Measuring the Driving Forces of Predictive Performance: Application to Credit Scoring

References

• Aas K, Jullum M, Løland A (2021) Explaining individual predictions when features are dependent: More accurate approximations to Shapley values. Artificial Intelligence 298:103502.Crossref, Google Scholar
• Bell L, Devanathan N, Boyd S (2024) Efficient Shapley performance attribution for least-squares regression. Statist. Comput. 34(5):149.Crossref, Google Scholar
• Bergstra J, Bengio Y (2012) Random search for hyper-parameter optimization. J. Machine Learn. Res. 12:281–305.Google Scholar
• Borup D, Coulombe G, Rapach DE, Schütte ECM, Schwenk-Nebbe S (2022) The anatomy of out-of-sample forecasting accuracy. Federal Reserve Bank of Atlanta, Atlanta.Google Scholar
• Bowen D, Ungar L (2020) Generalized shap: Generating multiple types of explanations in machine learning. Preprint, submitted June 12, https://arxiv.org/abs/2006.07155.Google Scholar
• Breiman L (2001) Random forests. Machine Learn. 45(1):5–32.Crossref, Google Scholar
• Candes E, Fan Y, Janson L, Lv J (2018) Panning for gold: ‘Model-x’ knockoffs for high dimensional controlled variable selection. J. Roy. Statist. Soc. Ser. B: Statist. Methodology 80(3):551–577.Crossref, Google Scholar
• Casalicchio G, Molnar C, Bischl B (2019) Visualizing the feature importance for black box models. Machine Learning and Knowledge Discovery in Databases (Springer International Publishing, Cham, Switzerland), 655–670.Crossref, Google Scholar
• Chen Y, Calabrese R, Martin-Barragan B (2024) Interpretable machine learning for imbalanced credit scoring data sets. Eur. J. Oper. Res. 312(1):357–372.Crossref, Google Scholar
• Covert I, Lundberg SM, Lee S-I (2020) Understanding global feature contributions with additive importance measures. Adv. Neural Inform. Processing Systems 33:17212–17223.Google Scholar
• Fryer D, Strümke I, Nguyen H (2021) Shapley values for feature selection: The good, the bad, and the axioms. IEEE Access 9:144352–144360.Crossref, Google Scholar
• Gelman A, Vehtari A (2021) What are the most important statistical ideas of the past 50 years? J. Amer. Statist. Assoc. 116(536):2087–2097.Crossref, Google Scholar
• Ghorbani A, Zou J (2019) Data Shapley: Equitable valuation of data for machine learning. Proc. Internat. Conf. Machine Learn. (PMLR, New York), 2242–2251.Google Scholar
• Giesecke K, Horel E, Jirachotkulthorn C (2025) AICO: Feature significance tests for supervised learning. Preprint, submitted June 29, https://arxiv.org/abs/2506.23396.Google Scholar
• Gromping U (2007a) Estimators of relative importance in linear regression based on variance decomposition. Amer. Statist. 61(2):139–147.Crossref, Google Scholar
• Gromping U (2007b) Relative importance for linear regression in R: The package relaimpo. J. Statist. Software 17(1):27.Google Scholar
• Gunnarsson BR, Vanden Broucke S, Baesens B, Óskarsdóttir M, Lemahieu W (2021) Deep learning for credit scoring: Do or don’t? Eur. J. Oper. Res. 295(1):292–305.Crossref, Google Scholar
• Horel E, Giesecke K (2022) Computationally efficient feature significance and importance for predictive models. Proc. 3rd ACM Internat. Conf. AI Finance (ACM, New York), 300–307.Google Scholar
• Huang Z (1997) Clustering large data sets with mixed numeric and categorical values. Proc. 1st Pacific-Asia Conf. Knowledge Discovery Data Mining (World Scientific, Singapore), 21–34.Google Scholar
• Hurlin C, Pérignon C, Saurin S (2026) The fairness of credit scoring models. Management Sci. 72(1):406–425.Link, Google Scholar
• Israeli O (2007) A Shapley-based decomposition of the R-square of a linear regression. J. Econom. Inequality 5(2):199–212.Crossref, Google Scholar
• Jiang C, Lu W, Wang Z, Ding Y (2023) Benchmarking state-of-the-art imbalanced data learning approaches for credit scoring. Expert Systems Appl. 213:118878.Crossref, Google Scholar
• Kumar IE, Venkatasubramanian S, Scheidegger C, Friedler SA (2020) Problems with Shapley-value-based explanations as feature importance measures. Proc. Internat. Conf. Machine Learn. 119:5491–5500.Google Scholar
• Lei J, G’Sell M, Rinaldo A, Tibshirani RJ, Wasserman L (2018) Distribution-free predictive inference for regression. J. Amer. Statist. Assoc. 113(523):1094–1111.Crossref, Google Scholar
• Lindeman RH, Merenda PF, Gold RZ (1980) Introduction to Bivariate and Multivariate Analysis, vol. 4 (Scott, Foresman Glenview, IL).Google Scholar
• Lipovetsky S, Conklin M (2001) Analysis of regression in game theory approach. Appl. Stochastic Models Bus. Industry 17(4):319–330.Crossref, Google Scholar
• Liu Y, Ročková V (2023) Variable selection via Thompson sampling. J. Amer. Statist. Assoc. 118(541):287–304.Crossref, Google Scholar
• Lundberg SM, Lee S-I (2017) A unified approach to interpreting model predictions. Adv. Neural Inform. Processing Systems 30:4768–4777.Google Scholar
• Lundberg SM, Erion GG, Lee S-I (2018) Consistent individualized feature attribution for tree ensembles. Preprint, submitted February 12, https://arxiv.org/abs/1802.03888.Google Scholar
• Moehle N, Boyd S, Ang A (2021) Portfolio performance attribution via Shapley value. Preprint, submitted February 11, https://arxiv.org/abs/2102.05799.Google Scholar
• Molnar C (2025) Interpretable Machine Learning. Accessed July 31, 2025, https://christophm.github.io/interpretable-ml-book/.Google Scholar
• Murdoch WJ, Singh C, Kumbier K, Abbasi-Asl R, Yu B (2019) Definitions, methods, and applications in interpretable machine learning. Proc. Natl. Acad. Sci. USA 116(44):22071–22080.Crossref, Google Scholar
• Owen AB, Prieur C (2017) On Shapley value for measuring importance of dependent inputs. SIAM/ASA J. Uncertainty Quant. 5(1):986–1002.Crossref, Google Scholar
• Park H-S, Jun C-H (2009) A simple and fast algorithm for k-medoids clustering. Expert Systems Appl. 36(2):3336–3341.Crossref, Google Scholar
• Pratt JW (1987) Dividing the indivisible: Using simple symmetry to partition variance explained. Proc. 2nd Internat. Tampere Conf. Statist., 245–260.Google Scholar
• Redell N (2019) Shapley decomposition of R-squared in machine learning models. Preprint, submitted August 26, https://arxiv.org/abs/1908.09718.Google Scholar
• Romano Y, Patterson E, Candès EJ (2019) Conformalized quantile regression. Proc. 33rd Internat. Conf. Neural Inform. Processing Systems 32:3543–3553.Google Scholar
• Shapley L (1953) A value for n-person games. Contributions Theory Games 2:307–317.Google Scholar
• Strumbelj E, Kononenko I (2010) An efficient explanation of individual classifications using game theory. J. Machine Learn. Res. 11:1–18.Google Scholar
• Sun Y, Wang L (2021) Stochastic tree search for estimating optimal dynamic treatment regimes. J. Amer. Statist. Assoc. 116(533):421–432.Crossref, Google Scholar
• Sundararajan M, Najmi A (2020) The many Shapley values for model explanation. Proc. 37th Internat. Conf. Machine Learn. (PMLR, New York), 9269–9278.Google Scholar
• Sundararajan M, Dhamdhere K, Agarwal A (2020) The Shapley Taylor interaction index. Proc. Internat. Conf. Machine Learn. (PMLR, New York), 9259–9268.Google Scholar
• Sundararajan M, Taly A, Yan Q (2017) Axiomatic attribution for deep networks. Proc. Internat. Conf. Machine Learn. (PMLR, New York), 3319–3328.Google Scholar
• Sutera A, Louppe G, Huynh-Thu VA, Wehenkel L, Geurts P (2021) From global to local MDI variable importances for random forests and when they are Shapley values. Adv. Neural Inform. Processing Systems 34:3533–3543.Google Scholar
• Verdinelli I, Wasserman L (2024) Feature importance: A closer look at Shapley values and loco. Statist. Sci. 39(4):623–636.Crossref, Google Scholar
• Visani G, Bagli E, Chesani F, Poluzzi A, Capuzzo D (2021) Statistical stability indices for lime: Obtaining reliable explanations for machine learning models. J. Oper. Res. Soc. 73(1):91–101.Crossref, Google Scholar
• Vovk V, Gammerman A, Shafer G (2005) Algorithmic Learning in a Random World (Springer, New York).Google Scholar
• Williamson B, Feng J (2020) Efficient nonparametric statistical inference on population feature importance using Shapley values. Proc. Internat. Conf. Machine Learn. (PMLR, New York), 10282–10291.Google Scholar
• Zhang H, Singh H, Ghassemi M, Joshi S (2023) “Why did the model fail?”: Attributing model performance changes to distribution shifts. Proc. 40th Internat. Conf. Machine Learn. (PMLR, New York), 41550–41578.Google Scholar
• Zhao Q, Hastie T (2021) Causal interpretations of black-box models. J. Bus. Econom. Statist. 39(1):272–281.Crossref, Google Scholar