Forecasting in social settings: the state of the art

Spyros Makridakis, Rob J. Hyndman, Fotios Petropoulos

Research output: Contribution to journalArticle

Abstract

This paper provides a non-systematic review of the progress of forecasting in social settings. It is aimed at someone outside the field of forecasting who wants to understand and appreciate the results of the M4 Competition, and forms a survey paper regarding the state of the art of this discipline. It discusses the recorded improvements in forecast accuracy over time, the need to capture forecast uncertainty, and things that can go wrong with predictions. Subsequently, the review classifies the knowledge achieved over recent years into (i) what we know, (ii) what we are not sure about, and (iii) what we don’t knowIn the first two areas, we explore the difference between explanation and prediction, the existence of an optimal model, the performance of machine learning methods on time series forecasting tasks, the difficulties of predicting non-stable environments, the performance of judgment, and the value added by exogenous variables. The article concludes with the importance of (thin and) fat tails, the challenges and advances in causal inference, and the role of luck.
Original languageEnglish
Pages (from-to)1-14
Number of pages14
JournalInternational Journal of Forecasting
Early online date26 Sep 2019
DOIs
Publication statusE-pub ahead of print - 26 Sep 2019

Keywords

  • Review
  • knowns and unknowns
  • accuracy
  • uncertainty
  • judgment
  • causality
  • machine learning

Cite this

Forecasting in social settings: the state of the art. / Makridakis, Spyros; Hyndman, Rob J.; Petropoulos, Fotios.

In: International Journal of Forecasting, 26.09.2019, p. 1-14.

Research output: Contribution to journalArticle

@article{fcc2f82efa1d49f28cb84e7ea92f8be2,
title = "Forecasting in social settings: the state of the art",
abstract = "This paper provides a non-systematic review of the progress of forecasting in social settings. It is aimed at someone outside the field of forecasting who wants to understand and appreciate the results of the M4 Competition, and forms a survey paper regarding the state of the art of this discipline. It discusses the recorded improvements in forecast accuracy over time, the need to capture forecast uncertainty, and things that can go wrong with predictions. Subsequently, the review classifies the knowledge achieved over recent years into (i) what we know, (ii) what we are not sure about, and (iii) what we don’t knowIn the first two areas, we explore the difference between explanation and prediction, the existence of an optimal model, the performance of machine learning methods on time series forecasting tasks, the difficulties of predicting non-stable environments, the performance of judgment, and the value added by exogenous variables. The article concludes with the importance of (thin and) fat tails, the challenges and advances in causal inference, and the role of luck.",
keywords = "Review, knowns and unknowns, accuracy, uncertainty, judgment, causality, machine learning",
author = "Spyros Makridakis and Hyndman, {Rob J.} and Fotios Petropoulos",
year = "2019",
month = "9",
day = "26",
doi = "10.1016/j.ijforecast.2019.05.011",
language = "English",
pages = "1--14",
journal = "International Journal of Forecasting",
issn = "0169-2070",
publisher = "Elsevier",

}

TY - JOUR

T1 - Forecasting in social settings: the state of the art

AU - Makridakis, Spyros

AU - Hyndman, Rob J.

AU - Petropoulos, Fotios

PY - 2019/9/26

Y1 - 2019/9/26

N2 - This paper provides a non-systematic review of the progress of forecasting in social settings. It is aimed at someone outside the field of forecasting who wants to understand and appreciate the results of the M4 Competition, and forms a survey paper regarding the state of the art of this discipline. It discusses the recorded improvements in forecast accuracy over time, the need to capture forecast uncertainty, and things that can go wrong with predictions. Subsequently, the review classifies the knowledge achieved over recent years into (i) what we know, (ii) what we are not sure about, and (iii) what we don’t knowIn the first two areas, we explore the difference between explanation and prediction, the existence of an optimal model, the performance of machine learning methods on time series forecasting tasks, the difficulties of predicting non-stable environments, the performance of judgment, and the value added by exogenous variables. The article concludes with the importance of (thin and) fat tails, the challenges and advances in causal inference, and the role of luck.

AB - This paper provides a non-systematic review of the progress of forecasting in social settings. It is aimed at someone outside the field of forecasting who wants to understand and appreciate the results of the M4 Competition, and forms a survey paper regarding the state of the art of this discipline. It discusses the recorded improvements in forecast accuracy over time, the need to capture forecast uncertainty, and things that can go wrong with predictions. Subsequently, the review classifies the knowledge achieved over recent years into (i) what we know, (ii) what we are not sure about, and (iii) what we don’t knowIn the first two areas, we explore the difference between explanation and prediction, the existence of an optimal model, the performance of machine learning methods on time series forecasting tasks, the difficulties of predicting non-stable environments, the performance of judgment, and the value added by exogenous variables. The article concludes with the importance of (thin and) fat tails, the challenges and advances in causal inference, and the role of luck.

KW - Review

KW - knowns and unknowns

KW - accuracy

KW - uncertainty

KW - judgment

KW - causality

KW - machine learning

U2 - 10.1016/j.ijforecast.2019.05.011

DO - 10.1016/j.ijforecast.2019.05.011

M3 - Article

SP - 1

EP - 14

JO - International Journal of Forecasting

JF - International Journal of Forecasting

SN - 0169-2070

ER -