Kybernetika 48 no. 1, 1-15, 2012

Holt-Winters method with general seasonality

Tomáš Hanzák

Abstract:

The paper suggests a generalization of widely used Holt-Winters smoothing and forecasting method for seasonal time series. The general concept of seasonality modeling is introduced both for the additive and multiplicative case. Several special cases are discussed, including a linear interpolation of seasonal indices and a usage of trigonometric functions. Both methods are fully applicable for time series with irregularly observed data (just the special case of missing observations was covered up to now). Moreover, they sometimes outperform the classical Holt-Winters method even for regular time series. A simulation study and real data examples compare the suggested methods with the classical one.

Keywords:

exponential smoothing, Holt-Winters method, irregular time series, seasonal indices, trigonometric functions

Classification:

62M10, 62M20, 60G35, 65D10

paper.pdf

References:

  1. M. Aldrin and E. Damsleth: Forecasting non-seasonal time series with missing observations. J. Forecasting 8 (1989), 97-116.   CrossRef
  2. C. Chatfield and M. Yar: Holt-Winters forecasting: some practical issues. The Statistician 37 (1988), 129-140.   CrossRef
  3. T. Cipra and T. Hanzák: Exponential smoothing for irregular time series. Kybernetika 44 (2008), 385-399.   CrossRef
  4. T. Cipra, J. Trujillo and A. Rubio: Holt-Winters method with missing observations. Management Sci. 41 (1995), 174-178.   CrossRef
  5. E. S. Gardner: Exponential smoothing: The state of the art. J. Forecasting 4 (1985), 1-28.   CrossRef
  6. E. S. Gardner: Exponential smoothing: The state of the art - Part II. Internat. J. Forecasting 22 (2006), 637-666.   CrossRef
  7. T. Hanzák: Improved Holt method for irregular time series. In: WDS'08 Proc. Contributed Papers, Part I - Mathematics and Computer Sciences, Matfyzpress, Prague 2008, pp. 62-67.   CrossRef
  8. C. C. Holt: Forecasting seasonals and trends by exponentially weighted moving averages. Internat. J. Forecasting 20 (2004), 5-10.   CrossRef
  9. R. J. Hyndman: Time Series Data Library, www.robhyndman.info/TSDL. Accessed on 26 June 2010.   CrossRef
  10. R. J. Hyndman, A. B. Koehler, R. D. Snyder and S. Grose: A state space framework for automatic forecasting using exponential smoothing methods. Internat. J. Forecasting 18 (2002), 439-454.   CrossRef
  11. T. Ratinger: Seasonal time series with missing observations. Appl. Math. 41 (1996), 41-55.   CrossRef
  12. J. W. Taylor: Short-term electricity demand forecasting using double seasonal exponential smoothing. J. Oper. Res. Soc. 54 (2003), 799-805.   CrossRef
  13. J. W. Taylor: A comparison of univariate time series methods for forecasting intraday arrivals at a call center. Management Sci. 54 (2008), 253-265.   CrossRef
  14. P. R. Winters: Forecasting sales by exponentially weighted moving averages. Management Sci. 6 (1960), 324-342.   CrossRef
  15. D. J. Wright: Forecasting data published at irregular time intervals using extension of Holt's method. Management Sci. 32 (1986), 499-510.   CrossRef