The M4 Competition: Results, findings, conclusion and way forward
The DOI matches its Crossref record; 10 of the 10 references it cites were found in a registry or an index.
Abstract
The M4 competition is the continuation of three previous competitions started more than 45 years ago whose purpose was to learn how to improve forecasting accuracy, and how such learning can be applied to advance the theory and practice of forecasting. The purpose of M4 was to replicate the results of the previous ones and extend them into three directions: First significantly increase the number of series, second include Machine Learning (ML) forecasting methods, and third evaluate both point forecasts and prediction intervals. The five major findings of the M4 Competitions are: 1. Out Of the 17 most accurate methods, 12 were “combinations” of mostly statistical approaches. 2. The biggest surprise was a “hybrid” approach that utilized both statistical and ML features. This method’s average sMAPE was close to 10% more accurate than the combination benchmark used to compare the submitted methods. 3. The second most accurate method was a combination of seven statistical methods and one ML one, with the weights for the averaging being calculated by a ML algorithm that was trained to minimize the forecasting. 4. The two most accurate methods also achieved an amazing success in specifying the 95% prediction intervals correctly. 5. The six pure ML methods performed poorly, with none of them being more accurate than the combination benchmark and only one being more accurate than Naive2. This paper presents some initial results of M4, its major findings and a logical conclusion. Finally, it outlines what the authors consider to be the way forward for the field of forecasting.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
The M3-Competition: results, conclusions and implications
Makridakis, S., & Hibon, M. (2000) · International Journal of Forecasting · doi:10.1016/s0169-2070(00)00057-1 · Cited by 1,695
Self-citationStatistical and machine learning forecasting methods: Concerns and ways forward
Makridakis, S., Spiliotis, E., & Assimakopoulos, V. (2018) · PLOS ONE · doi:10.1371/journal.pone.0194889 · Cited by 1,528
Self-citationThe accuracy of extrapolation (time series) methods: Results of a forecasting competition
Makridakis, S., Andersen, A., Carbone, R., Fildes, R., Hibon, M., Lewandowski, R., Newton, J., Parzen, E., & Winkler, R. (1982) · Journal of Forecasting · doi:10.1002/for.3980010202 · Cited by 1,462
Self-citationforecast: Forecasting Functions for Time Series and Linear Models
Hyndman, R., Athanasopoulos, G., Bergmeir, C., Caceres, G., Chhay, L., Kuroptev, K., Mücke, M., O'Hara-Wild, M., Petropoulos, F., Razbash, S., et al. (2009) · CRAN: Contributed Packages · doi:10.32614/cran.package.forecast · Cited by 791
An Empirical Comparison of Machine Learning Models for Time Series Forecasting
Ahmed, N. K., Atiya, A. F., Gayar, N. E., & El-Shishiny, H. (2010) · Econometric Reviews · doi:10.1080/07474938.2010.481556 · Cited by 789
Accuracy of Forecasting: An Empirical Investigation
Makridakis, S., Hibon, M., & Moser, C. (1979) · Journal of the Royal Statistical Society. Series A (General) · doi:10.2307/2345077 · Cited by 514
Self-citationThe M2-competition: A real-time judgmentally based forecasting study
Makridakis, S., Chatfield, C., Hibon, M., Lawrence, M., Mills, T., Ord, K., & Simmons, L. F. (1993) · International Journal of Forecasting · doi:10.1016/0169-2070(93)90044-n · Cited by 263
Self-citation‘Horses for Courses’ in demand forecasting
Petropoulos, F., Makridakis, S., Assimakopoulos, V., & Nikolopoulos, K. (2014) · European Journal of Operational Research · doi:10.1016/j.ejor.2014.02.036 · Cited by 201
Self-citationVisualising forecasting algorithm performance using time series instance spaces
Kang, Y., Hyndman, R. J., & Smith‐Miles, K. (2017) · International Journal of Forecasting · doi:10.1016/j.ijforecast.2016.09.004 · Cited by 157
Objectivity, reproducibility and replicability in forecasting research
Makridakis, S., Assimakopoulos, V., & Spiliotis, E. (2018) · International Journal of Forecasting · doi:10.1016/j.ijforecast.2018.05.001 · Cited by 44
Self-citation
Cited by
DeepAR: Probabilistic forecasting with autoregressive recurrent networks
Salinas, D., Flunkert, V., Gasthaus, J., & Januschowski, T. (2020) · International Journal of Forecasting · doi:10.1016/j.ijforecast.2019.07.001 · Cited by 2,661
Recurrent Neural Networks for Time Series Forecasting: Current status and future directions
Hewamalage, H., Bergmeir, C., & Bandara, K. (2021) · International Journal of Forecasting · doi:10.1016/j.ijforecast.2020.06.008 · Cited by 1,371
Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting
Li, S., Jin, X., Xuan, Y., Zhou, X., Chen, W., Wang, Y.-X., & Yan, X. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.00235 · Cited by 1,053
The M4 Competition: 100,000 time series and 61 forecasting methods
Makridakis, S., Spiliotis, E., & Assimakopoulos, V. (2020) · International Journal of Forecasting · doi:10.1016/j.ijforecast.2019.04.014 · Cited by 969
Self-citationAn Introduction to Machine Learning
Badillo, S., Banfai, B., Birzele, F., Davydov, I. I., Hutchinson, L., Kam‐Thong, T., Siebourg‐Polster, J., Steiert, B., & Zhang, J. D. (2020) · Clinical Pharmacology & Therapeutics · doi:10.1002/cpt.1796 · Cited by 799
A hybrid method of exponential smoothing and recurrent neural networks for time series forecasting
Smyl, S. (2020) · International Journal of Forecasting · doi:10.1016/j.ijforecast.2019.03.017 · Cited by 761
Applications of deep learning in stock market prediction: Recent progress
Jiang, W. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.115537 · Cited by 714
An Experimental Review on Deep Learning Architectures for Time Series Forecasting
Lara-Benítez, P., Carranza-García, M., & Riquelme, J. C. (2021) · International Journal of Neural Systems · doi:10.1142/s0129065721300011 · Cited by 532
A review of artificial neural network models for ambient air pollution prediction
Cabaneros, S. M., Calautit, J. K., & Hughes, B. R. (2019) · Environmental Modelling & Software · doi:10.1016/j.envsoft.2019.06.014 · Cited by 459
Probabilistic forecasting with temporal convolutional neural network
Chen, Y., Kang, Y., Chen, Y., & Wang, Z. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.03.011 · Cited by 393
Deep Learning for Time Series Forecasting: Tutorial and Literature Survey
Benidis, K., Rangapuram, S. S., Flunkert, V., Wang, Y., Maddix, D., Turkmen, C., Gasthaus, J., Bohlke-Schneider, M., Salinas, D., Stella, L., et al. (2023) · ACM Computing Surveys · doi:10.1145/3533382 · Cited by 382
Retail forecasting: Research and practice
Fildes, R., Ma, S., & Kolassa, S. (2022) · International Journal of Forecasting · doi:10.1016/j.ijforecast.2019.06.004 · Cited by 346
TSGBench: Time Series Generation Benchmark
Ang, Y., Huang, Q., Bao, Y., Tung, A. K. H., & Huang, Z. (2023) · Proceedings of the VLDB Endowment · doi:10.14778/3632093.3632097 · Cited by 38
Younis, H. A., Ruhaiyem, N. I. R., Badr, A. A., Abdul-Hassan, A. K., Alfadli, I. M., Binjumah, W. M., Altuwaijri, E. A., & Nasser, M. (2023) · Processes · doi:10.3390/pr11051488 · Cited by 21
Quantifying machine influence over human forecasters
Abeliuk, A., Benjamin, D. M., Morstatter, F., & Galstyan, A. (2020) · Scientific Reports · doi:10.1038/s41598-020-72690-4 · Cited by 18
“They found that general models, like the ARIMA one used in this work, perform the best across a wide array of problems, even in comparison to more complex machine learning models 68,69 .”
Method
Metadata
- Authors
- Makridakis, S., Spiliotis, E., & Assimakopoulos, V.
- Venue
- International Journal of Forecasting
- Publisher
- Elsevier BV
- Published
- 2018-10
- Volume
- 34
- Issue
- 4
- Pages
- 802-808
- ISSN
- 0169-2070 (ISSN-L)
- OpenAlex
- W2808800115
- Cited by
- 629
- References
- 10
- FWCI
- 49.57
- Type
- journal article
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.