Home > Research > Publications & Outputs > Online Detection of Forecast Model Inadequacies...

Links

Text available via DOI:

View graph of relations

Online Detection of Forecast Model Inadequacies Using Forecast Errors

Research output: Contribution to Journal/MagazineJournal articlepeer-review

E-pub ahead of print

Standard

Online Detection of Forecast Model Inadequacies Using Forecast Errors. / Grundy, Thomas; Killick, Rebecca; Svetunkov, Ivan.
In: Journal of Time Series Analysis, 11.06.2025.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Grundy T, Killick R, Svetunkov I. Online Detection of Forecast Model Inadequacies Using Forecast Errors. Journal of Time Series Analysis. 2025 Jun 11. Epub 2025 Jun 11. doi: 10.1111/jtsa.12843

Author

Bibtex

@article{3ee0e69153da4bb0b32c78920f0ed73f,
title = "Online Detection of Forecast Model Inadequacies Using Forecast Errors",
abstract = "In many organizations, accurate forecasts are essential for making informed decisions in a variety of applications, from inventory management to staffing optimization. Whatever forecasting model is used, changes in the underlying process can lead to inaccurate forecasts, which will be damaging to decision‐making. At the same time, models are becoming increasingly complex, and identifying change through direct modeling is problematic. We present a novel framework for online monitoring of forecasts to ensure they remain accurate. By utilizing sequential changepoint techniques on the forecast errors, our framework allows for the real‐time identification of potential changes in the process caused by various external factors. We show theoretically that some common changes in the underlying process will manifest in the forecast errors and can be identified faster by identifying shifts in the forecast errors than within the original modeling framework. Moreover, we demonstrate the effectiveness of this framework on numerous forecasting approaches through simulations and show its effectiveness over alternative approaches. Finally, we present two concrete examples, one from Royal Mail parcel delivery volumes and one from NHS A&E admissions relating to gallstones.",
keywords = "concept drift, structural break, sequential change, changepoint",
author = "Thomas Grundy and Rebecca Killick and Ivan Svetunkov",
year = "2025",
month = jun,
day = "11",
doi = "10.1111/jtsa.12843",
language = "English",
journal = "Journal of Time Series Analysis",
issn = "0143-9782",
publisher = "Wiley-Blackwell",

}

RIS

TY - JOUR

T1 - Online Detection of Forecast Model Inadequacies Using Forecast Errors

AU - Grundy, Thomas

AU - Killick, Rebecca

AU - Svetunkov, Ivan

PY - 2025/6/11

Y1 - 2025/6/11

N2 - In many organizations, accurate forecasts are essential for making informed decisions in a variety of applications, from inventory management to staffing optimization. Whatever forecasting model is used, changes in the underlying process can lead to inaccurate forecasts, which will be damaging to decision‐making. At the same time, models are becoming increasingly complex, and identifying change through direct modeling is problematic. We present a novel framework for online monitoring of forecasts to ensure they remain accurate. By utilizing sequential changepoint techniques on the forecast errors, our framework allows for the real‐time identification of potential changes in the process caused by various external factors. We show theoretically that some common changes in the underlying process will manifest in the forecast errors and can be identified faster by identifying shifts in the forecast errors than within the original modeling framework. Moreover, we demonstrate the effectiveness of this framework on numerous forecasting approaches through simulations and show its effectiveness over alternative approaches. Finally, we present two concrete examples, one from Royal Mail parcel delivery volumes and one from NHS A&E admissions relating to gallstones.

AB - In many organizations, accurate forecasts are essential for making informed decisions in a variety of applications, from inventory management to staffing optimization. Whatever forecasting model is used, changes in the underlying process can lead to inaccurate forecasts, which will be damaging to decision‐making. At the same time, models are becoming increasingly complex, and identifying change through direct modeling is problematic. We present a novel framework for online monitoring of forecasts to ensure they remain accurate. By utilizing sequential changepoint techniques on the forecast errors, our framework allows for the real‐time identification of potential changes in the process caused by various external factors. We show theoretically that some common changes in the underlying process will manifest in the forecast errors and can be identified faster by identifying shifts in the forecast errors than within the original modeling framework. Moreover, we demonstrate the effectiveness of this framework on numerous forecasting approaches through simulations and show its effectiveness over alternative approaches. Finally, we present two concrete examples, one from Royal Mail parcel delivery volumes and one from NHS A&E admissions relating to gallstones.

KW - concept drift

KW - structural break

KW - sequential change

KW - changepoint

U2 - 10.1111/jtsa.12843

DO - 10.1111/jtsa.12843

M3 - Journal article

JO - Journal of Time Series Analysis

JF - Journal of Time Series Analysis

SN - 0143-9782

ER -