Correlation analysis of forecasting methods

The case of the M4 competition

Pantelis Agathangelou, Demetris Trihinas, Ioannis Katakis

    Research output: Contribution to journalComment/debate

    Abstract

    This commentary introduces a correlation analysis of the top-10 ranked forecasting methods that participated in the M4 forecasting competition. The “M” competitions attempt to promote and advance research in the field of forecasting by inviting both industry and academia to submit forecasting algorithms for evaluation over a large corpus of real-world datasets. After performing the initial analysis to derive the errors of each method, we proceed to investigate the pairwise correlations among them in order to understand the extent to which they produce errors in similar ways. Based on our results, we conclude that there is indeed a certain degree of correlation among the top-10 ranked methods, largely due to the fact that many of them consist of a combination of well-known, statistical and machine learning techniques. This fact has a strong impact on the results of the correlation analysis, and therefore leads to similar forecasting error patterns.

    Original languageEnglish
    JournalInternational Journal of Forecasting
    DOIs
    Publication statusPublished - 1 Jan 2019

    Fingerprint

    Forecasting method
    Correlation analysis
    Forecasting competitions
    Forecasting error
    M-Competition
    Statistical learning
    Machine learning
    Evaluation
    Industry

    Keywords

    • Forecasting
    • M4
    • Machine learning
    • Makridakis
    • Statistics

    Cite this

    @article{bceba5f8ef524b9486221921d648e560,
    title = "Correlation analysis of forecasting methods: The case of the M4 competition",
    abstract = "This commentary introduces a correlation analysis of the top-10 ranked forecasting methods that participated in the M4 forecasting competition. The “M” competitions attempt to promote and advance research in the field of forecasting by inviting both industry and academia to submit forecasting algorithms for evaluation over a large corpus of real-world datasets. After performing the initial analysis to derive the errors of each method, we proceed to investigate the pairwise correlations among them in order to understand the extent to which they produce errors in similar ways. Based on our results, we conclude that there is indeed a certain degree of correlation among the top-10 ranked methods, largely due to the fact that many of them consist of a combination of well-known, statistical and machine learning techniques. This fact has a strong impact on the results of the correlation analysis, and therefore leads to similar forecasting error patterns.",
    keywords = "Forecasting, M4, Machine learning, Makridakis, Statistics",
    author = "Pantelis Agathangelou and Demetris Trihinas and Ioannis Katakis",
    year = "2019",
    month = "1",
    day = "1",
    doi = "10.1016/j.ijforecast.2019.03.020",
    language = "English",
    journal = "International Journal of Forecasting",
    issn = "0169-2070",
    publisher = "Elsevier",

    }

    Correlation analysis of forecasting methods : The case of the M4 competition. / Agathangelou, Pantelis; Trihinas, Demetris; Katakis, Ioannis.

    In: International Journal of Forecasting, 01.01.2019.

    Research output: Contribution to journalComment/debate

    TY - JOUR

    T1 - Correlation analysis of forecasting methods

    T2 - The case of the M4 competition

    AU - Agathangelou, Pantelis

    AU - Trihinas, Demetris

    AU - Katakis, Ioannis

    PY - 2019/1/1

    Y1 - 2019/1/1

    N2 - This commentary introduces a correlation analysis of the top-10 ranked forecasting methods that participated in the M4 forecasting competition. The “M” competitions attempt to promote and advance research in the field of forecasting by inviting both industry and academia to submit forecasting algorithms for evaluation over a large corpus of real-world datasets. After performing the initial analysis to derive the errors of each method, we proceed to investigate the pairwise correlations among them in order to understand the extent to which they produce errors in similar ways. Based on our results, we conclude that there is indeed a certain degree of correlation among the top-10 ranked methods, largely due to the fact that many of them consist of a combination of well-known, statistical and machine learning techniques. This fact has a strong impact on the results of the correlation analysis, and therefore leads to similar forecasting error patterns.

    AB - This commentary introduces a correlation analysis of the top-10 ranked forecasting methods that participated in the M4 forecasting competition. The “M” competitions attempt to promote and advance research in the field of forecasting by inviting both industry and academia to submit forecasting algorithms for evaluation over a large corpus of real-world datasets. After performing the initial analysis to derive the errors of each method, we proceed to investigate the pairwise correlations among them in order to understand the extent to which they produce errors in similar ways. Based on our results, we conclude that there is indeed a certain degree of correlation among the top-10 ranked methods, largely due to the fact that many of them consist of a combination of well-known, statistical and machine learning techniques. This fact has a strong impact on the results of the correlation analysis, and therefore leads to similar forecasting error patterns.

    KW - Forecasting

    KW - M4

    KW - Machine learning

    KW - Makridakis

    KW - Statistics

    UR - http://www.scopus.com/inward/record.url?scp=85068570399&partnerID=8YFLogxK

    U2 - 10.1016/j.ijforecast.2019.03.020

    DO - 10.1016/j.ijforecast.2019.03.020

    M3 - Comment/debate

    JO - International Journal of Forecasting

    JF - International Journal of Forecasting

    SN - 0169-2070

    ER -