Replications of forecasting research

H. Evanschitzky, J. Armstrong

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

We have examined the frequency of replications published in the two leading forecasting journals, the International Journal of Forecasting (IJF) and the Journal of Forecasting (JoF). Replications in the IJF and JoF between 1996 and 2008 comprised 8.4% of the empirical papers. Various other areas of management science have values ranging from 2.2% in the Journal of Marketing Research to 18.1% in the American Economic Review. We also found that 35.3% of the replications in forecasting journals provided full support for the findings of the initial study, 45.1% provided partial support, and 19.6% provided no support. Given the importance of replications, we recommend various steps to encourage replications, such as requiring a full disclosure of the methods and data used for all published papers, and inviting researchers to replicate specific important papers.
LanguageEnglish
Pages4-8
JournalInternational Journal of Forecasting
Volume26
Issue number1
Early online date16 Oct 2009
DOIs
Publication statusPublished - Mar 2010

Fingerprint

Replication
Economics
Management science
Disclosure
Marketing research

Keywords

  • replication research
  • research policy
  • census study
  • forecasting

Cite this

Evanschitzky, H. ; Armstrong, J. / Replications of forecasting research. In: International Journal of Forecasting. 2010 ; Vol. 26, No. 1. pp. 4-8.
@article{5a2a787e7a9546538c0e0ec81e9af92a,
title = "Replications of forecasting research",
abstract = "We have examined the frequency of replications published in the two leading forecasting journals, the International Journal of Forecasting (IJF) and the Journal of Forecasting (JoF). Replications in the IJF and JoF between 1996 and 2008 comprised 8.4{\%} of the empirical papers. Various other areas of management science have values ranging from 2.2{\%} in the Journal of Marketing Research to 18.1{\%} in the American Economic Review. We also found that 35.3{\%} of the replications in forecasting journals provided full support for the findings of the initial study, 45.1{\%} provided partial support, and 19.6{\%} provided no support. Given the importance of replications, we recommend various steps to encourage replications, such as requiring a full disclosure of the methods and data used for all published papers, and inviting researchers to replicate specific important papers.",
keywords = "replication research, research policy, census study, forecasting",
author = "H. Evanschitzky and J. Armstrong",
year = "2010",
month = "3",
doi = "10.1016/j.ijforecast.2009.09.003",
language = "English",
volume = "26",
pages = "4--8",
journal = "International Journal of Forecasting",
issn = "0169-2070",
number = "1",

}

Replications of forecasting research. / Evanschitzky, H.; Armstrong, J.

In: International Journal of Forecasting, Vol. 26, No. 1, 03.2010, p. 4-8.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Replications of forecasting research

AU - Evanschitzky, H.

AU - Armstrong, J.

PY - 2010/3

Y1 - 2010/3

N2 - We have examined the frequency of replications published in the two leading forecasting journals, the International Journal of Forecasting (IJF) and the Journal of Forecasting (JoF). Replications in the IJF and JoF between 1996 and 2008 comprised 8.4% of the empirical papers. Various other areas of management science have values ranging from 2.2% in the Journal of Marketing Research to 18.1% in the American Economic Review. We also found that 35.3% of the replications in forecasting journals provided full support for the findings of the initial study, 45.1% provided partial support, and 19.6% provided no support. Given the importance of replications, we recommend various steps to encourage replications, such as requiring a full disclosure of the methods and data used for all published papers, and inviting researchers to replicate specific important papers.

AB - We have examined the frequency of replications published in the two leading forecasting journals, the International Journal of Forecasting (IJF) and the Journal of Forecasting (JoF). Replications in the IJF and JoF between 1996 and 2008 comprised 8.4% of the empirical papers. Various other areas of management science have values ranging from 2.2% in the Journal of Marketing Research to 18.1% in the American Economic Review. We also found that 35.3% of the replications in forecasting journals provided full support for the findings of the initial study, 45.1% provided partial support, and 19.6% provided no support. Given the importance of replications, we recommend various steps to encourage replications, such as requiring a full disclosure of the methods and data used for all published papers, and inviting researchers to replicate specific important papers.

KW - replication research

KW - research policy

KW - census study

KW - forecasting

UR - http://www.scopus.com/inward/record.url?scp=74149086787&partnerID=8YFLogxK

U2 - 10.1016/j.ijforecast.2009.09.003

DO - 10.1016/j.ijforecast.2009.09.003

M3 - Article

VL - 26

SP - 4

EP - 8

JO - International Journal of Forecasting

T2 - International Journal of Forecasting

JF - International Journal of Forecasting

SN - 0169-2070

IS - 1

ER -