A Simple Combination of Univariate Models

Fotios Petropoulos, Ivan Svetunkov

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper describes the approach that we implemented for producing the point forecasts and prediction intervals for our M4-competition submission. The proposed simple combination of univariate models (SCUM) is a median combination of the point forecasts and prediction intervals of four models, namely exponential smoothing, complex exponential smoothing, automatic autoregressive integrated moving average and dynamic optimised theta. Our submission performed very well in the M4-competition, being ranked 6 th for the point forecasts (with a small difference compared to the 2 nd submission) and prediction intervals and 2 nd and 3 rd for the point forecasts of the weekly and quarterly data respectively.

Original languageEnglish
Pages (from-to)110-115
Number of pages6
JournalInternational Journal of Forecasting
Volume36
Issue number1
Early online date17 Apr 2019
DOIs
Publication statusE-pub ahead of print - 17 Apr 2019

Keywords

  • ARIMA
  • Complex exponential smoothing
  • ETS
  • M4-competition
  • Median combination
  • Theta method

ASJC Scopus subject areas

  • Business and International Management

Cite this

A Simple Combination of Univariate Models. / Petropoulos, Fotios; Svetunkov, Ivan.

In: International Journal of Forecasting, Vol. 36, No. 1, 01.01.2020, p. 110-115.

Research output: Contribution to journalArticle

Petropoulos, Fotios ; Svetunkov, Ivan. / A Simple Combination of Univariate Models. In: International Journal of Forecasting. 2020 ; Vol. 36, No. 1. pp. 110-115.
@article{0d772dceda3c47eb84e6e2193febd5e1,
title = "A Simple Combination of Univariate Models",
abstract = "This paper describes the approach that we implemented for producing the point forecasts and prediction intervals for our M4-competition submission. The proposed simple combination of univariate models (SCUM) is a median combination of the point forecasts and prediction intervals of four models, namely exponential smoothing, complex exponential smoothing, automatic autoregressive integrated moving average and dynamic optimised theta. Our submission performed very well in the M4-competition, being ranked 6 th for the point forecasts (with a small difference compared to the 2 nd submission) and prediction intervals and 2 nd and 3 rd for the point forecasts of the weekly and quarterly data respectively.",
keywords = "ARIMA, Complex exponential smoothing, ETS, M4-competition, Median combination, Theta method",
author = "Fotios Petropoulos and Ivan Svetunkov",
year = "2019",
month = "4",
day = "17",
doi = "10.1016/j.ijforecast.2019.01.006",
language = "English",
volume = "36",
pages = "110--115",
journal = "International Journal of Forecasting",
issn = "0169-2070",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - A Simple Combination of Univariate Models

AU - Petropoulos, Fotios

AU - Svetunkov, Ivan

PY - 2019/4/17

Y1 - 2019/4/17

N2 - This paper describes the approach that we implemented for producing the point forecasts and prediction intervals for our M4-competition submission. The proposed simple combination of univariate models (SCUM) is a median combination of the point forecasts and prediction intervals of four models, namely exponential smoothing, complex exponential smoothing, automatic autoregressive integrated moving average and dynamic optimised theta. Our submission performed very well in the M4-competition, being ranked 6 th for the point forecasts (with a small difference compared to the 2 nd submission) and prediction intervals and 2 nd and 3 rd for the point forecasts of the weekly and quarterly data respectively.

AB - This paper describes the approach that we implemented for producing the point forecasts and prediction intervals for our M4-competition submission. The proposed simple combination of univariate models (SCUM) is a median combination of the point forecasts and prediction intervals of four models, namely exponential smoothing, complex exponential smoothing, automatic autoregressive integrated moving average and dynamic optimised theta. Our submission performed very well in the M4-competition, being ranked 6 th for the point forecasts (with a small difference compared to the 2 nd submission) and prediction intervals and 2 nd and 3 rd for the point forecasts of the weekly and quarterly data respectively.

KW - ARIMA

KW - Complex exponential smoothing

KW - ETS

KW - M4-competition

KW - Median combination

KW - Theta method

UR - http://www.scopus.com/inward/record.url?scp=85064319785&partnerID=8YFLogxK

U2 - 10.1016/j.ijforecast.2019.01.006

DO - 10.1016/j.ijforecast.2019.01.006

M3 - Article

VL - 36

SP - 110

EP - 115

JO - International Journal of Forecasting

JF - International Journal of Forecasting

SN - 0169-2070

IS - 1

ER -