A note on upper bounds for forecast-value-added relative to naive forecasts

Research output: Contribution to journalArticle

Abstract

In Forecast Value Added (FVA) analysis the accuracy of relatively sophisticated forecasting methods is compared to that of naïve 1 forecasts to see if the extra costs and effort of implementing them is justified. In this note we derive a ratio that indicates the upper bound of a forecasting method's accuracy relative to naïve 1 forecasts when the mean squared error is used to measure one-period-ahead accuracy. The ratio is applicable when a series is stationary or when its first differences are
LanguageEnglish
Pages1082-1084
JournalJournal of the Operational Research Society
Volume68
Issue number9
Early online date27 Mar 2017
DOIs
StatusPublished - Sep 2017

Fingerprint

Value engineering
Costs
Upper bound
Value added
Forecasting method
Mean squared error

Cite this

A note on upper bounds for forecast-value-added relative to naive forecasts. / Goodwin, Paul; Petropoulos, Fotios; Hyndman, Rob J.

In: Journal of the Operational Research Society, Vol. 68, No. 9, 09.2017, p. 1082-1084.

Research output: Contribution to journalArticle

@article{a2b084c786254b36becdc7435f45908a,
title = "A note on upper bounds for forecast-value-added relative to naive forecasts",
abstract = "In Forecast Value Added (FVA) analysis the accuracy of relatively sophisticated forecasting methods is compared to that of na{\"i}ve 1 forecasts to see if the extra costs and effort of implementing them is justified. In this note we derive a ratio that indicates the upper bound of a forecasting method's accuracy relative to na{\"i}ve 1 forecasts when the mean squared error is used to measure one-period-ahead accuracy. The ratio is applicable when a series is stationary or when its first differences are",
author = "Paul Goodwin and Fotios Petropoulos and Hyndman, {Rob J.}",
year = "2017",
month = "9",
doi = "10.1057/s41274-017-0218-3",
language = "English",
volume = "68",
pages = "1082--1084",
journal = "Journal of the Operational Research Society",
issn = "0160-5682",
publisher = "Palgrave Macmillan",
number = "9",

}

TY - JOUR

T1 - A note on upper bounds for forecast-value-added relative to naive forecasts

AU - Goodwin,Paul

AU - Petropoulos,Fotios

AU - Hyndman,Rob J.

PY - 2017/9

Y1 - 2017/9

N2 - In Forecast Value Added (FVA) analysis the accuracy of relatively sophisticated forecasting methods is compared to that of naïve 1 forecasts to see if the extra costs and effort of implementing them is justified. In this note we derive a ratio that indicates the upper bound of a forecasting method's accuracy relative to naïve 1 forecasts when the mean squared error is used to measure one-period-ahead accuracy. The ratio is applicable when a series is stationary or when its first differences are

AB - In Forecast Value Added (FVA) analysis the accuracy of relatively sophisticated forecasting methods is compared to that of naïve 1 forecasts to see if the extra costs and effort of implementing them is justified. In this note we derive a ratio that indicates the upper bound of a forecasting method's accuracy relative to naïve 1 forecasts when the mean squared error is used to measure one-period-ahead accuracy. The ratio is applicable when a series is stationary or when its first differences are

UR - https://doi.org/10.1057/s41274-017-0218-3

U2 - 10.1057/s41274-017-0218-3

DO - 10.1057/s41274-017-0218-3

M3 - Article

VL - 68

SP - 1082

EP - 1084

JO - Journal of the Operational Research Society

T2 - Journal of the Operational Research Society

JF - Journal of the Operational Research Society

SN - 0160-5682

IS - 9

ER -