Exponential smoothing model selection for forecasting

Abstract

Applications of exponential smoothing to forecast time series usually rely on three basic methods: simple exponential smoothing, trend corrected exponential smoothing and a seasonal variation thereof. A common approach to select the method appropriate to a particular time series is based on prediction validation on a withheld part of the sample using criteria such as the mean absolute percentage error. A second approach is to rely on the most appropriate general case of the three methods. For annual series this is trend corrected exponential smoothing: for sub-annual series it is the seasonal adaptation of trend corrected exponential smoothing. The rationale for this approach is that a general method automatically collapses to its nested counterparts when the pertinent conditions pertain in the data. A third approach may be based on an information criterion when maximum likelihood methods are used in conjunction with exponential smoothing to estimate the smoothing parameters. In this paper, such approaches for selecting the appropriate forecasting method are compared in a simulation study. They are also compared on real time series from the M3 forecasting competition. The results indicate that the information criterion approach appears to provide the best basis for an automated approach to method selection, provided that it is based on Akaike's information criterion.

Another look at measures of forecast accuracy

“…Billah et al (2005) used a similar error measure when they computed the absolute value of the forecast error as a percentage of the insample standard deviation. However, this approach has the disadvantage that the denominator grows with the sample size for non-stationary series containing a unit root.…”

Analysis and forecast of Turkey unemployment rate

“…They are commonly used in inventory control systems where many items are to be forecast and low cost is a primary concern. No clear theory exists for deciding which of these information criteria is best suited for choosing the appropriate method of exponential smoothing [3].…”

Previsión de demanda intermitente con métodos de series de tiempo y redes neuronales artificiales: Estudio de caso

“…La suavización exponencial (SE) es uno de los principales métodos clásicos de previsión cuantitativa, es sencillo, robusto y ampliamente utilizado para la previsión de la demanda [6]. Los valores pasados de la serie de tiempo son ponderados y desvalorizados en relación a los más recientes.…”