Home > Research > Publications & Outputs > 'Horses for courses' in demand forecasting
View graph of relations

'Horses for courses' in demand forecasting

Research output: Contribution to journalJournal articlepeer-review

Published

Standard

'Horses for courses' in demand forecasting. / Petropoulos, Fotios; Makridakis, Spyros; Assimakopoulos, Vassilios; Nikolopoulos, Konstantinos.

In: European Journal of Operational Research, Vol. 237, No. 1, 16.08.2014, p. 152-163.

Research output: Contribution to journalJournal articlepeer-review

Harvard

Petropoulos, F, Makridakis, S, Assimakopoulos, V & Nikolopoulos, K 2014, ''Horses for courses' in demand forecasting', European Journal of Operational Research, vol. 237, no. 1, pp. 152-163. https://doi.org/10.1016/j.ejor.2014.02.036

APA

Petropoulos, F., Makridakis, S., Assimakopoulos, V., & Nikolopoulos, K. (2014). 'Horses for courses' in demand forecasting. European Journal of Operational Research, 237(1), 152-163. https://doi.org/10.1016/j.ejor.2014.02.036

Vancouver

Petropoulos F, Makridakis S, Assimakopoulos V, Nikolopoulos K. 'Horses for courses' in demand forecasting. European Journal of Operational Research. 2014 Aug 16;237(1):152-163. https://doi.org/10.1016/j.ejor.2014.02.036

Author

Petropoulos, Fotios ; Makridakis, Spyros ; Assimakopoulos, Vassilios ; Nikolopoulos, Konstantinos. / 'Horses for courses' in demand forecasting. In: European Journal of Operational Research. 2014 ; Vol. 237, No. 1. pp. 152-163.

Bibtex

@article{3d27cfb91c0c4e96a58075a7fc61d107,
title = "'Horses for courses' in demand forecasting",
abstract = "Forecasting as a scientific discipline has progressed a lot in the last forty years, with Nobel prizes being awarded for seminar work in the field, most notably to Engle, Granger and Kahneman. Despite these advances, even today we are unable to answer a very simple question, the one that is always the first tabled during discussions with practitioners: “what is the best method for my data?”. In essence, as there are horses for courses, there must also be forecasting methods that are more tailored to some types of data, and, therefore, enable practitioners to make informed method selection when facing new data. The current study attempts to shed light on this direction via identifying the main determinants of forecasting accuracy, through simulations and empirical investigations involving fourteen popular forecasting methods (and combinations of them), seven time series features (seasonality, trend, cycle, randomness, number of observations, inter-demand interval and coefficient of variation) and one strategic decision (the forecasting horizon). Our main findings dictate that forecasting accuracy is influenced as follows: a) for fast-moving data, cycle and randomness have the biggest (negative) effect and the longer the forecasting horizon, the more accuracy decreases; b) for intermittent data, inter-demand interval has bigger (negative) impact than the coefficient of variation; c) for all types of data, increasing the length of a series has a small positive effect.",
keywords = "Forecasting methods, Time series methods, Forecasting accuracy, M- Competitions, Simulation",
author = "Fotios Petropoulos and Spyros Makridakis and Vassilios Assimakopoulos and Konstantinos Nikolopoulos",
year = "2014",
month = aug,
day = "16",
doi = "10.1016/j.ejor.2014.02.036",
language = "English",
volume = "237",
pages = "152--163",
journal = "European Journal of Operational Research",
issn = "0377-2217",
publisher = "Elsevier Science B.V.",
number = "1",

}

RIS

TY - JOUR

T1 - 'Horses for courses' in demand forecasting

AU - Petropoulos, Fotios

AU - Makridakis, Spyros

AU - Assimakopoulos, Vassilios

AU - Nikolopoulos, Konstantinos

PY - 2014/8/16

Y1 - 2014/8/16

N2 - Forecasting as a scientific discipline has progressed a lot in the last forty years, with Nobel prizes being awarded for seminar work in the field, most notably to Engle, Granger and Kahneman. Despite these advances, even today we are unable to answer a very simple question, the one that is always the first tabled during discussions with practitioners: “what is the best method for my data?”. In essence, as there are horses for courses, there must also be forecasting methods that are more tailored to some types of data, and, therefore, enable practitioners to make informed method selection when facing new data. The current study attempts to shed light on this direction via identifying the main determinants of forecasting accuracy, through simulations and empirical investigations involving fourteen popular forecasting methods (and combinations of them), seven time series features (seasonality, trend, cycle, randomness, number of observations, inter-demand interval and coefficient of variation) and one strategic decision (the forecasting horizon). Our main findings dictate that forecasting accuracy is influenced as follows: a) for fast-moving data, cycle and randomness have the biggest (negative) effect and the longer the forecasting horizon, the more accuracy decreases; b) for intermittent data, inter-demand interval has bigger (negative) impact than the coefficient of variation; c) for all types of data, increasing the length of a series has a small positive effect.

AB - Forecasting as a scientific discipline has progressed a lot in the last forty years, with Nobel prizes being awarded for seminar work in the field, most notably to Engle, Granger and Kahneman. Despite these advances, even today we are unable to answer a very simple question, the one that is always the first tabled during discussions with practitioners: “what is the best method for my data?”. In essence, as there are horses for courses, there must also be forecasting methods that are more tailored to some types of data, and, therefore, enable practitioners to make informed method selection when facing new data. The current study attempts to shed light on this direction via identifying the main determinants of forecasting accuracy, through simulations and empirical investigations involving fourteen popular forecasting methods (and combinations of them), seven time series features (seasonality, trend, cycle, randomness, number of observations, inter-demand interval and coefficient of variation) and one strategic decision (the forecasting horizon). Our main findings dictate that forecasting accuracy is influenced as follows: a) for fast-moving data, cycle and randomness have the biggest (negative) effect and the longer the forecasting horizon, the more accuracy decreases; b) for intermittent data, inter-demand interval has bigger (negative) impact than the coefficient of variation; c) for all types of data, increasing the length of a series has a small positive effect.

KW - Forecasting methods

KW - Time series methods

KW - Forecasting accuracy

KW - M- Competitions

KW - Simulation

U2 - 10.1016/j.ejor.2014.02.036

DO - 10.1016/j.ejor.2014.02.036

M3 - Journal article

VL - 237

SP - 152

EP - 163

JO - European Journal of Operational Research

JF - European Journal of Operational Research

SN - 0377-2217

IS - 1

ER -