Pith. sign in

REVIEW 3 major objections 5 minor 41 references

Explaining deep neural network models for electricity price forecasting with XAI

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Why day-ahead electricity price AI favors recency over curve matching

desk verdict A useful applied XAI study with a promising visualization toolkit, but the central market claims rest on an estimator-definition mismatch that should be resolved before the conclusions are cited. read the letter →

arxiv 2506.19894 v1 pith:P3H3CQ3L submitted 2025-06-24 cs.LG cs.AIstat.ML

Antoine Pesenti , Aidan OSullivan This is my paper · ORCID
classification cs.LGcs.AIstat.ML
keywords electricitypriceforecastingexplainableAISHAPSSHAPdeepneuralnetworksday-aheadmarketfeatureattributiondynamics
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

This paper sets out to show that explainable-AI attributions can expose what black-box neural networks have learned about wholesale electricity markets, turning accurate but opaque price forecasters into a source of market insight. Applying SHAP, gradient, and heatmap methods to deep networks for Germany, France, Belgium, Nord Pool, and ComEd, it claims that the five forecasters do not behave like naive curve matchers. The most recent price available is the most important variable in every market, and the models learn market-specific dynamics such as early-morning load driving later prices down in France, Belgium, and ComEd, and renewable generation dominating price moves in Germany. The paper introduces SSHAP values and SSHAP lines, super-variable aggregates of SHAP values, as a way to compress and visualize the high-dimensional 24-input-by-24-output attribution tables. The paper argues these explanations reveal genuine market fundamentals rather than arbitrary model quirks.

What carries the argument

The engine of the analysis is Shapley-value attribution and its aggregation. SHAP assigns each of the 120 to 241 input variables a per-output-hour contribution to each of the 24 forecast hours, with the efficiency property that the sum of all SHAP values equals the model prediction minus its mean. The paper defines a super-variable as any partition of the input variables, typically grouping the 24 hourly lags of one underlying series such as Price D-1 or Load Forecast D, and defines the SSHAP value of a super-variable as the sum of the SHAP values of its members; because the partition is exclusive and complete, efficiency survives the aggregation. The SSHAP line is then the Gaussian-kernel-weighted average of SSHAP values plotted against the actual price, so that if the model is well calibrated the summed lines approximate the identity line. Heatmaps of SHAP and gradient values, arranged as input-hour by output-hour tables, make the difference between column effects (recent hours influencing the whole day) and diagonal effects (same-hour matching) visible.

What would settle it

Retrain the same five architectures on price histories whose daily order is shuffled (or on a held-out period) and rerun the SSHAP analysis: if the 'last price available' dominance and the early-morning-load negative rectangle disappear or reverse, those explanations are artifacts of training-set memorization rather than market fundamentals. A second check is to recompute the attributions with many more Monte-Carlo samples or with exact conditional Shapley values on a subset of instances and see whether the reported SSHAP lines shift materially.

Watch

Extended reading notes

Core claim

The central claim is that the trained day-ahead price forecasters, once explained, reveal real and interpretable market behavior. Across all five markets, the most recent observed day-ahead price is the most important input, and the models prefer that recent information over a price-curve-matching strategy that would pair each forecast hour with the same hour from yesterday. The explanations also show distinct market personalities: Germany's prices are driven heavily by renewable generation forecasts, Nord Pool is comparatively simple with the previous day's price dominating, and France, Belgium, and ComEd all contain a learned early-morning load effect in which a high load forecast in the first hours pushes prices down for the rest of the day. In addition, the paper claims the SSHAP value---the sum of SHAP values over a partition of variables into super-variables---and the SSHAP line, a Gaussian-weighted average of those values against actual price, are new explanatory tools that retain Shapley efficiency while reducing the dimensionality of the attribution problem.

Load-bearing premise

The load-bearing premise is that the approximate SHAP values computed on the training sets faithfully reflect what the networks actually learned, so that the patterns they reveal are genuine market dynamics rather than estimation error or overfitting noise.

Editorial extensions

If this is right

  • In all five day-ahead markets, the most recent observed price is the strongest single predictor, with its influence concentrated on the early forecast hours and fading later in the day.
  • The models' preference for recency over same-hour curve matching means that 'last price available' style features deserve more weight in benchmark and production forecasters than the common naive same-hour benchmark suggests.
  • High early-morning load forecasts in France, Belgium, and ComEd are learned as a downward pressure on prices for the rest of the day, which is a concrete, market-specific mechanism worth testing outside the model.
  • Germany's model is renewable-dominated while Nord Pool's is price-dominated, so market complexity and generation mix are reflected in the attribution structure and in the proposed complexity measures.
  • SSHAP super-variable aggregation lets practitioners summarize 120 to 241 input by 24 output attributions into a small number of interpretable curves without losing the additive efficiency identity.

Reading between the lines

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

  • SSHAP lines could serve as a calibration diagnostic: where the summed lines deviate from the identity line, the model is systematically over- or under-predicting in that price band, pointing to where features or training data are missing.
  • The recency result suggests adding even fresher information---for example intraday prices or the latest load and renewable updates---could push day-ahead accuracy further, a hypothesis the paper does not test.
  • The early-morning-load signature in France, Belgium, and ComEd is a simple checkable rule: unusually high overnight load forecasts should imply lower afternoon prices, independently of the neural networks.
  • The SSHAP construction is generic for any tabular model with grouped inputs, though its usefulness depends on choosing partitions that correspond to meaningful underlying variables.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper trains the two-hidden-layer DNN day-ahead forecasting models from the epftoolbox benchmark for five electricity markets (Germany, France, Belgium, Nord Pool, ComEd) and explains the fitted models with SHAP values, gradients, and heatmaps. It introduces SSHAP values as sums of SHAP values over 24-hour super-variables and SSHAP lines as Gaussian-smoothed conditional averages of SSHAP values against actual price. On this basis, Sections 5 and 6 claim that the last available price is the most important input in all five markets, that the models prefer recency over a naive hour-matching 'curve' strategy, and that early-morning load forecasts push down later prices in France, Belgium, and ComEd.

Significance. If the explanations were computed with a valid conditional Shapley estimator on a representative evaluation set, the cross-market comparison would be a useful empirical contribution: it uses open benchmark data and models, covers five markets, and offers a clean visualization layer for high-dimensional tabular DNNs. The additive construction of SSHAP values (Eq. 2) is exact and convenient. However, the central market-behavior conclusions are not yet established because the estimator actually used appears to be the interventional permutation estimator rather than the conditional estimator claimed in Section 4.1. The paper's strengths are its reproducible data and models, the multi-market scope, and the new visualization tools; its main weakness is that the headline findings rest on an estimator-definition mismatch and on explanations computed on training data.

major comments (3)
  1. [Section 4.1; Section 6, first bullet] The paper claims to estimate conditional Shapley values but cites [41] and the SHAP library, whose model-agnostic permutation sampler is interventional: coalition complements are drawn from the empirical marginal distribution of the background data, not from the conditional distribution P(X_barS | X_S). For the strongly correlated hourly price and load inputs of these models, interventional SHAP can attribute the joint effect of a correlated block to the single feature most correlated with the output, which would mechanically produce the column effect and the 'last price available is the most important variable' conclusion reported in Sections 5 and 6. The manuscript does not state the SHAP library version, the number of Monte-Carlo samples, or the background dataset used, so there is no convergence or variance evidence. Please either implement a genuine conditional estimator (for example, Gaussian or nearest-neighbor conditional imputation) and show that the heatmaps are stable, or explicitly reframe all market conclusions as statements about interventional SHAP under an independence assumption, accompanied by a sensitivity analysis to the background distribution.
  2. [Section 3.2; Table 2] The models are explained on their training sets, and Table 2 reports much better errors than the benchmark test-set results in [35]. A model that overfits four years of historical data can exhibit attribution patterns that reflect training noise rather than market fundamentals, yet Section 6 offers these patterns as insights into how the markets work. The authors should either recompute the explanations on a held-out period (or on the original test set) and demonstrate that the same conclusions hold, or explicitly restrict the claims to properties of the trained models and discuss the training-to-test performance gap as a limitation.
  3. [Section 4.3; Eq. (3)] Equation (3) is presented as a prediction ('we can derive from Eq. (2) that the sum of the SSHAP lines should be almost a straight line with a slope of 1'). This is not an independent empirical test: it is an algebraic consequence of Eq. (2) whenever the SSHAP lines are weighted averages of SSHAP values over the same training instances and the model output is close to the target. The presentation should call Eq. (3) a self-consistency check of the efficiency property, not a prediction that validates the method.
minor comments (5)
  1. [Section 5.1.3] The sentence 'a complex offsetting interaction, between.' is incomplete; it should end with an explicit reference to the two renewable forecast variables being offset.
  2. [Section 4.3; Table 3] The Gaussian bandwidth of 5 used in the SSHAP line definition and the e0.5/MWh importance threshold in Table 3 are introduced without justification; please report sensitivity analyses for these hyperparameters.
  3. [Section 5.5.1] The 'extreme SHAP values of a few hundred dollars per MWh' are labeled abnormal but are then averaged; reporting medians and robust dispersion measures would clarify whether the average absolute values are driven by a small number of unstable estimates.
  4. [Section 2] The discussion of [33] notes a conflicting conclusion (prices three days earlier are most important for Nord Pool) but does not attempt to reconcile it with the present finding that Price D-1 dominates; a comparison paragraph would strengthen the paper.
  5. [Abstract] The dataset link in the abstract contains a space in the URL ('4624 805') and should be corrected.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation is present: SSHAP values are explicit additive aggregations of SHAP values, and the reported market insights are empirical explanations rather than quantities forced by construction or by self-citation.

full rationale

The paper's derivation chain is self-contained and presents no circular step. SSHAP values are introduced by definition as the sum of SHAP values over pre-specified super-variables (Section 4.3), so the efficiency identity in Eq. (2) follows directly from the SHAP efficiency property rather than from any fitted parameter; Eq. (3) is an explicit consistency check that assumes the model is approximately calibrated (average forecast close to actual price in each price band), not a prediction extracted from the same data. The market-level findings in Sections 5-6, such as the importance of the last available price and the French and Belgian early-morning load effect, are empirical readings of the SHAP and gradient heatmaps computed for externally provided benchmark models; no quantity reported as a finding is a re-labeling of a fitted parameter, and no inference is justified solely by a self-citation. The super-variable splits, for example Load Forecast D H0-H4 versus H5-H23, are motivated by observed gradient signs, but this is a presentation choice for a descriptive visualization and does not make the subsequent SSHAP line offsetting a construction artifact. A possible mismatch between the paper's stated "conditional expectation" and the Monte-Carlo estimator of [41] would be a methodological correctness concern about estimator validity, not a circularity of the derivation, and is therefore not scored here.

Assumptions & free parameters 3 free parameters · 5 assumptions · 2 invented entities

The central analysis rests on standard Shapley additivity, plus a set of unverified practical choices: the SHAP library's conditional sampling, data-dependent super-variable splits, and a hand-chosen kernel bandwidth. These choices are not predictions; they shape what the explanations claim about the markets.

free parameters (3)
  • Gaussian kernel bandwidth = 5 (units unspecified)
    Used for SSHAP lines in Section 4.3; chosen by hand with no sensitivity analysis or unit specification for different markets/currencies.
  • Important-variable threshold = 0.5 EUR/MWh
    Used in Table 3 to count 'important variables per hour'; arbitrary threshold applied uniformly across markets.
  • Super-variable split points = H5 for France/Belgium load, H6 for ComEd load
    Chosen after observing gradient heatmap sign changes (Sections 5.2.2, 5.3.2, 5.5.2); data-dependent and used to support the early-morning-load conclusions.
assumptions (5)
  • standard math Shapley value efficiency and additivity (Eq. 1)
    The SSHAP formulation in Section 4.3 relies on the efficiency property; this is standard cooperative game theory from [38].
  • domain assumption The DNN models trained per [35] are representative of good EPF models.
    Section 5 states [35] 'has shown that this model outperforms other approaches'; the paper does not revalidate on test data.
  • domain assumption Monte-Carlo sampling from the SHAP library yields accurate conditional Shapley values.
    Section 4.1 states conditional expectation is used and estimated by the method of [41], but no convergence checks, sample counts, or error bars are reported.
  • domain assumption The datasets (prices, load forecasts, generation forecasts) from epftoolbox are reliable and complete.
    Section 3.1 uses the public benchmark data without independent verification; this is standard for benchmark studies.
  • ad hoc to paper Gaussian smoothing with bandwidth 5 produces a faithful representation of the SSHAP-price relationship.
    Section 4.3 defines SSHAP lines with this kernel; the choice is not justified and is applied across different price scales (Euros vs Dollars).
invented entities (2)
  • SSHAP values
    purpose: Summarize SHAP contributions of a group of hourly inputs (super-variable) into one number, reducing dimensionality.
    Defined in Section 4.3 as sums of SHAP values; purely descriptive, no falsifiable prediction outside the model.
  • SSHAP lines
    purpose: Visualize the dependence of average SSHAP values on actual price using Gaussian smoothing.
    Defined in Section 4.3 with bandwidth 5; a visualization tool, no external benchmark.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Explaining deep neural network models for electricity price forecasting with XAI." pith.science (2026). https://pith.science/paper/P3H3CQ3L

@misc{pith2026250619894,
  author       = {Pith},
  title        = {Pith review of: Explaining deep neural network models for electricity price forecasting with XAI},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/P3H3CQ3L}},
  note         = {Machine review of arXiv:2506.19894}
}
read the original abstract

Electricity markets are highly complex, involving lots of interactions and complex dependencies that make it hard to understand the inner workings of the market and what is driving prices. Econometric methods have been developed for this, white-box models, however, they are not as powerful as deep neural network models (DNN). In this paper, we use a DNN to forecast the price and then use XAI methods to understand the factors driving the price dynamics in the market. The objective is to increase our understanding of how different electricity markets work. To do that, we apply explainable methods such as SHAP and Gradient, combined with visual techniques like heatmaps (saliency maps) to analyse the behaviour and contributions of various features across five electricity markets. We introduce the novel concepts of SSHAP values and SSHAP lines to enhance the complex representation of high-dimensional tabular models.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 26 canonical work pages

  1. [41]

    Explaining prediction models and individual pre- dictions with feature contributions

    Štrumbelj E, Kononenko I. Explaining prediction models and individual pre- dictions with feature contributions. Knowl Inf Syst 2014;41(3):647–65. http:// dx.doi.org/10.1007/s10115-013-0679-x, URL http://link.springer.com/10.1007/ s10115-013-0679-x

    doi:10.1007/s10115-013-0679-x 2014

  2. [35]

    Forecasting day-ahead elec- tricity prices: A review of state-of-the-art algorithms, best practices and an open-access benchmark

    Lago J, Marcjasz G, De Schutter B, Weron R. Forecasting day-ahead elec- tricity prices: A review of state-of-the-art algorithms, best practices and an open-access benchmark. Appl Energy 2021;293:116983. http://dx.doi.org/ 10.1016/j.apenergy.2021.116983, URL https://linkinghub.elsevier.com/retrieve/ pii/S0306261921004529

    arXiv 2021

  3. [1]

    Lessons learned from electricity market liberalization

    Joskow PL. Lessons learned from electricity market liberalization. En- ergy J 2008;29(2_suppl):9–42. http://dx.doi.org/10.5547/ISSN0195-6574-EJ- Vol29-NoSI2-3, URL https://journals.sagepub.com/doi/10.5547/ISSN0195-6574- EJ-Vol29-NoSI2-3

    work page doi:10.5547/issn0195-6574-ej- 2008

  4. [2]

    Market design and price behavior in restructured electricity mar- kets: An international comparison

    Wolak FA. Market design and price behavior in restructured electricity mar- kets: An international comparison. In: Faruqui A, Eakin K, editors. Pricing in competitive electricity markets. Boston, MA: Springer US; 2000, p. 127–52. http://dx.doi.org/10.1007/978-1-4615-4529-3_8, URL http://link.springer.com/ 10.1007/978-1-4615-4529-3_8

    work page doi:10.1007/978-1-4615-4529-3_8

  5. [3]

    Electricity markets around the world

    Mayer K, Trück S. Electricity markets around the world. J Commod Mark 2018;9:77–100. http://dx.doi.org/10.1016/j.jcomm.2018.02.001, URL https:// www.sciencedirect.com/science/article/pii/S2405851318300059

    work page doi:10.1016/j.jcomm.2018.02.001 2018

  6. [4]

    The merit order and price-setting dynamics in European electricity markets

    Gasparella A, Koolen D, Zucker A. The merit order and price-setting dynamics in European electricity markets. 2023, URL https://publications.jrc.ec.europa.eu/ repository/handle/JRC134300

    work page 2023

  7. [5]

    Why we need to stick with uniform-price auc- tions in electricity markets

    Cramton P, Stoft S. Why we need to stick with uniform-price auc- tions in electricity markets. Electr J 2007;20(1):26–37. http://dx.doi.org/10. 1016/j.tej.2006.11.011, URL https://www.sciencedirect.com/science/article/pii/ S1040619006001527

    work page 2007

  8. [6]

    Electricity 2024: Analysis and forecast to 2026

    IEA. Electricity 2024: Analysis and forecast to 2026. 2024, URL https: //iea.blob.core.windows.net/assets/18f3ed24-4b26-4c83-a3d2-8a1be51c8cc8/ Electricity2024-Analysisandforecastto2026.pdf

    work page 2024

Show all 41 references
  1. [7]

    A solution to global warming, air pollution, and energy insecurity for 149 countries

    Jacobson MZ. A solution to global warming, air pollution, and energy insecurity for 149 countries. 2023, URL https://web.stanford.edu/group/efmh/jacobson/ Articles/I/149Country/24-WWS-149Countries.pdf

    2023

  2. [8]

    Forecasting electricity prices

    Maciejowska K, Uniejewski B, Weron R. Forecasting electricity prices. 2022, http://dx.doi.org/10.48550/arXiv.2204.11735, URL http://arxiv.org/abs/2204. 11735. arXiv:2204.11735 [q-fin]

    2022 doi arXiv

  3. [9]

    Forecasting prices in electricity markets: Needs, tools and limitations

    Gil HA, Gómez-Quiles C, Gómez-Expósito A, Santos JR. Forecasting prices in electricity markets: Needs, tools and limitations. In: Sorokin A, Rebennack S, Pardalos PM, Iliadis NA, Pereira MVF, editors. Handbook of networks in power systems i. Berlin, Heidelberg: Springer; 2012,...

    2012

  4. [10]

    ARIMA models to predict next- day electricity prices

    Contreras J, Espinola R, Nogales F, Conejo A. ARIMA models to predict next- day electricity prices. IEEE Trans Power Syst 2003;18(3):1014–20. http://dx. doi.org/10.1109/TPWRS.2002.804943, URL https://ieeexplore.ieee.org/abstract/ document/1216141. Conference Name: IEEE Transac...

    2003

  5. [11]

    Forecasting electricity prices for a day-ahead pool-based electric energy market

    Conejo AJ, Contreras J, Espínola R, Plazas MA. Forecasting electricity prices for a day-ahead pool-based electric energy market. Int J Fore- cast 2005;21(3):435–62. http://dx.doi.org/10.1016/j.ijforecast.2004.12.005, URL https://www.sciencedirect.com/science/article/pii/S01692...

    2005 doi

  6. [12]

    Electricity price forecasting in deregu- lated markets: A review and evaluation

    Aggarwal SK, Saini LM, Kumar A. Electricity price forecasting in deregu- lated markets: A review and evaluation. Int J Electr Power Energy Syst 2009;31(1):13–22. http://dx.doi.org/10.1016/j.ijepes.2008.09.003, URL https:// linkinghub.elsevier.com/retrieve/pii/S0142061508000884

    2009 doi

  7. [13]

    Electricity price forecasting: A review of the state-of-the-art with a look into the future

    Weron R. Electricity price forecasting: A review of the state-of-the-art with a look into the future. Int J Forecast 2014;30(4):1030–81. http://dx.doi.org/10. 1016/j.ijforecast.2014.08.008, URL https://linkinghub.elsevier.com/retrieve/pii/ S0169207014001083

    2014

  8. [14]

    The state of the art electricity load and price forecasting for the modern whole- sale electricity market

    Laitsos V, Vontzos G, Paraschoudis P, Tsampasis E, Bargiotas D, Tsoukalas LH. The state of the art electricity load and price forecasting for the modern whole- sale electricity market. Energies 2024;17(22):5797. http://dx.doi.org/10.3390/ en17225797, URL https://www.mdpi.com/1...

    2024

  9. [15]

    Integrated forecasting method for wind energy management: A case study in China

    Dong Y, Zhang L, Liu Z, Wang J. Integrated forecasting method for wind energy management: A case study in China. Processes 2020;8(1):35. http://dx.doi.org/ 10.3390/pr8010035, URL https://www.mdpi.com/2227-9717/8/1/35. Number: 1 Publisher: Multidisciplinary Digital Publishing Institute

    2020 doi

  10. [16]

    Decomposition-selection-ensemble predic- tion system for short-term wind speed forecasting

    Jiang P, Liu Z, Wang J, Zhang L. Decomposition-selection-ensemble predic- tion system for short-term wind speed forecasting. Electr Power Syst Res 2022;211:108186. http://dx.doi.org/10.1016/j.epsr.2022.108186, URL https:// www.sciencedirect.com/science/article/pii/S03787796220...

    2025

  11. [17]

    Hybrid model for profit-driven churn prediction based on cost minimization and return maximization

    Jiang P, Liu Z, Zhang L, Wang J. Hybrid model for profit-driven churn prediction based on cost minimization and return maximization. Expert Syst Appl 2023;228:120354. http://dx.doi.org/10.1016/j.eswa.2023.120354, URL https:// www.sciencedirect.com/science/article/pii/S0957417423008564

    2023

  12. [18]

    Predicting dissolved oxygen level using Young’s double-slit experiment optimizer-based weighting model

    Dong Y, Sun Y, Liu Z, Du Z, Wang J. Predicting dissolved oxygen level using Young’s double-slit experiment optimizer-based weighting model. J Environ Manag 2024;351:119807. http://dx.doi.org/10.1016/j.jenvman.2023. 119807, URL https://linkinghub.elsevier.com/retrieve/pii/S0301...

    2024 doi

  13. [19]

    Quantifying uncertainties of neural network-based electricity price forecasts

    Khosravi A, Nahavandi S, Creighton D. Quantifying uncertainties of neural network-based electricity price forecasts. Appl Energy 2013;112:120–9. http: //dx.doi.org/10.1016/j.apenergy.2013.05.075, URL https://www.sciencedirect. com/science/article/pii/S0306261913004881

    2013 doi

  14. [20]

    Forecasting spot electricity prices: Deep learning approaches and empirical comparison of traditional algorithms

    Lago J, De Ridder F, De Schutter B. Forecasting spot electricity prices: Deep learning approaches and empirical comparison of traditional algorithms. Appl En- ergy 2018;221:386–405. http://dx.doi.org/10.1016/j.apenergy.2018.02.069, URL https://linkinghub.elsevier.com/retrieve/...

    2018 doi

  15. [21]

    Bench- marking and survey of explanation methods for black box models

    Bodria F, Giannotti F, Guidotti R, Naretto F, Pedreschi D, Rinzivillo S. Bench- marking and survey of explanation methods for black box models. 2021, URL http://arxiv.org/abs/2102.13076. arXiv:2102.13076 [cs]

    2021 arXiv

  16. [22]

    Interpretable machine learning

    Molnar C. Interpretable machine learning. Molnar; 2022, URL https:// christophm.github.io/interpretable-ml-book/

    2022

  17. [23]

    Explainable artificial intelligence (XAI) techniques for energy and power sys- tems: Review, challenges and opportunities

    Machlev R, Heistrene L, Perl M, Levy K, Belikov J, Mannor S, Levron Y. Explainable artificial intelligence (XAI) techniques for energy and power sys- tems: Review, challenges and opportunities. Energy AI 2022;9:100169. http:// dx.doi.org/10.1016/j.egyai.2022.100169, URL https:...

    2022

  18. [24]

    A unified approach to interpreting model predictions

    Lundberg S, Lee S-I. A unified approach to interpreting model predictions. 2017, URL http://arxiv.org/abs/1705.07874. arXiv:1705.07874 [cs, stat]

    2017 arXiv

  19. [25]

    ‘‘Why should i trust you?’’: Explaining the predictions of any classifier

    Ribeiro MT, Singh S, Guestrin C. ‘‘Why should i trust you?’’: Explaining the predictions of any classifier. In: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. San Francisco California USA: ACM; 2016, p. 1135–44. http://dx.do...

    2016

  20. [26]

    Explainable profit-driven hotel booking cancellation prediction based on heterogeneous stacking-based ensemble clas- sification

    Liu Z, De Bock KW, Zhang L. Explainable profit-driven hotel booking cancellation prediction based on heterogeneous stacking-based ensemble clas- sification. European J Oper Res 2025;321(1):284–301. http://dx.doi.org/10. 1016/j.ejor.2024.08.026, URL https://www.sciencedirect.co...

    2025

  21. [28]

    An improved and explainable electricity price forecasting model via SHAP- based error compensation approach

    Heistrene L, Belikov J, Baimel D, Katzir L, Machlev R, Levy K, Mannor S, Lev- ron Y. An improved and explainable electricity price forecasting model via SHAP- based error compensation approach. IEEE Trans Artif Intell 2024;1–11. http://dx. doi.org/10.1109/TAI.2024.3455313, URL...

    2024

  22. [29]

    Random forests

    Breiman L. Random forests. Mach Learn 2001;45(1):5–32. http://dx.doi.org/10. 1023/A:1010933404324

    2001

  23. [30]

    All models are wrong, but many are useful: Learning a variable’s importance by studying an entire class of prediction models simultaneously

    Fisher A, Rudin C, Dominici F. All models are wrong, but many are useful: Learning a variable’s importance by studying an entire class of prediction models simultaneously. J Mach Learn Res 2019;20(177):1–81, URL http://jmlr.org/ papers/v20/18-760.html

    2019

  24. [31]

    Understanding electricity prices beyond the merit order principle using explainable AI

    Trebbien J, Rydin Gorjão L, Praktiknjo A, Schäfer B, Witthaut D. Understanding electricity prices beyond the merit order principle using explainable AI. Energy AI 2023;13:100250. http://dx.doi.org/10.1016/j.egyai.2023.100250, URL https: //linkinghub.elsevier.com/retrieve/pii/S...

    2023

  25. [32]

    Consistent individualized feature attribution for tree ensembles

    Lundberg SM, Erion GG, Lee S-I. Consistent individualized feature attribution for tree ensembles. 2019, http://dx.doi.org/10.48550/arXiv.1802.03888, URL http://arxiv.org/abs/1802.03888. arXiv:1802.03888 [cs]

    2019 doi arXiv

  26. [33]

    Bridging accuracy and explainabil- ity in electricity price forecasting

    Mascarenhas MM, Amelin M, Kazmi H. Bridging accuracy and explainabil- ity in electricity price forecasting. In: 2024 20th international conference on the European energy market. EEM, 2024, p. 1–6. http://dx.doi.org/ 10.1109/EEM60825.2024.10608857, URL https://ieeexplore.ieee.o...

    2024

  27. [34]

    Electricity price forecasting on the day-ahead market using machine learning

    Tschora L, Pierre E, Plantevit M, Robardet C. Electricity price forecasting on the day-ahead market using machine learning. Appl Energy 2022;313:118752. http: //dx.doi.org/10.1016/j.apenergy.2022.118752, URL https://linkinghub.elsevier. com/retrieve/pii/S0306261922002057

    2022

  28. [36]

    The role of natural gas in setting elec- tricity prices in Europe

    Zakeri B, Staffell I, Dodds PE, Grubb M, Ekins P, Jääskeläinen J, Cross S, Helin K, Castagneto Gissey G. The role of natural gas in setting elec- tricity prices in Europe. Energy Rep 2023;10:2778–92. http://dx.doi.org/10. 1016/j.egyr.2023.09.069, URL https://www.sciencedirect....

    2023

  29. [37]

    Algorithms for hyper-parameter optimization

    Bergstra JS, Bardenet R, Bengio Y, Kégl B. Algorithms for hyper-parameter optimization. Adv Neural Inf Process Syst 2011;24. URL https://core.ac.uk/ download/pdf/46766638.pdf

    2011

  30. [38]

    A value for n-person games (1953)

    Shapley LS. A value for n-person games (1953). In: Roth AE, edi- tor. The Shapley value: essays in honor of lloyd s. Shapley. Cambridge: Cambridge University Press; 1988, p. 31–40. http://dx.doi.org/10.1017/ CBO9780511528446.003, URL https://www.cambridge.org/core/books/shaple...

    1953

  31. [39]

    True to the model or true to the data? 2020, URL http://arxiv.org/abs/2006.16234

    Chen H, Janizek JD, Lundberg S, Lee S-I. True to the model or true to the data? 2020, URL http://arxiv.org/abs/2006.16234. arXiv:2006.16234 [cs, stat]

    2020 arXiv

  32. [40]

    Algorithms to estimate Shapley value feature attributions

    Chen H, Covert IC, Lundberg SM, Lee S-I. Algorithms to estimate Shapley value feature attributions. Nat Mach Intell 2023;5(6):590–601. http://dx.doi.org/10. 1038/s42256-023-00657-x, URL https://www.nature.com/articles/s42256-023- 00657-x

    2023

  33. [42]

    Electricity price forecasting: The dawn of machine learning

    Jedrzejewski A, Lago J, Marcjasz G, Weron R. Electricity price forecasting: The dawn of machine learning. IEEE Power Energy Mag 2022;20(3):24–31. http://dx.doi.org/10.1109/MPE.2022.3150809, URL https://ieeexplore.ieee.org/ document/9761111/. Conference Name: IEEE Power and Ene...

    2025

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.