Pith. sign in

REVIEW 3 major objections 6 minor 59 references

Financial Data Analysis with Robust Federated Logistic Regression

T0 review · 3 major / 6 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Federated logistic regression with median or trimmed-mean aggregation matches centralized logistic regression, decision trees, and k-nearest neighbors in AUC on four financial datasets, retaining that performance when a fifth of clients…

desk verdict Known robust federated aggregation rules applied to logistic regression for finance; useful benchmark, but the claim of comparability to DT/KNN is contradicted by the paper's own LoanPred results and Eq (5) has a real mathematical error. read the letter →

arxiv 2504.20250 v1 pith:HTPIC2OQ submitted 2025-04-28 cs.LG q-fin.GNq-fin.STstat.APstat.ML

Kun Yang , Nikhil Krishnan , Sanjeev R. Kulkarni This is my paper · ORCID
classification cs.LGq-fin.GNq-fin.STstat.APstat.ML
keywords FederatedLearningLogisticRegressionRobustAggregationFinancialDataAnalysisOutlierRobustnessNon-IIDAUCInterpretability
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper aims to show that financial classification can be done well without ever pooling raw customer data. It proposes a federated logistic regression framework in which each client trains a local linear model and sends only coefficient vectors to a central server. Across four public datasets covering loan default, deposit subscription, credit scoring, and credit risk, the authors report AUC comparable to centralized logistic regression, decision tree, and k-nearest-neighbor baselines for both binary and multi-class tasks. They further claim that replacing the usual average of client updates with a coordinate-wise median or a trimmed mean keeps that AUC stable when up to 20% of clients are corrupted, while plain averaging fails. If these results hold, a simple and interpretable linear model can deliver privacy, outlier robustness, and centralized-level accuracy on the same financial problems.

What carries the argument

The load-bearing mechanism is server-side aggregation of client coefficient vectors. After local logistic-regression updates, the server either averages the per-coordinate values, takes their per-coordinate median, or removes the top and bottom $\alpha$ fraction of each coordinate before averaging (trimmed mean). In the robust variants, one corrupted local model cannot shift the global model the way it shifts a mean, and the paper sets $\alpha \ge p_{\mathrm{out}}/2$ so that trimming removes at least the corrupted updates. For multi-class problems the same machinery is repeated for each one-vs-rest binary classifier, and the client only transmits $C$ coefficient vectors per round.

What would settle it

On the BankMarketing dataset with $M=100$ clients, sample size $s=100$, and 10% of clients corrupted, run FLR-trim_mean with a fixed trimming fraction $\alpha=0.05$ instead of one chosen from the true corruption rate. If AUC drops by more than about 0.02 below the no-outlier value of 0.83, the paper's rule $\alpha \ge p_{\mathrm{out}}/2$ is load-bearing and the practical robustness claim is only conditional.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that a straightforward federated linear model is enough to reproduce centralized performance on these financial benchmarks. With no outliers, all three aggregation strategies (mean, median, trimmed mean) give nearly identical results, matching centralized logistic regression's AUC on BankMarketing, LoanPred, and CreditScore within 0.01 and on CreditRisk within 0.02, under both IID and non-IID data splits. When outliers are injected, mean aggregation degrades as the corruption rate grows, while median and trimmed-mean aggregation keep AUC largely unchanged up to $p_{\mathrm{out}} = 0.2$, with median slightly ahead. The multi-class version trains one-vs-rest binary models in parallel, so communication grows linearly with the number of classes. The coefficients themselves serve as interpretable feature-importance reports.

Load-bearing premise

The trimmed-mean robustness result assumes the server knows roughly how many clients are corrupt, because the trimming fraction is set from that number; without that knowledge the reported outlier resilience is not established, although the median-aggregation results do not need that assumption.

Editorial extensions

If this is right

  • Financial institutions can train a classification model with centralized-level AUC while raw customer data never leaves local devices; only coefficient vectors cross the network.
  • Median aggregation is a simple defense against outlier clients: it preserves AUC up to 20% corruption without requiring the server to know the exact corruption rate.
  • Multi-class credit-scoring tasks fit the same framework by training one-vs-rest logistic models in parallel, at a communication cost that grows linearly with class count.
  • Coefficient-based feature importance is available directly from the federated model, so decisions can be explained even though training is distributed.
  • Non-IID client data degrades performance only slightly relative to IID data, so the method does not depend on a uniform partition of the population.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the reported experiments, a natural stress test is to run the same protocol on raw, unfiltered financial features; the paper first removes highly correlated features, and that preprocessing may be hiding part of the advantage non-linear centralized models would have.
  • Because the trimmed-mean rule requires knowing the corruption fraction, an untested practical variant is to set the trimming fraction adaptively from the spread of client updates; the reported median results suggest per-coordinate median may already give this robustness without tuning.
  • The protocol's small payload (one coefficient vector per class per round) suggests it could scale to much larger client populations than the 200-client maximum tested here, provided communication is the bottleneck.
  • Rerunning the released code with a temporal split—training on earlier months and testing on later ones—would indicate whether the centralized-level AUC is a property of the datasets or of the method itself.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper proposes a federated logistic regression (FLR) framework for financial data analysis, with three server-side aggregation rules: mean, coordinate-wise median, and trimmed mean. The method is evaluated on four public financial datasets (two binary, two multi-class) under IID and non-IID client partitions, both without outliers and with simulated outlier clients. The results are compared against centralized logistic regression (LR), decision trees (DT), and k-nearest neighbors (KNN). The central claim is that FLR achieves comparable classification performance to these centralized algorithms and remains robust to outliers when median or trimmed-mean aggregation is used. The paper also provides feature-importance analysis and makes code and datasets publicly available.

Significance. If the performance-parity claim were supported, the paper would provide a useful practical result: a simple, interpretable linear model can be trained in a federated fashion without a major accuracy loss relative to centralized non-linear models, while robust aggregation offers protection against malicious clients. The experimental design is straightforward and the reported metrics (ACC, F1, AUC with means and 1.96-sigma intervals) are appropriate. The authors also commit to releasing code and data, which is commendable for reproducibility. However, the paper's headline claim is contradicted by its own experimental results on one of the two binary datasets, and the mathematical formulation of the likelihood contains a serious error. These issues prevent the stated significance from being realized without substantial revision.

major comments (3)
  1. [Abstract and Conclusion vs. Tables 3–4] The abstract and conclusion claim 'comparable performance' to LR, DT, and KNN in both binary and multi-class tasks. This is contradicted by the paper's own Tables 3 and 4 on the LoanPred dataset. For LoanPred, all FLR variants in Table 4 achieve AUC 0.53 ± 0.01 under both IID and non-IID partitions, while centralized DT and KNN in Table 3 achieve 0.85 ± 0.00 and 0.87 ± 0.01, respectively. This is a 0.32–0.34 gap in AUC, far outside the reported confidence intervals. Since the claim is unqualified and covers both binary datasets, the failure on one of them is decisive. The text in Section 5.3 that explicitly states 'our proposed framework can achieve comparable performance in terms of AUC with the baseline performance obtained by LR, DT and KNN' is therefore not supported by the reported results.
  2. [Section 3.2.1, Eqs. (5)–(6)] The log-likelihood expression in Eq. (5) is mathematically incorrect. For independent Bernoulli outcomes, the log-likelihood should be ln L = Σ_i [Y_i ln p_i + (1 − Y_i) ln(1 − p_i)]. The paper instead writes ln(L) = Σ_i [Y_i p + (1 − Y_i)(1 − p)], omitting the logarithms and using a single scalar p without an index. Moreover, Eq. (6) sets up an 'argmin' of the log-likelihood, whereas maximum likelihood estimation maximizes it. Although the implementation described later uses the cross-entropy loss, the stated derivation is invalid and should be corrected or removed.
  3. [Section 3.2.1 and Section 5.3 (Federated LR with Outliers)] The robustness demonstration for FLR-trim_mean depends on knowledge of the outlier fraction. Section 3.2.1 states that the trimming fraction α should satisfy α ≥ pout/2, where pout is the percentage of adversarial clients. In the outlier experiments of Section 5.3, pout is varied and the trimming fraction is evidently chosen using the known pout. The paper provides no procedure for estimating pout in practice, nor does it report the exact α values used. Consequently, the robustness result for the trimmed-mean variant is only established under an oracle assumption about the attack strength. The FLR-median results do not have this limitation, but the paper's general robustness claim covers both robust aggregation methods without qualification.
minor comments (6)
  1. [Abstract] The phrase 'Logistical Regression' should be 'Logistic Regression'.
  2. [Section 4.2] The equation label 'AUCmarco' contains a typo; it should be 'AUCmacro'.
  3. [Section 5.2] The text says the outlier percentage pout is varied 'from 0 to 0.2 with a step size of 0.5'; the step size should presumably be 0.05 to cover the stated range.
  4. [Section 5.2] The paper states that each experiment is conducted '10 times' using seeds 'spanning from 0 to 1000, incremented by 100', which yields 11 seeds. Please align the number of runs with the seed list.
  5. [Algorithm 1] The pseudocode indentation is unclear: the 'if t % 10 == 0' block appears to enclose the client selection, initialization, and server update, implying updates occur only every 10 iterations. This contradicts the textual description in Section 3.2.1, where the server updates in every iteration. Please redraw the algorithm with explicit indentation.
  6. [Throughout] There are several typographical and grammatical errors, e.g., 'dta' for 'data', 'we emphasis on' for 'we emphasize', and 'the number of iterationsTc' in Algorithm 1 is not used in the pseudocode. A careful proofread is needed.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the paper's central claims are empirical comparisons against external centralized baselines, not predictions derived from their own inputs.

full rationale

The paper contains no first-principles derivation whose output is equivalent to its input. FLR is standard logistic regression trained under a federated aggregation scheme; its no-outlier results (Table 4) match centralized LR (Table 3) because the two procedures are the same model trained on the same data, which is a sanity check rather than a circular prediction. The comparison against Decision Tree and K-Nearest Neighbors is an external benchmark on public datasets, so the central 'comparable performance' claim is empirically checkable rather than constructed from the method's own assumptions. The robust aggregation section defines the trimmed mean with α ≥ p_out/2 (Section 3.2.1), and experiments are run with known outlier fractions; this is a limitation of the robustness demonstration, since the server must know p_out, but it is a stated experimental precondition, not a fitted parameter renamed as a prediction. The only self-citation by the authors ([58], on greedy centroid initialization for federated k-means) appears in the Discussion as a pointer to future initialization work and is not load-bearing for any reported result. The paper does contain substantive correctness problems, notably the incorrect log-likelihood in Eq. (5) and the Table 4 LoanPred AUC of 0.53 vs. centralized DT/KNN AUC of 0.85/0.87, which contradicts the headline 'comparable to DT and KNN' claim; however, those are internal-consistency and correctness issues, not circularity. No step in the claimed derivation chain reduces to its own inputs.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The method relies on standard logistic regression and robust averaging; no new free entities. The main hidden knob is the trimmed fraction alpha, which is chosen using the outlier fraction; other hyperparameters are standard defaults.

free parameters (5)
  • trimmed fraction alpha = not reported (depends on pout)
    Used by FLR-trim_mean; rule alpha >= pout/2 in Section 3.2.1, but the exact value in experiments is not stated. The robustness results for trimmed mean depend on this choice.
  • learning rate eta = not reported
    Set fixed and same across clients (Section 3.2.1: 'we use the same and fixed eta in this work'). Value not given; affects convergence.
  • L2 regularization strength C = scikit-learn default (C=1.0)
    ℓ2 penalty on loss (Section 3.2.1); experiments use scikit-learn defaults, so regularization strength is a chosen hyperparameter.
  • local update iterations Tc = 10
    Number of local parameter updates per communication round (Section 5.2 Algorithm Setting). Hand-selected.
  • communication rounds T = 100
    Maximum server-client iterations (Section 5.2). Hand-selected.
assumptions (5)
  • standard math Data points are treated as independent for maximum likelihood estimation.
    Used to factor the likelihood in Eq (4).
  • domain assumption The logistic regression assumptions (linearity of logit, no multicollinearity, sufficient sample size) are required for the model to be valid.
    Section 3.3 lists them; Section 5.3 shows linearity fails for all BankMarketing features, yet the model is still evaluated.
  • ad hoc to paper Adversarial clients can be simulated as partitions/devices that send corrupted parameters, and the server knows or can estimate pout to set alpha.
    Section 5.2 With Outliers and Section 3.2.1; the trimmed mean needs alpha >= pout/2, so the robustness result assumes this knowledge.
  • domain assumption The federated algorithm converges to a usable global model without formal convergence guarantees.
    No convergence analysis is provided; the paper stops after T=100 rounds (Section 5.2).
  • domain assumption Global standardization can be computed from client statistics without violating privacy.
    Algorithm 1 lines 7-8 compute all clients' mu and sigma at the server; the paper does not discuss the privacy cost of sharing these statistics.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Financial Data Analysis with Robust Federated Logistic Regression." pith.science (2026). https://pith.science/paper/HTPIC2OQ

@misc{pith2026250420250,
  author       = {Pith},
  title        = {Pith review of: Financial Data Analysis with Robust Federated Logistic Regression},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HTPIC2OQ}},
  note         = {Machine review of arXiv:2504.20250}
}
read the original abstract

In this study, we focus on the analysis of financial data in a federated setting, wherein data is distributed across multiple clients or locations, and the raw data never leaves the local devices. Our primary focus is not only on the development of efficient learning frameworks (for protecting user data privacy) in the field of federated learning but also on the importance of designing models that are easier to interpret. In addition, we care about the robustness of the framework to outliers. To achieve these goals, we propose a robust federated logistic regression-based framework that strives to strike a balance between these goals. To verify the feasibility of our proposed framework, we carefully evaluate its performance not only on independently identically distributed (IID) data but also on non-IID data, especially in scenarios involving outliers. Extensive numerical results collected from multiple public datasets demonstrate that our proposed method can achieve comparable performance to those of classical centralized algorithms, such as Logistical Regression, Decision Tree, and K-Nearest Neighbors, in both binary and multi-class classification tasks.

Figures

Figures reproduced from arXiv: 2504.20250 by the authors.

Figure 1
Figure 1. Feature correlation of CreditRisk after iteratively removing the features with higher VIFs (>10). [PITH_FULL_IMAGE:figures/full_fig_p010_1.png] view at source ↗
Figure 2
Figure 2. Different cases on non-IID data. Feature Importance Report Here, we analyze the feature importance obtained by our proposed framework with￾out outliers. More specifically, we focus on analyzing the coefficients of each feature obtained by our proposed FLR framework in a multi-class classification task. According to [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Feature importances obtained by FLR with median aggregation on CreditScore with non-IID data, but no [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

59 extracted references · 55 canonical work pages

  1. [1]

    Ahmed, M

    S. Ahmed, M. M. Alshater, A. El Ammari, H. Hammami, Artificial intelligence and machine learning in finance: A bibliometric review, Research in International Business and Finance 61 (2022) 101646

    work page 2022

  2. [2]

    S. K. Sahu, A. Mokhade, N. D. Bokde, An overview of machine learning, deep learning, and reinforcement learning-based techniques in quantitative finance: recent progress and challenges, Applied Sciences 13 (3) (2023) 1956

    work page 2023

  3. [3]

    Akomea-Frimpong, D

    I. Akomea-Frimpong, D. Adeabah, D. Ofosu, E. J. Tenakwah, A review of studies on green finance of banks, research gaps and future directions, Journal of Sustainable Finance & Investment 12 (4) (2022) 1241–1264

    work page 2022

  4. [4]

    Weber, K

    P. Weber, K. V . Carl, O. Hinz, Applications of explainable artificial intelligence in finance—a systematic review of finance, information systems, and computer science literature, Management Review Quarterly 74 (2) (2024) 867–907

    work page 2024

  5. [5]

    J. W. Goodell, S. Kumar, W. M. Lim, D. Pattnaik, Artificial intelligence and machine learning in finance: Iden- tifying foundations, themes, and research clusters from bibliometric analysis, Journal of Behavioral and Experi- mental Finance 32 (2021) 100577

    work page 2021

  6. [6]

    Malhotra, D

    R. Malhotra, D. Malhotra, K. Malhotra, Predicting credit outlook of banking and nonbanking finance companies: A comparative analysis of machine learning models., Journal of Financial Data Science 6 (3) (2024)

    work page 2024

  7. [7]

    S. N. Hojaji, M. Yahyazadehfar, B. Abedin, Machine learning in behavioral finance: A systematic literature review., Journal of Financial Data Science 4 (3) (2022)

    work page 2022

  8. [8]

    I. H. Sarker, Machine learning: Algorithms, real-world applications and research directions, SN computer science 2 (3) (2021) 160

    work page 2021

Show all 59 references
  1. [9]

    D. Xu, Y . Tian, A comprehensive survey of clustering algorithms, Annals of Data Science 2 (2) (2015) 165–193

    2015

  2. [10]

    E. Min, X. Guo, Q. Liu, G. Zhang, J. Cui, J. Long, A survey of clustering with deep learning: From the perspec- tive of network architecture, IEEE Access 6 (2018) 39501–39514

    2018

  3. [11]

    A. M. Ozbayoglu, M. U. Gudelek, O. B. Sezer, Deep learning for financial applications: A survey, Applied Soft Computing 93 (2020) 106384

    2020

  4. [12]

    Cao, Ai in finance: challenges, techniques, and opportunities, ACM Computing Surveys (CSUR) 55 (3) (2022) 1–38

    L. Cao, Ai in finance: challenges, techniques, and opportunities, ACM Computing Surveys (CSUR) 55 (3) (2022) 1–38

    2022

  5. [13]

    Konecny, H

    J. Konecny, H. B. McMahan, F. X. Yu, P. Richtarik, A. T. Suresh, D. Bacon, Federated learning: Strategies for improving communication efficiency, in: Proc. Adv. Neural Inf. Proc. Syst., NeurIPS, Barcelona, Spain, 2016

    2016

  6. [14]

    J. Liu, J. Huang, Y . Zhou, X. Li, S. Ji, H. Xiong, D. Dou, From distributed machine learning to federated learning: A survey, Knowledge and Information Systems (2022) 1–33

    2022

  7. [15]

    Mothukuri, R

    V . Mothukuri, R. M. Parizi, S. Pouriyeh, Y . Huang, A. Dehghantanha, G. Srivastava, A survey on security and privacy of federated learning, Future Generation Computer Systems 115 (2021) 619–640

    2021

  8. [16]

    Burkart, M

    N. Burkart, M. F. Huber, A survey on the explainability of supervised machine learning, Journal of Artificial Intelligence Research 70 (2021) 245–317. 14 Financial Data Analysis with Robust Federated Logistic Regression

    2021

  9. [17]

    GDPR, General data protection regulation, https://gdpr.eu/, accessed: 2023-09-01 (2023)

    2023

  10. [18]

    HIPAA, The health insurance portability and accountability act, https://www.hhs.gov/hipaa/index.html, accessed: 2023-09-01 (2023)

    2023

  11. [19]

    Bhatore, L

    S. Bhatore, L. Mohan, Y . R. Reddy, Machine learning techniques for credit risk evaluation: a systematic literature review, Journal of Banking and Financial Technology 4 (2020) 111–138

    2020

  12. [20]

    Dixon, I

    M. Dixon, I. Halperin, P. Bilokon, Machine Learning in Finance: From Theory to Practice, Springer Cham, 2020. URL https://link.springer.com/book/10.1007/978-3-030-41068-1

    2020 doi

  13. [21]

    Kurani, A

    A. Kurani, A. Vakharia, M. Shah, A comprehensive comparative study of artificial neural network (ann) and support vector machines (svm) on stock forecasting, Annals of Data Science 10 (2023) 183–208

    2023

  14. [22]

    J. R. Coakley, C. E. Brown, Artificial neural networks in accounting and finance: modeling issues, Intelligent Systems in Accounting, Finance and Management 9 (2) (2000) 119–144. doi:https://doi.org/10.1002/ 1099-1174(200006)9:2<119::AID-ISAF182>3.0.CO;2-Y . URL https://online...

    2000 doi

  15. [23]

    Hambly, R

    B. Hambly, R. Xu, H. Yang, Recent advances in reinforcement learning in finance, Mathematical Finance 33 (3) (2023) 437–503. arXiv:https://onlinelibrary.wiley.com/doi/pdf/10.1111/mafi.12382, doi:https://doi.org/10.1111/mafi.12382. URL https://onlinelibrary.wiley.com/doi/abs/10...

    2023 doi

  16. [24]

    Moscato, A

    V . Moscato, A. Picariello, G. Sperlí, A benchmark of machine learning approaches for credit score prediction, Expert Systems with Applications 165 (2021) 113986

    2021

  17. [25]

    A. K. Jain, Data clustering: 50 years beyond k-means, Pattern recognition letters 31 (8) (2010) 651–666

    2010

  18. [26]

    Agarwal, M

    P. Agarwal, M. A. Alam, R. Biswas, Issues, challenges and tools of clustering algorithms, International Journal of Computer Science Issues (IJCSI) 8 (3) (2011) 523

    2011

  19. [27]

    Kant, Data center evolution: A tutorial on state of the art, issues, and challenges, Computer Networks 53 (17) (2009) 2939–2965

    K. Kant, Data center evolution: A tutorial on state of the art, issues, and challenges, Computer Networks 53 (17) (2009) 2939–2965

    2009

  20. [28]

    Q. Yang, Y . Liu, T. Chen, Y . Tong, Federated machine learning: Concept and applications, ACM Transactions on Intelligent Systems and Technology (TIST) 10 (2) (2019) 1–19

    2019

  21. [29]

    G. Long, Y . Tan, J. Jiang, C. Zhang, Federated learning for open banking, in: Federated Learning: Privacy and Incentive, Springer, 2020, pp. 240–254

    2020

  22. [30]

    Imteaj, M

    A. Imteaj, M. H. Amini, Leveraging asynchronous federated learning to predict customers financial distress, Intelligent Systems with Applications 14 (2022) 200064

    2022

  23. [31]

    C. M. Lee, J. Delgado Fernandez, S. Potenciano Menci, A. Rieger, G. Fridgen, Federated learning for credit risk assessment, in: Proceedings of the 56th Hawaii International Conference on System Sciences, ICSS, 2023, p. 10

    2023

  24. [32]

    Shingi, A federated learning based approach for loan defaults prediction, in: 2020 International Conference on Data Mining Workshops (ICDMW), IEEE, IEEE, 2020, pp

    G. Shingi, A federated learning based approach for loan defaults prediction, in: 2020 International Conference on Data Mining Workshops (ICDMW), IEEE, IEEE, 2020, pp. 362–368

    2020

  25. [33]

    Belle, I

    V . Belle, I. Papantonis, Principles and practice of explainable machine learning, Frontiers in big Data (2021) 39

    2021

  26. [34]

    Blanco-Justicia, J

    A. Blanco-Justicia, J. Domingo-Ferrer, Machine learning explainability through comprehensible decision trees, in: Machine Learning and Knowledge Extraction: Third IFIP TC 5, TC 12, WG 8.4, WG 8.9, WG 12.9 In- ternational Cross-Domain Conference, CD-MAKE 2019, Canterbury, UK, A...

    2019

  27. [35]

    Maalouf, Logistic regression in data analysis: an overview, International Journal of Data Analysis Techniques and Strategies 3 (3) (2011) 281–299

    M. Maalouf, Logistic regression in data analysis: an overview, International Journal of Data Analysis Techniques and Strategies 3 (3) (2011) 281–299

    2011

  28. [36]

    S. A. Czepiel, Maximum likelihood estimation of logistic regression models: theory and implementation, Avail- able at czep. net/stat/mlelr. pdf 83 (2002)

    2002

  29. [37]

    Sutskever, J

    I. Sutskever, J. Martens, G. Dahl, G. Hinton, On the importance of initialization and momentum in deep learning, in: International conference on machine learning, PMLR, PMLR, 2013, pp. 1139–1147

    2013

  30. [38]

    J. W. Tukey, et al., Exploratory data analysis, V ol. 2, Reading, MA, 1977

    1977

  31. [39]

    L. Liu, H. Jiang, P. He, W. Chen, X. Liu, J. Gao, J. Han, On the variance of the adaptive learning rate and beyond, arXiv preprint arXiv:1908.03265 (2019)

    2019 arXiv

  32. [40]

    Galar, A

    M. Galar, A. Fernández, E. Barrenechea, H. Bustince, F. Herrera, An overview of ensemble methods for binary classifiers in multi-class problems: Experimental study on one-vs-one and one-vs-all schemes, Pattern Recogni- tion 44 (8) (2011) 1761–1776. 15 Financial Data Analysis w...

    2011

  33. [41]

    J. K. Harris, Primer on binary logistic regression, Family Medicine and Community Health 9 (Suppl 1) (2021)

    2021

  34. [42]

    R. G. Long, The crux of the method: assumptions in ordinary least squares and logistic regression, Psychological reports 103 (2) (2008) 431–434

    2008

  35. [43]

    C.-Y . J. Peng, K. L. Lee, G. M. Ingersoll, An introduction to logistic regression analysis and reporting, The journal of educational research 96 (1) (2002) 3–14

    2002

  36. [44]

    Schreiber-Gregory, H

    D. Schreiber-Gregory, H. Jackson, K. Bader, Logistic and linear regression assumptions: Violation recognition and control, Henry M Jackson Foundation (2018)

    2018

  37. [45]

    Vittinghoff, C

    E. Vittinghoff, C. E. McCulloch, Relaxing the rule of ten events per variable in logistic and cox regression, American journal of epidemiology 165 (6) (2007) 710–718

    2007

  38. [46]

    Wikipedia, F-score, https://en.wikipedia.org/wiki/F-score, accessed: 2023-08-01 (2023)

    2023

  39. [47]

    Wikipedia, Auc, https://en.wikipedia.org/wiki/Receiver_operating_characteristic, accessed: 2023-08-01 (2023)

    2023

  40. [48]

    banking institution, Bank marketing, https://archive.ics.uci.edu/dataset/222/bank+marketing, accessed: 2023-08-01 (2014)

    P. banking institution, Bank marketing, https://archive.ics.uci.edu/dataset/222/bank+marketing, accessed: 2023-08-01 (2014)

    2014

  41. [49]

    Univ.AI, Loan prediction based on customer behavior, https://www.kaggle.com/datasets/subhamjain/ loan-prediction-based-on-customer-behavior , accessed: 2021-02-01 (202x)

    2021

  42. [50]

    global financial company, Credit score classification,https://www.kaggle.com/datasets/parisrohan/ credit-score-classification, accessed: 2021-02-01 (202x)

    U. global financial company, Credit score classification,https://www.kaggle.com/datasets/parisrohan/ credit-score-classification, accessed: 2021-02-01 (202x)

    2021

  43. [51]

    global financial company, Credit risk dataset, https://www.kaggle.com/datasets/ranadeep/ credit-risk-dataset, accessed: 2021-02-01 (202x)

    U. global financial company, Credit risk dataset, https://www.kaggle.com/datasets/ranadeep/ credit-risk-dataset, accessed: 2021-02-01 (202x)

    2021

  44. [52]

    R. M. O’brien, A caution regarding rules of thumb for variance inflation factors, Quality & quantity 41 (2007) 673–690

    2007

  45. [53]

    Fox, Applied Regression Analysis and Generalized Linear Models, Sage publications, 2016

    J. Fox, Applied Regression Analysis and Generalized Linear Models, Sage publications, 2016

    2016

  46. [54]

    B. Zhu, L. Wang, Q. Pang, S. Wang, J. Jiao, D. Song, M. I. Jordan, Byzantine-robust federated learning with optimal statistical rates, in: International Conference on Artificial Intelligence and Statistics, PMLR, PMLR, 2023, pp. 3151–3178

    2023

  47. [55]

    Pillutla, S

    K. Pillutla, S. M. Kakade, Z. Harchaoui, Robust aggregation for federated learning, IEEE Transactions on Signal Processing 70 (2022) 1142–1154

    2022

  48. [56]

    M. S. Jere, T. Farnan, F. Koushanfar, A taxonomy of attacks on federated learning, IEEE Security & Privacy 19 (2) (2020) 20–28

    2020

  49. [57]

    L. Yang, A. Shami, On hyperparameter optimization of machine learning algorithms: Theory and practice, Neu- rocomputing 415 (2020) 295–316

    2020

  50. [58]

    K. Yang, M. M. Amiri, S. R. Kulkarni, Greedy centroid initialization for federated k-means, in: 2023 57th Annual Conference on Information Sciences and Systems (CISS), IEEE, IEEE, 2023, pp. 1–6

    2023

  51. [59]

    loan_amnt

    Y . Liu, Y . Liu, Z. Liu, Y . Liang, C. Meng, J. Zhang, Y . Zheng, Federated forest, IEEE Transactions on Big Data 8 (3) (2020) 843–854. 16 Financial Data Analysis with Robust Federated Logistic Regression A Extra Results A.1 Multicollinearity on CreditRisk Figure A.1 illustra...

    2020

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.