Why and how does exponential smoothing fail? An in depth comparison of ATA-simple and simple exponential smoothing.

Abstract

Even though exponential smoothing (ES) is publicized as one of the most successful forecasting methods in the time series literature and it is widely used in practice due to its simplicity, its accuracy can be affected by the initialization and optimization procedures followed. It also suffers from some fundamental problems that can be seen clearly when its weighting scheme is studied closely. Exponential smoothing fails to account for the number of data points that can contribute to the forecast when assigning weights to historical data. ATA smoothing has been proposed as an alternative forecasting method and is shown to perform better than ES when the accuracies are compared on empirical data. In this paper, the properties of ATA that make it stand out from ES models will be discussed by just comparing the simple versions of both models. Empirical performance of the two simple models will also be compared based on popular error metrics.

Keywords

• Exponential smoothing,Forecasting,Initial value,M3-competition,Time series

References

1. Bailey, W. N., 1935. Generalized hypergeometric series. University Press Cambridge.
2. Box, G. E., Jenkins, G. M., Reinsel, G., 1970. Forecasting and control. Time Series Analysis 3, 75.
3. Brown, R. G., 1959. Statistical forecasting for inventory control. McGraw/Hill.
4. De Gooijer, J. G., Hyndman, R. J., 2005. 25 years of iif time series forecasting: a selective review. Tinbergen Institute Discussion Papers No. TI, 05-068.
5. Gardner, E. S., 1985. Exponential smoothing: The state of the art. Journal of forecasting 4 (1), 1-28.
6. Gardner, E. S., 2006. Exponential smoothing: The state of the artpart ii. International journal of forecasting 22 (4), 637-666.
7. Gardner Jr, E. S., McKenzie, E., 1985. Forecasting trends in time series. Management Science 31 (10), 1237-1246.
8. Goodwin, P., et al., 2010. The holt-winters approach to exponential smoothing: 50 years old and going strong. Foresight 19, 30-33.

9. Hyndman, R., Koehler, A. B., Ord, J. K., Snyder, R. D., 2008. Forecasting with exponential smoothing: the state space approach. Springer-Verlag.
10. Hyndman, R. J., Koehler, A. B., 2006. Another look at measures of forecast accuracy. International journal of forecasting 22 (4), 679-688.
11. Hyndman, R. J., Koehler, A. B., Snyder, R. D., Grose, S., 2002. A state space framework for automatic forecasting using exponential smoothing methods. International Journal of Forecasting 18 (3), 439-454.
12. Johnson, N. L., Kotz, S., 1977. Urn models and their application: an approach to modern discrete probability theory. Vol. 77. Wiley New York.
13. Makridakis, S., Hibon, M., 1991. Exponential smoothing: The eect of initial values and loss functions on post-sample forecasting accuracy. International Journal of Forecasting 7 (3), 317-330.
14. Makridakis, S., Hibon, M., 2000. The m3-competition: results, conclusions and implications. International journal of forecasting 16 (4), 451-476.
15. Makridakis, S. G., Andersen, A., Carbone, R., Fildes, R., Hibon, M., Lewandowski, R., Newton, J., Parzen, E., Winkler, R., 1984. The forecasting accuracy of major time series methods. Wiley.
16. Pegels, C. C., 1969. On startup or learning curves: An expanded view. AIIE Transactions 1 (3), 216-222.
17. Taylor, J. W., 2003. Exponential smoothing with a damped multiplicative trend. International journal of Forecasting 19 (4), 715-725.
18. Yapar, G., 2016. Modied simple exponential smoothing. Hacettepe University Journal of Mathematics and Statistics Early access, 0.
19. Yapar, G., Capar, S., Selamlar, H. T., Yavuz, I., 2016. Modied holt's linear trend method. Hacettepe University Journal of Mathematics and Statistics Accepted, 0.