. 1, Jan. 2021, doi: 10.1186/s43093-020-00047-4.

S. Lundberg and S. Lee, “A Unified Approach to Interpreting Model Predictions,” arXiv (Cornell University), May 2017, doi: 10.48550/arxiv.1705.07874.

M. Izzeldin, E. Mamatzakis, A. Murphy, V. Pappas, and E. G. Tsionas, “An innovative Bayesian multiple indicator-multiple cause analysis of bank productivity,” Review of Quantitative Finance and Accounting, Oct. 2025, doi: 10.1007/s11156-025-01445-x.

M. Koetter, “The Stability of Efficiency Rankings When Risk-Preferences and Objectives are Different,” SSRN Electronic Journal, Jan. 2006, doi: 10.2139/ssrn.2793981.

J. P. Hughes and L. J. Mester, “Measuring the Performance of Banks: Theory, Practice, Evidence, and Some Policy Implications,” SSRN Electronic Journal, Jan. 2013, doi: 10.2139/ssrn.2306003.

J. P. Hughes and L. J. Mester, “Efficiency in Banking: Theory, Practice, and Evidence,” Federal Reserve Bank of Philadelphia, Jan. 2008. doi: 10.21799/frbp.wp.2008.01.

X. Zhou, Z. Xu, J. Chai, L. Yao, S. Wang, and B. Lev, “Efficiency evaluation for banking systems under uncertainty: A multi-period three-stage DEA model,” Omega, vol. 85, p. 68, May 2018, doi: 10.1016/j.omega.2018.05.012.

A. P. Balcerzak et al., “Non-Parametric Approach to Measuring the Efficiency of Banking Sectors in European Union Countries,” Acta Polytechnica Hungarica, vol. 14, no. 7, Feb. 2018, doi: 10.12700/aph.14.7.2017.7.4.

J. Tanwar, H. Seth, A. K. Vaish, and N. V. M. Rao, “Revisiting the Efficiency of Indian Banking Sector: An Analysis of Comparative Models Through Data Envelopment Analysis,” Indian Journal of Finance and Banking, vol. 4, no. 1, p. 92, May 2020, doi: 10.46281/ijfb.v4i1.585.

R. Färe, S. Grosskopf, and D. Margaritis, “Directional Distance Functions Revisited: Selective Overview and Update,” Data Envelopment Analysis Journal, vol. 1, no. 2, p. 57, Jan. 2015, doi: 10.1561/103.00000006.

R. Färe and S. Grosskopf, “Theory and Application of Directional Distance Functions,” Journal of Productivity Analysis, vol. 13, no. 2, p. 93, Mar. 2000, doi: 10.1023/a:1007844628920.

C. Daraio, L. Simar, and P. W. Wilson, “Fast and efficient computation of directional distance estimators,” Annals of Operations Research, vol. 288, no. 2, p. 805, Feb. 2019, doi: 10.1007/s10479-019-03163-9.

B. K. Sahoo, M. Mehdiloozad, and K. Tone, “Cost, revenue and profit efficiency measurement in DEA: A directional distance function approach,” European Journal of Operational Research, vol. 237, no. 3, p. 921, Feb. 2014, doi: 10.1016/j.ejor.2014.02.017.

C. Yang, “An enhanced DEA model for decomposition of technical efficiency in banking,” Annals of Operations Research, vol. 214, no. 1, p. 167, Jul. 2011, doi: 10.1007/s10479-011-0926-z.

T. Le, “The efficiency effects of bank mergers: An analysis of case studies in Vietnam,” Risk Governance and Control Financial Markets & Institutions, vol. 7, no. 1, p. 61, Jan. 2017, doi: 10.22495/rgcv7i1art8.

L. T. Vu, N. T. T. Nguyen, and L. H. Dinh, “Measuring banking efficiency in Vietnam: parametric and non-parametric methods,” Banks and Bank Systems, vol. 14, no. 1, p. 55, Feb. 2019, doi: 10.21511/bbs.14(1).2019.06.

P. H. Nguyen and D. T. B. Pham, “The cost efficiency of Vietnamese banks – the difference between DEA and SFA,” Journal of Economics and Development, vol. 22, no. 2, p. 209, May 2020, doi: 10.1108/jed-12-2019-0075.

C. Wang, N.-A.-T. Nguyen, T. Dang, and T.-T.-Q. Trinh, “A Decision Support Model for Measuring Technological Progress and Productivity Growth: The Case of Commercial Banks in Vietnam,” Axioms, vol. 10, no. 3, p. 131, Jun. 2021, doi: 10.3390/axioms10030131.

B. Kelly, S. Malamud, and K. Zhou, “The Virtue of Complexity in Return Prediction,” The Journal of Finance, vol. 79, no. 1, p. 459, Dec. 2023, doi: 10.1111/jofi.13298.

S. Giglio, B. Kelly, and D. Xiu, “Factor Models, Machine Learning, and Asset Pricing,” Annual Review of Financial Economics, vol. 14, no. 1, p. 337, Aug. 2022, doi: 10.1146/annurev-financial-101521-104735.

B. Kelly and D. Xiu, “Financial Machine Learning,” Foundations and Trends® in Finance, vol. 13, p. 205, Jan. 2023, doi: 10.1561/0500000064.

N. Gafsi, “Machine Learning Approaches to Credit Risk: Comparative Evidence from Participation and Conventional Banks in the UK,” Journal of risk and financial management, vol. 18, no. 7, p. 345, Jun. 2025, doi: 10.3390/jrfm18070345.

G. Manthoulis, M. Doumpos, C. Zopounidis, and E. Galariotis, “An ordinal classification framework for bank failure prediction: Methodology and empirical evidence for US banks,” European Journal of Operational Research, vol. 282, no. 2, p. 786, Sep. 2019, doi: 10.1016/j.ejor.2019.09.040.

P. Appiahene, Y. M. Missah, and U. Najim, “Predicting Bank Operational Efficiency Using Machine Learning Algorithm: Comparative Study of Decision Tree, Random Forest, and Neural Networks,” Advances in Fuzzy Systems, vol. 2020, p. 1, Jul. 2020, doi: 10.1155/2020/8581202.

Q. H. Nguyen, H. V. Trinh, V. P. Truong, and T. T. M. Ly, “Early Warning System for Debt Group Migration: The Case of One Commercial Bank in Vietnam,” Foundations of Management, vol. 16, no. 1, p. 195, Jan. 2024, doi: 10.2478/fman-2024-0012.

S. Boubaker, T. Le, T. Ngo, and R. Manita, “Predicting the performance of MSMEs: a hybrid DEA-machine learning approach,” Annals of Operations Research, vol. 350, no. 2, p. 555, Feb. 2023, doi: 10.1007/s10479-023-05230-8.

T. van Trung and N. A. N. Vuong, “DEVELOPMENT OF A CREDIT SCORING MODEL USING MACHINE LEARNING FOR COMMERCIAL BANKS IN VIETNAM,” Advances and Applications in Statistics, vol. 92, no. 1, p. 107, Nov. 2024, doi: 10.17654/0972361725006.

L. Yang and A. Shami, “On hyperparameter optimization of machine learning algorithms: Theory and practice,” Neurocomputing, vol. 415, p. 295, Jul. 2020, doi: 10.1016/j.neucom.2020.07.061.

B. Bischl et al., “Hyperparameter optimization: Foundations, algorithms, best practices, and open challenges,” Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery, vol. 13, no. 2, Jan. 2023, doi: 10.1002/widm.1484.

L. Franceschi et al., “Hyperparameter Optimization in Machine Learning,” arXiv (Cornell University), Oct. 2024, doi: 10.48550/arxiv.2410.22854.

P.-D. Arsenault, S. Wang, and J.-M. Patenaude, “A Survey of Explainable Artificial Intelligence (XAI) in Financial Time Series Forecasting,” ACM Computing Surveys, vol. 57, no. 10. Association for Computing Machinery, p. 1, Apr. 18, 2025. doi: 10.1145/3729531.

P.-D. Arsenault, S. Wang, and J.-M. Patenande, “A Survey of Explainable Artificial Intelligence (XAI) in Financial Time Series Forecasting,” arXiv (Cornell University), Jul. 2024, doi: 10.48550/arxiv.2407.15909.

B. Bilodeau, N. Jaques, P. W. Koh, and B. Kim, “Impossibility theorems for feature attribution,” Proceedings of the National Academy of Sciences, vol. 121, no. 2, Jan. 2024, doi: 10.1073/pnas.2304406120.

C. Yang, M. Z. Abedin, H. Zhang, F. Weng, and P. Hájek, “An interpretable system for predicting the impact of COVID-19 government interventions on stock market sectors,” Annals of Operations Research, vol. 347, no. 2, p. 1031, Apr. 2023, doi: 10.1007/s10479-023-05311-8.

Y. Xia, Z. Liao, J. Xu, and Y. Li, “FROM CREDIT SCORING TO REGULATORY SCORING: COMPARING CREDIT SCORING MODELS FROM A REGULATORY PERSPECTIVE,” Technological and Economic Development of Economy, vol. 28, no. 6, p. 1954, Dec. 2022, doi: 10.3846/tede.2022.17045.

P. Bracke, A. Datta, C. Jung, and S. Sen, “Machine Learning Explainability in Finance: An Application to Default Risk Analysis,” SSRN Electronic Journal, Jan. 2019, doi: 10.2139/ssrn.3435104.

F. J. Bargagli-Stoffi, F. Incerti, M. Riccaboni, and A. Rungi, “Machine learning for zombie hunting: predicting distress from firms’ accounts and missing values,” Industrial and Corporate Change, vol. 33, no. 5, p. 1063, Oct. 2023, doi: 10.1093/icc/dtad049.

M. Alexandre, T. C. Silva, C. Connaughton, and F. A. Rodrigues, “The drivers of systemic risk in financial networks: a data-driven machine learning analysis,” Chaos Solitons & Fractals, vol. 153, p. 111588, Nov. 2021, doi: 10.1016/j.chaos.2021.111588.

S. Consoli, D. R. Recupero, and M. Saisana, Data Science for Economics and Finance. 2021. doi: 10.1007/978-3-030-66891-4.

H. Hewamalage, K. Ackermann, and C. Bergmeir, “Forecast evaluation for data scientists: common pitfalls and best practices,” Data Mining and Knowledge Discovery, vol. 37, no. 2, p. 788, Dec. 2022, doi: 10.1007/s10618-022-00894-5.

T. Dierckx, J. Davis, and W. Schoutens, “Using Machine Learning and Alternative Data to Predict Movements in Market Risk,” arXiv (Cornell University), Sep. 2020, doi: 10.48550/arxiv.2009.07947.

V. Cerqueira, L. Torgo, and I. Mozetič, “Evaluating time series forecasting models: an empirical study on performance estimation methods,” Machine Learning, vol. 109, no. 11, p. 1997, Oct. 2020, doi: 10.1007/s10994-020-05910-7.

C. Bergmeir, R. J. Hyndman, and B. Koo, “A note on the validity of cross-validation for evaluating autoregressive time series prediction,” Computational Statistics & Data Analysis, vol. 120, p. 70, Nov. 2017, doi: 10.1016/j.csda.2017.11.003.

C. Rao, T. Xue, M. Kan, P. Zhou, and Y. Lan, “A new hybrid neural network framework inspired by biological systems for advanced financial forecasting,” Scientific Reports, vol. 15, no. 1, p. 36985, Oct. 2025, doi: 10.1038/s41598-025-21842-5.

T. Berger, “On the information content of explainable artificial intelligence for quantitative approaches in finance,” OR Spectrum, vol. 47, no. 1, p. 177, Jun. 2024, doi: 10.1007/s00291-024-00769-9.

S. Kumar, “Explainable AI in Financial Forecasting Using Time Series Analysis,” International Journal for Research in Applied Science and Engineering Technology, vol. 13, no. 4, p. 7155, Apr. 2025, doi: 10.22214/ijraset.2025.70080.




DOI: https://doi.org/10.47738/jads.v7i2.1244

Refbacks

  • There are currently no refbacks.


Barcode

Journal of Applied Data Sciences

ISSN:2723-6471 (Online)
Publisher:Bright Publisher
Website:http://bright-journal.org/JADS
Email:taqwa@amikompurwokerto.ac.id (principal contact)
  support@bright-journal.org (technical issues)

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0