pith. sign in

arxiv: 2309.16238 · v1 · submitted 2023-09-28 · 📊 stat.AP

Human spatial dynamics for electricity demand forecasting: the case of France during the 2022 energy crisis

Nathan Doum\`eche (LPSM , EDF R&D OSIRIS) , Yann Allioux (EDF R&D OSIRIS) , Yannig Goude (EDF R&D OSIRIS) , Stefania Rubrichi This is my paper

Pith reviewed 2026-05-24 06:48 UTC · model grok-4.3

classification 📊 stat.AP
keywords electricity demand forecastingmobility indicesmobile network dataenergy crisissobriety periodFrancework dynamicsstatistical models
0
0 comments X

The pith

Mobility indices from mobile network data improve electricity demand forecasting accuracy during France's 2022 energy crisis.

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

The paper examines how standard electricity demand models, which rely on meteorological and calendar data, had difficulty adapting to sudden shifts in consumption during the 2022 energy crisis and associated sobriety incentives in France. It shows that mobility indices derived from mobile network data, reflecting changes in work behaviors and spatial dynamics, lead to better forecast performance than state-of-the-art approaches during this atypical winter period. The work first documents the observed drop in French electricity consumption and then demonstrates how these indices capture relevant work dynamics. A sympathetic reader would care because accurate forecasts matter for energy security when relying on intermittent renewables and when consumption patterns can change abruptly due to external shocks.

Core claim

Mobility indices based on mobile network data significantly improve the performance of the state-of-the-art models in electricity demand forecasting during the sobriety period, by capturing work dynamics that standard models miss after the documented drop in consumption during winter 2022-2023 in France.

What carries the argument

Mobility indices from mobile network data that capture spatial work dynamics affecting electricity consumption.

If this is right

  • Electricity demand models can better adapt to abrupt behavioral changes triggered by energy crises or government incentives.
  • The specific contribution of work-related movements to electricity demand can be isolated and quantified.
  • Forecasts become more reliable for grid planning when consumption deviates from historical meteorological and calendar patterns.
  • The impact of sobriety measures on national consumption can be assessed with greater precision.

Where Pith is reading between the lines

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

  • The same mobility-based adjustment could be tested for demand forecasting in other European countries that experienced similar 2022 consumption drops.
  • Real-time mobility streams might enable shorter-horizon forecasts that respond to daily work-pattern fluctuations.
  • This approach could be extended to other utility sectors where spatial population shifts drive resource use, such as water or transport demand.

Load-bearing premise

The mobility indices from mobile network data accurately capture the relevant work dynamics affecting electricity consumption without significant confounding factors or data biases.

What would settle it

Adding the mobility indices to the forecasting models produces no improvement or a degradation in accuracy on out-of-sample data from the 2022-2023 period or a comparable behavioral-shift episode.

read the original abstract

Accurate electricity demand forecasting is crucial to meet energy security and efficiency, especially when relying on intermittent renewable energy sources. Recently, massive savings have been observed in Europe, following an unprecedented global energy crisis. However, assessing the impact of such crisis and of government incentives on electricity consumption behaviour is challenging. Moreover, standard statistical models based on meteorological and calendar data have difficulty adapting to such brutal changes. Here, we show that mobility indices based on mobile network data significantly improve the performance of the state-of-the-art models in electricity demand forecasting during the sobriety period. We start by documenting the drop in the French electricity consumption during the winter of 2022-2023. We then show how our mobile network data captures work dynamics and how adding these mobility indices outperforms the state-of-the-art during this atypical period. Our results characterise the effect of work behaviours on the electricity demand.

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 paper documents a drop in French electricity consumption during the 2022-2023 winter crisis and claims that mobility indices derived from mobile network data capture work dynamics, thereby significantly improving the performance of state-of-the-art statistical models for electricity demand forecasting during the sobriety period. It positions these indices as a way to adapt models to abrupt behavioral changes beyond standard meteorological and calendar covariates.

Significance. If the central empirical claim holds with rigorous validation, the work would demonstrate a practical channel for incorporating high-frequency behavioral proxies into energy forecasting, addressing a known limitation of standard models during crises. This could inform policy on demand response and grid stability, particularly if the mobility data provide falsifiable, out-of-sample gains net of existing covariates.

major comments (3)
  1. [Abstract] Abstract and introduction: The headline claim that mobility indices 'significantly improve' SOTA model performance is stated without any quantitative metrics (e.g., RMSE, MAE, or percentage improvement), error bars, cross-validation details, or explicit model-comparison tables. This absence makes the central empirical result impossible to evaluate and is load-bearing for the paper's contribution.
  2. [Data and methods] Section describing mobility index construction (likely §2 or §3): No information is provided on index construction details, whether the indices are residualized against the calendar/weather covariates already present in the baseline model, or validation against independent labor-market or building-occupancy data. Without these steps, the reported forecasting gain risks being an artifact of correlation with the crisis period itself rather than an isolated work-dynamics channel.
  3. [Results] Results section (likely §4): The paper asserts outperformance during the atypical period but does not report whether the mobility indices add explanatory power after controlling for price signals, media campaigns, or general mobility reductions, nor does it include placebo tests or falsification exercises that would rule out omitted-variable bias.
minor comments (2)
  1. [Abstract] The abstract refers to 'the sobriety period' without a precise date range or definition; this should be clarified with explicit calendar bounds.
  2. [Methods] Notation for the mobility indices and the baseline model should be introduced consistently with equation numbers if any are present.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed report. The comments highlight important areas for clarification and strengthening of the empirical claims. We address each major comment below and will incorporate revisions to improve the manuscript's transparency and robustness.

read point-by-point responses

  1. Referee: [Abstract] Abstract and introduction: The headline claim that mobility indices 'significantly improve' SOTA model performance is stated without any quantitative metrics (e.g., RMSE, MAE, or percentage improvement), error bars, cross-validation details, or explicit model-comparison tables. This absence makes the central empirical result impossible to evaluate and is load-bearing for the paper's contribution.

    Authors: We agree that the abstract and introduction would benefit from explicit quantitative support for the headline claim. While the results section contains model comparisons with RMSE, MAE, and cross-validation details, these are not summarized upfront. We will revise the abstract and introduction to report key metrics (e.g., percentage RMSE reduction during the sobriety period relative to the baseline), reference the relevant tables, and note the cross-validation scheme used. revision: yes

  2. Referee: [Data and methods] Section describing mobility index construction (likely §2 or §3): No information is provided on index construction details, whether the indices are residualized against the calendar/weather covariates already present in the baseline model, or validation against independent labor-market or building-occupancy data. Without these steps, the reported forecasting gain risks being an artifact of correlation with the crisis period itself rather than an isolated work-dynamics channel.

    Authors: The manuscript describes the mobile-network data sources and aggregation into daily mobility indices intended to proxy work-related presence. We will expand this section with precise construction steps (including any filtering or normalization) and explicitly state that the indices are not residualized against calendar/weather covariates, as they are meant to capture residual behavioral variation. Where feasible, we will add correlations with available labor-market aggregates; however, building-occupancy validation data are not available to us and will be noted as a limitation. revision: partial

  3. Referee: [Results] Results section (likely §4): The paper asserts outperformance during the atypical period but does not report whether the mobility indices add explanatory power after controlling for price signals, media campaigns, or general mobility reductions, nor does it include placebo tests or falsification exercises that would rule out omitted-variable bias.

    Authors: The results focus on out-of-sample gains during the 2022-2023 period. To address omitted-variable concerns, we will add regressions and forecasting exercises that include price signals and media-campaign indicators (where daily data exist) and report incremental explanatory power. We will also include placebo tests on pre-crisis periods to assess whether the mobility indices spuriously improve fit outside the sobriety window. revision: yes

Circularity Check

0 steps flagged

No significant circularity; empirical gains from external mobility indices

full rationale

The paper reports an empirical forecasting exercise in which mobility indices derived from mobile-network data are added as covariates to existing state-of-the-art electricity-demand models and shown to improve accuracy during the 2022-2023 French sobriety period. No equations, self-definitional constructions, fitted parameters renamed as predictions, or load-bearing self-citations appear in the abstract or description; the claimed improvement rests on external data sources and out-of-sample performance comparisons rather than any reduction of the result to its own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The paper relies on standard statistical modeling assumptions and the validity of mobile data as a proxy for work behavior.

pith-pipeline@v0.9.0 · 5715 in / 888 out tokens · 14435 ms · 2026-05-24T06:48:43.714659+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

95 extracted references · 95 canonical work pages

  1. [1]

    & Walker, G

    Shove, E. & Walker, G. What is energy for? Social practice and energy demand. Theory, culture & society 31, 41–58 (2014)

    work page 2014

  2. [2]

    & Gazzani, A

    Ferriani, F. & Gazzani, A. The impact of the war in Ukraine on energy prices: consequences for firms’ financial performance. International Economics 174, 221– 230 (2023)

    work page 2023

  3. [3]

    & Hirth, L

    Ruhnau, O., Stiewe, C., Muessel, J. & Hirth, L. Natural gas savings in Germany during the 2022 energy crisis. Nature energy (2023)

    work page 2022

  4. [4]

    European Commission. Report from the Commission to the European Par- liament and the Council on the review of emergency interventions to address high energy prices in accordance with Council Regulation (EU) 2022/1854 (05/06/2023). URL https://energy.ec.europa.eu/system/files/2023-06/COM 2023 302 1 EN ACT part1 v2.pdf

    work page 2022

  5. [5]

    R´ eactualisation des perspectives pour le syst` eme ´ electrique pour l’hiver 2022-2023 (f´ evrier 2023)

    RTE. R´ eactualisation des perspectives pour le syst` eme ´ electrique pour l’hiver 2022-2023 (f´ evrier 2023). Tech. Rep., R´ eseau de Transport d’´Electricit´ e (2023). URL https://assets.rte-france.com/prod/public/2023-02/ 2023-02-24-reactualisation-hiver.pdf

    work page 2022

  6. [6]

    Plan de sobri´ et´ e ´ energ´ etique

    French government. Plan de sobri´ et´ e ´ energ´ etique. Press release (10/06/2022). URL https://www.ecologie.gouv.fr/sites/default/files/dp-plan-sobriete.pdf

    work page 2022

  7. [7]

    Consommation ´ electrique : le plan de sobri´ et´ e fournit ses premiers effets

    Techniques de l’ing´ enieur. Consommation ´ electrique : le plan de sobri´ et´ e fournit ses premiers effets. Avail- able: https://www.techniques-ingenieur.fr/actualite/articles/ consommation-electrique-le-plan-de-sobriete-fournit-ses-premiers-effets-117348/ (2022). [Accessed: May 11, 2023]

    work page 2022

  8. [8]

    As Russia chokes Europe’s gas, France enters era of energy sobriety

    The New York Times. As Russia chokes Europe’s gas, France enters era of energy sobriety. Available: https://www.nytimes.com/2022/09/05/business/ russia-gas-europe-france.html (2022). [Accessed: May 11, 2023]

    work page 2022

  9. [9]

    ´Electricit´ e : la baisse inattendue de la consom- mation a permis d’´ eviter les coupures cet hiver

    Le Monde. ´Electricit´ e : la baisse inattendue de la consom- mation a permis d’´ eviter les coupures cet hiver. Available: 13 https://www.lemonde.fr/planete/article/2023/03/16/electricite-la- baisse-inattendue-de-la-consommation-a-permis-d-eviter-les-coupures-cet- hiver 6165773 3244.html (2023). [Accessed: May 11, 2023]

    work page 2023

  10. [10]

    & Mekhilef, S

    Abdelaziz, E., Saidur, R. & Mekhilef, S. A review on energy saving strate- gies in industrial sector. Renewable and Sustainable Energy Reviews 15, 150–168 (2011)

    work page 2011

  11. [11]

    A roadmap for rapid decarbonization.Science 355, 1269–1271 (2017)

    Rockstr¨ om, J.et al. A roadmap for rapid decarbonization.Science 355, 1269–1271 (2017)

    work page 2017

  12. [12]

    Hoegh-Guldberg, O. et al. The human imperative of stabilizing global climate change at 1.5°C. Science 365, eaaw6974 (2019)

    work page 2019

  13. [13]

    & Lucille, P.-L

    de Maere d’Aertrycke, G., Smeers, Y., de Peufeilhoux, H. & Lucille, P.-L. The role of electrification in the decarbonization of central-western Europe. Energies 13 (2020)

    work page 2020

  14. [14]

    Omar, H., Graetz, G. & Ho, M. Decarbonizing with Nuclear Power, Current Builds, and Future Trends, Ch. 6, 103–151 (John Wiley & Sons, Ltd, 2022)

    work page 2022

  15. [15]

    Energy pathways to 2050

    RTE. Energy pathways to 2050. Tech. Rep., R´ eseau de Transport d’´Electricit´ e (2022). URL https://assets.rte-france.com/prod/public/2022-01/ Energy%20pathways%202050 Key%20results.pdf

    work page 2050

  16. [16]

    Rolnick, D. et al. Tackling climate change with machine learning. ACM Computing Surveys (CSUR) 55, 1–96 (2022)

    work page 2022

  17. [17]

    & Francisco, A

    Pinheiro, M., Madeira, S. & Francisco, A. Short-term electricity load fore- casting—a systematic approach from system level to secondary substations. Applied Energy 332, 120493 (2023)

    work page 2023

  18. [18]

    Hong, T. et al. Energy forecasting: A review and outlook. IEEE Open Access Journal of Power and Energy 7, 376–388 (2020)

    work page 2020

  19. [19]

    & Adekoya, A

    Nti, I., Teimeh, M., Nyarko-Boateng, O. & Adekoya, A. Electricity load forecasting: a systematic review. Journal of Electrical Systems and Information Technology 7, 2314–7172 (2020)

    work page 2020

  20. [20]

    & Dragan, D

    Hammad, M., Jereb, B., Rosi, B. & Dragan, D. Methods and models for electric load forecasting: a comprehensive review. Logistics, Supply Chain, Sustainability and Global Challenges 11, 51–76 (2020)

    work page 2020

  21. [21]

    & Goude, Y

    Obst, D., de Vilmarest, J. & Goude, Y. Adaptive methods for short- term electricity load forecasting during covid-19 lockdown in France. IEEE Transactions on Power Systems 36, 4754–4763 (2021). 14

    work page 2021

  22. [22]

    & Krings, G

    Blondel, V., Decuyper, A. & Krings, G. Understanding vehicular routing behavior with location-based service data. EPJ Data Science 4 (2015)

    work page 2015

  23. [23]

    Deville, P. et al. Dynamic population mapping using mobile phone data. Proceedings of the National Academy of Sciences of the United States of America 111, 15888–15893 (2014)

    work page 2014

  24. [24]

    Blumenstock, J., Cadamuro, G. & On, R. Predicting poverty and wealth from mobile phone metadata. Science 350, 1073–1076 (2015)

    work page 2015

  25. [25]

    Lorenzo, G. et al. AllAboard: Visual exploration of cellphone mobility data to optimise public transport. IEEE Transactions on Visualization and Computer Graphics 22 (2016)

    work page 2016

  26. [26]

    & Schreeb, J

    Bengtsson, L., Lu, X., Thorson, A., Garfield, R. & Schreeb, J. Improved response to disasters and outbreaks by tracking population movements with mobile phone network data: a post-earthquake geospatial study in Haiti. PLoS Medicine 8, e100–1083 (2011)

    work page 2011

  27. [27]

    Inferring patterns of internal migration from mobile phone call records: evidence from Rwanda

    Blumenstock, J. Inferring patterns of internal migration from mobile phone call records: evidence from Rwanda. Information Technology for Development 18, 107–125 (2012)

    work page 2012

  28. [28]

    & Musolesi, M

    Rubrichi, S., Smoreda, Z. & Musolesi, M. A comparison of spatial-based tar- geted disease mitigation strategies using mobile phone data. EPJ Data Science 7 (2018)

    work page 2018

  29. [29]

    & Colizza, V

    Pullano, G., Valdano, E., Scarpa, N., Rubrichi, S. & Colizza, V. Evaluating the effect of demographic factors, socioeconomic factors, and risk aversion on mobility during the covid-19 epidemic in France under lockdown: a population- based study. Lancet Digit Health 2, e638–e649 (2020)

    work page 2020

  30. [30]

    & Gonz´ alez, M

    Xu, Y., Clemente, R. & Gonz´ alez, M. Understanding vehicular routing behavior with location-based service data. EPJ Data Science 10, 1–17 (2021)

    work page 2021

  31. [31]

    Filazzola, A. et al. Using smartphone-gps data to quantify human activity in green spaces. PLOS Computational Biology 18, 1–20 (2022)

    work page 2022

  32. [32]

    & Cecilia, J

    Terroso-S´ aenz, F., Mu˜ noz, A., Fern´ andez-Pedauye, J. & Cecilia, J. Human mobil- ity prediction with region-based flows and water consumption. IEEE Access 9, 88651–88663 (2021)

    work page 2021

  33. [33]

    Smolak, K. et al. Applying human mobility and water consumption data for short-term water demand forecasting using classical and machine learning models. Urban Water Journal 17, 32–42 (2020). 15

    work page 2020

  34. [34]

    & Zhang, B

    Chen, Y., Yang, W. & Zhang, B. Using mobility for electrical load forecasting during the covid-19 pandemic. arXiv:2006.08826 (2020)

    work page arXiv 2006

  35. [35]

    Ruan, G. et al. A cross-domain approach to analyzing the short-run impact of covid-19 on the U.S. electricity sector. Joule 4, 2322–2337 (2020)

    work page 2020

  36. [36]

    & Mcardle, G

    Zarbakhsh, N., Misaghian, M. & Mcardle, G. Human mobility-based fea- tures to analyse the impact of covid-19 on power system operation of ireland. IEEE Open Access Journal of Power and Energy 9, 213–225 (2022)

    work page 2022

  37. [37]

    & Goude, Y

    Antoniadis, A., Gaucher, S. & Goude, Y. Hierarchical transfer learning with applications to electricity load forecasting. International Journal of Forecasting (2023)

    work page 2023

  38. [38]

    Flux vision

    Business, O. Flux vision. URL https://www.orange-business.com/en/solutions/ data-intelligence-iot/flux-vision

    work page

  39. [39]

    L´ evy, J., Piantoni, J. C. S. & Fran¸ cois, . Who lives where? Counting, locating, and observing France’s real inhabitants. SocArXiv (2023)

    work page 2023

  40. [40]

    & Eckley, I

    Killick, R. & Eckley, I. Changepoint: an R package for changepoint analysis. Journal of Statistical Software 58, 1–19 (2014)

    work page 2014

  41. [41]

    & Cook, D

    Aminikhanghahi, S. & Cook, D. A survey of methods for time series change point detection. Knowledge and Information Systems 51, 339–367 (2017)

    work page 2017

  42. [42]

    Winter 2022-2023

    RTE. Winter 2022-2023. Tech. Rep., R´ eseau de Transport d’´Electricit´ e (2023). URL https://www.rte-france.com/actualites/ bilan-hiver-2022-2023-coupures-electricite-evitees-grace-baisse-consommation

    work page 2022

  43. [43]

    The global normalcy index

    The Economist. The global normalcy index. Available: https://www. economist.com/graphic-detail/tracking-the-return-to-normalcy-after-covid-19? utm medium=pr&utm source=inf-a (2023). [Accessed: July 11, 2023]

    work page 2023

  44. [44]

    Adaptive load forecasting methodology based on generalized additive model with automatic variable selection

    Krstonijevi´ c, S. Adaptive load forecasting methodology based on generalized additive model with automatic variable selection. Sensors (Basel) (2022)

    work page 2022

  45. [45]

    & Chaturvedi, D

    Singh, A., Ibraheem, Khatoon, S., Muazzam, M. & Chaturvedi, D. Load forecasting techniques and methodologies: a review. 2012 2nd International Conference on Power, Control and Embedded Systems 1–10 (2012)

    work page 2012

  46. [46]

    Wang, Z. et al. Benchmarks and custom package for electrical load forecasting. arXiv:2307.07191 (2023)

    work page arXiv 2023

  47. [47]

    & Goude, Y

    Vilmarest, J. & Goude, Y. State-space models for online post-covid electricity load forecasting competition. IEEE Open Access Journal of Power and Energy 9, 192–201 (2022). 16

    work page 2022

  48. [48]

    & Poggi, J.-M

    Goehry, B., Yan, H., Goude, Y., Massart, P. & Poggi, J.-M. Random forests for time series. REVSTAT-Statistical Journal 21, 283–302 (2023)

    work page 2023

  49. [49]

    & Roeder, K

    Wasserman, L. & Roeder, K. High-dimensional variable selection. The Annals of Statistics 37, 2178–2201 (2009)

    work page 2009

  50. [50]

    & Wei, F

    Huang, J., Horowitz, J. & Wei, F. Variable selection in nonparametric additive models. Annals of statistics 38, 2282–2313 (2010)

    work page 2010

  51. [51]

    & Wood, S

    Marra, G. & Wood, S. Practical variable selection for generalized additive models. Computational Statistics & Data Analysis 55, 2372–2387 (2011)

    work page 2011

  52. [52]

    & Tuleau-Malot, C

    Genuer, R., Poggi, J.-M. & Tuleau-Malot, C. Variable selection using random forests. Pattern Recognition Letters 31, 2225–2236 (2010)

    work page 2010

  53. [53]

    & Ding, C

    Peng, H., Long, F. & Ding, C. Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy. IEEE Transactions on Pattern Analysis and Machine Intelligence 27, 1226–1238 (2005)

    work page 2005

  54. [54]

    & Liu, X

    Han, M., Ren, W. & Liu, X. Joint mutual information-based input variable selection for multivariate time series modeling. Engineering Applications of Artificial Intelligence 37, 250–257 (2015)

    work page 2015

  55. [55]

    Jay, N. D. et al. mRMRe: an R package for parallelized mRMR ensemble feature selection. Bioinformatics 29, 2365–2368 (2013)

    work page 2013

  56. [56]

    & Dincer, I

    Abu-Rayash, A. & Dincer, I. Analysis of the electricity demand trends amidst the covid-19 coronavirus pandemic. Energy Research & Social Science 68, 101682 (2020)

    work page 2020

  57. [57]

    & Xing, B

    Ku, A., Qiu, Y., Lou, J., Nock, D. & Xing, B. Changes in hourly electricity consumption under covid mandates: a glance to future hourly residential power consumption pattern with remote work in Arizona. Applied Energy 310, 118539 (2022)

    work page 2022

  58. [58]

    Enquˆ ete mobilit´ e des personnes 2018-2019

    MTECT. Enquˆ ete mobilit´ e des personnes 2018-2019. Avail- able: https://www.statistiques.developpement-durable.gouv.fr/ resultats-detailles-de-lenquete-mobilite-des-personnes-de-2019. [Accessed: July 11, 2023]

    work page 2018

  59. [59]

    New normal at work in a post-covid world: work–life balance and labor markets

    Vyas, L. New normal at work in a post-covid world: work–life balance and labor markets. Policy and Society 41, 155–167 (2022)

    work page 2022

  60. [60]

    Telework and productivity three years after the start of the pandemic

    Pora, P. Telework and productivity three years after the start of the pandemic. Economie et Statistique / Economics and Statistics 593 (2023). 17

    work page 2023

  61. [61]

    & Fan, S

    Hong, T. & Fan, S. Probabilistic electric load forecasting: a tutorial review. International Journal of Forecasting 32, 914–938 (2016)

    work page 2016

  62. [62]

    Jours f´ eri´ es en France

    Etalab. Jours f´ eri´ es en France. Available: https://www.data.gouv.fr/fr/datasets/ jours-feries-en-france/ (2023). [Accessed: May 11, 2023]

    work page 2023

  63. [63]

    Vacances scolaires par zones

    Etalab. Vacances scolaires par zones. Available: https://www.data.gouv.fr/fr/ datasets/vacances-scolaires-par-zones/ (2023). [Accessed: May 11, 2023]

    work page 2023

  64. [64]

    Donn´ ees SYNOP essentielles OMM

    M´ et´ eo-France. Donn´ ees SYNOP essentielles OMM. Available: https: //public.opendatasoft.com/explore/dataset/donnees-synop-essentielles-omm/ (2023). [Accessed: May 11, 2023]

    work page 2023

  65. [65]

    ´ eCO2mix

    RTE. ´ eCO2mix. Available: https://public.opendatasoft.com/explore/dataset/ donnees-synop-essentielles-omm/ (2023). [Accessed: May 11, 2023]

    work page 2023

  66. [66]

    Resampling methods for dependent data 1 edn

    Lahiri, S. Resampling methods for dependent data 1 edn. Springer Series in Statistics (Springer New York, NY, 2003)

    work page 2003

  67. [67]

    & Wintenberger, O

    Vilmarest, J., Browell, J., Fasiolo, M., Goude, Y. & Wintenberger, O. Adaptive probabilistic forecasting of electricity (net-)load. arXiv:2301.10090 (2023)

    work page arXiv 2023

  68. [68]

    Taieb, S. B. & Hyndman, R. A gradient boosting approach to the kaggle load forecasting competition. International Journal of Forecasting 30, 382–394 (2014)

    work page 2014

  69. [69]

    Michailesco, O

    F. Michailesco, O. S. Migrations r´ esidentielles post-covid : l’attractivit´ e du p´ eriurbain l´ eg` erement renforc´ ee. Tech. Rep., Insee (2023). [Accessed: Sept 21, 2023]

    work page 2023

  70. [70]

    & Rossell´ o, J

    Bakhat, M. & Rossell´ o, J. Estimation of tourism-induced electricity consumption: The case study of balearics islands, spain. Energy Economics 33, 437–444 (2011)

    work page 2011

  71. [71]

    & Lam, K

    Lai, T., To, W., Lo, W., Choy, Y. & Lam, K. The causal relationship between electricity consumption and economic growth in a gaming and tourism center: the case of Macao SAR, the People’s Republic of China. Energy 36, 1134–1142 (2011)

    work page 2011

  72. [72]

    Arriv´ ees dans l’hˆ otellerie - Total - France m´ etropolitaine

    INSEE. Arriv´ ees dans l’hˆ otellerie - Total - France m´ etropolitaine. Avail- able: https://www.insee.fr/fr/statistiques/serie/010598571 (2023). ID 010598571, [Accessed: May 11, 2023]

    work page arXiv 2023

  73. [73]

    Emmanuel, T. et al. A survey on missing data in machine learning. Journal of Big Data 8 (2021)

    work page 2021

  74. [74]

    & Marlin, B

    Shukla, S. & Marlin, B. A survey on principles, models and methods for learning from irregularly sampled time series (2021). 18

    work page 2021

  75. [75]

    & Rubin, D

    Little, R. & Rubin, D. Statistical analysis with missing data Vol. 793 (John Wiley & Sons, 2019)

    work page 2019

  76. [76]

    Ma, J. et al. Transfer learning for long-interval consecutive missing values imputation without external features in air pollution time series. Advanced Engineering Informatics 44, 101092 (2020)

    work page 2020

  77. [77]

    Zhang, Y. et al. Missing value imputation in multivariate time series with end- to-end generative adversarial networks. Information Sciences 551, 67–82 (2021)

    work page 2021

  78. [78]

    & Scornet, E

    Ayme, A., Boyer, C., Dieuleveut, A. & Scornet, E. Naive imputation implicitly regularizes high-dimensional linear models. arXiv:2301.13585 (2023)

    work page arXiv 2023

  79. [79]

    & Jenkins, G

    Box, G. & Jenkins, G. M. Time Series Analysis: Forecasting and Control (Holden-Day, 1976)

    work page 1976

  80. [80]

    Wood, S. N. Generalized additive models: an introduction with R (CRC press, 2017)

    work page 2017

Showing first 80 references.