Anda di halaman 1dari 4

See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/243764019

The Forecasting Accuracy of Major Time-Series


Methods
Article in Journal of the American Statistical Association March 1986
DOI: 10.2307/2288019

CITATIONS

READS

94

239

9 authors, including:
Spyros Makridakis

Robert Fildes

Neapolis University

Lancaster University

152 PUBLICATIONS 7,173 CITATIONS

132 PUBLICATIONS 3,600 CITATIONS

SEE PROFILE

SEE PROFILE

Michele Hibon

Emanuel Parzen

INSEAD

Texas A&M University

22 PUBLICATIONS 2,473 CITATIONS

92 PUBLICATIONS 4,049 CITATIONS

SEE PROFILE

SEE PROFILE

Available from: Spyros Makridakis


Retrieved on: 21 August 2016

The Forecasting Accuracy of Major Time Series Methods. by S. Makridakis; A. Andersen; R.


Carbone; R. Fildes; M. Hibon; R. Lewandowski; J. Newton; E. Parzen; R. Winkler
Review by: Oliver D. Anderson
Journal of the Royal Statistical Society. Series D (The Statistician), Vol. 34, No. 2 (1985), pp.
261-262
Published by: Wiley for the Royal Statistical Society
Stable URL: http://www.jstor.org/stable/2988175 .
Accessed: 05/01/2015 13:36
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Wiley and Royal Statistical Society are collaborating with JSTOR to digitize, preserve and extend access to
Journal of the Royal Statistical Society. Series D (The Statistician).

http://www.jstor.org

This content downloaded from 194.221.86.126 on Mon, 5 Jan 2015 13:36:18 PM


All use subject to JSTOR Terms and Conditions

Book Reviews

261

the monographis publishedin the veryusefulSage series'QuantitativeApplications


in the Social Sciences'. Examples are given but it is oftenthe case that aftera good
deal of notation and theorythe reader gets to an example which statessomething
along the lines 'fromequation x.x it can be easily derivedbe that C =some figure'.
to see how to
Withoutunderstanding
the priortheoryand notationit is oftendifficult
performthe sometimescomplex calculations. Presumably to include all detailed
workingsfor each example would have extendedthe textconsiderablyand perhaps
beyond an acceptable length. Statisticians,though, will have little difficultyin
understanding
thismonograph.
I also noted severaltypographicalerrorsbut theseare perhapsminorquibbles as on
the whole thisis a welcome additionto the literatureon thisarea of statistics.
ROGER TARLING

Home OfficeResearchand PlanningUnit

The ForecastingAccuracyof Major Time Series Methods


S. MAKRIDAKIS, A. ANDERSEN, R. CARBONE, R. FILDES, M. HIBON, R.
LEWANDOWSKI, J. NEWTON, E. PARZEN & R. WINKLER
JohnWiley & Sons, New York, 1984
pp. viii+ 301, ?19.50
ISBN 0 471 90327 2.
This is a beautifullyproduced book, mainly of re-typesetpapers already in the
literature.As I have seen virtuallyall the materialbefore,I did not rereadit, so can
not commenton thelevel of misprints-althoughWileytendto be prettygood on that
score.
I would not considerbuyingthebook myself.I was originallyinvitedto take partin
the 1001 case
the main project,and spend six monthsat INSEAD 'Box-Jenkinsing'
series. This I declined, as I feltthe studywas undesigned,and could provide little
valid comparison between the various contendingforecastingapproaches. What I
underestimatedwas the lead author's abilityto get thingsstartedand keep them
rolling.SpyrosMakridakisis a veryremarkableman.
I stillfeelthattheForecastingCompetitionwas misconceivedin detail;and, relative
to the othereffortinvolved,could veryeasily have been properlydesignedand more
appropriatelyspecified.Do we really want to compare a host of methodsapplied
and withoutimagination?(Compare the commentsmade by Paul
indiscriminately
Newbold and David Pack in the 1983 Commentary.Also see relevantremarksby the
late GwilymJenkins,1982.) However,withoutMakridakis'drivingforce,the experithe impact
mentwould not have taken place; and, as is usual withhis involvements,
has in factbeen enormous.If only to get literallythousandsof people discussingthe
problemsinvolved.
Wiley know theyhave a best seller,and copies of thisbook will be snapped up by
everyoneconcernedwithforecasting,
exceptStatisticians.For thisreason,if no other,
Statisticians(especiallythosein Academia) should read thiswork.What is the use of
being veryscholarly,if no-one (exceptotherscholars)pays the least attentionto what
you write?This is a book which will influencepeople (and should be allowed to do
so). It is not just mathematicalrigour which is importantin practice: it is the
implementationof ideas that counts. Also, some of the discussionsare quite sound
even on an academic level. Contributorslike Allan Andersenand Emanuel Parzen,
forinstance,are quite respectablestatisticalauthors.

This content downloaded from 194.221.86.126 on Mon, 5 Jan 2015 13:36:18 PM


All use subject to JSTOR Terms and Conditions

262

Book Reviews

Contents:
Forecasting:State of theArt (S. Makridakis)
ForecastingwithEconomic Methods:FolkloreversusFact (J. S. Armstrong)
Accuracyof Forecasting:An EmpiricalInvestigation(S. Makridakisand M. Hibon)
The Accuracy of Extrapolation(Time Series) Methods: Results of a Forecasting
Competition(S. Makridakis,A. Andersen,R. Carbone, R. Fildes, M. Hibon, R.
Lewandowski,J. Newton,E. Parzen,and R. Winkler.
Forecasting:The Box-JenkinsApproach (A. Andersenand A. Weiss)
AEP Filtering(R. Carbone, R. Bilongo,P. Piat-Corson,and S. Nadeau)
Bayesian Forecasting(R. Fildes)
Naive, Moving Average, ExponentialSmoothing,and RegressionMethods (M. Hibon)
Lewandowski'sFORSYS Method (R. Lewandowski)
Sales Forecastingby FORSYS (R. Lewandowski)
Forecastingand Time SeriesModel Types of 111 EconomicTime Series(H.J. Newton
and E. Parzen)
CombiningForecasts(R. L. Winkler)
OLIVER D. ANDERSON

TSA&F, Nottingham

References
in organisations,Journalof Forecasting,1, pp.
JENKINS,G.M. (1982) Some practicalaspects of forcasting
3-21.

PACK, D.J. (1983) in, Commentaryon the Makridakis Time Series Competition
(M-Competition),Journalof Forecasting,2, pp. 259-311.

NEWBOLD, P. &

This content downloaded from 194.221.86.126 on Mon, 5 Jan 2015 13:36:18 PM


All use subject to JSTOR Terms and Conditions

Anda mungkin juga menyukai