Open AccessNote—Seasonal Exponential Smoothing with Damped Trends
Author(s) -
Everette S. Gardner,
E. McKenzie
Publication year - 1989
Publication title -
management science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 4.954
H-Index - 255
eISSN - 1526-5501
pISSN - 0025-1909
DOI - 10.1287/mnsc.35.3.372
Subject(s) - exponential smoothing , multiplicative function , smoothing , exponential function , econometrics , series (stratigraphy) , ex ante , mathematics , stability (learning theory) , mathematical optimization , computer science , statistics , economics , geology , mathematical analysis , paleontology , machine learning , macroeconomics
In this paper we apply the strategy of trend-damping to the popular Winters exponential smoothing systems for seasonal time series. Efficient model formulations are derived for both multiplicative and additive seasonal patterns. An algorithm is given to test the stability of the models in cases where predetermined smoothing parameters are used. Empirical results are presented to show that trend-damping improves ex ante forecast accuracy in seasonal data, especially at long leadtimes.forecasting: time series