Comparer des méthodes

	La méthode Theta	Modèle ARIMA (Autoregressive Integrated Moving Average)	ETS Model
Domaine	Économétrie	Économétrie	Économétrie
Famille	Regression model	Regression model	Regression model
Année d'origine	2000	2015	2008
Auteur d'origine	Assimakopoulos & Nikolopoulos	Box & Jenkins (Box-Jenkins methodology)	Hyndman, Koehler, Ord & Snyder (state space framework)
Type	Univariate time-series forecasting model	Univariate time-series model	Exponential smoothing state space model
Source fondatrice	Assimakopoulos, V. & Nikolopoulos, K. (2000). The Theta Model: A Decomposition Approach to Forecasting. International Journal of Forecasting, 16(4), 521-530.	Box, G. E. P., Jenkins, G. M., Reinsel, G. C. & Ljung, G. M. (2015). Time Series Analysis: Forecasting and Control (5th ed.). Wiley. ISBN: 978-1118675021	Hyndman, R. J., Koehler, A. B., Ord, J. K. & Snyder, R. D. (2008). Forecasting with Exponential Smoothing: The State Space Approach. Springer.
Alias	theta model, theta forecasting, Theta Yöntemi — M3 Tahmin Yarışması Birincisi	Box-Jenkins model, ARIMA(p,d,q), ARIMA Modeli	exponential smoothing state space model, innovations state space model, Holt-Winters family, ETS — Hata/Trend/Mevsimsellik Üstel Düzleştirme
Apparentées	4	5	5
Résumé	The Theta Method is a univariate time-series forecasting model introduced by Assimakopoulos and Nikolopoulos in 2000. It decomposes a series into two theta lines that capture its long-run trend and its short-run dynamics, forecasts each line separately, and combines them by a weighted average. Its simplicity and accuracy made it the winner of the M3 forecasting competition.	ARIMA is a univariate time-series forecasting model that combines autoregressive, integrated (differencing), and moving-average components to predict a single continuous series from its own past. It is the centrepiece of the Box-Jenkins methodology set out in Box, Jenkins, Reinsel & Ljung's Time Series Analysis (5th ed., 2015).	ETS is a comprehensive exponential smoothing framework that automatically selects additive or multiplicative combinations of the error (E), trend (T) and seasonal (S) components of a time series. Formalised as an innovations state space model by Hyndman, Koehler, Ord and Snyder in 2008, it unifies and generalises the Holt-Winters family of forecasting methods.