Kybernetika 51 no. 6, 973-993, 2015

Exponential smoothing based on L-estimation

Přemysl Bejda and Tomáš CipraDOI: 10.14736/kyb-2015-6-0973

Abstract:

Robust methods similar to exponential smoothing are suggested in this paper. First previous results for exponential smoothing in $L_1$ are generalized using the regression quantiles, including a generalization to more parameters. Then a method based on the classical sign test is introduced that should deal not only with outliers but also with level shifts, including a detection of change points. Properties of various approaches are investigated by means of a simulation study. A real data example is used as an illustration.

Keywords:

exponential smoothing, change point, quantiles, robust methods, sign test

Classification:

62M10, 62M20, 62N02

References:

  1. R. G. Brown: Smoothing, Forecasting and Prediction of Descrete Time Series. Prentice-Hall, Englewood Cliffs, NJ 1962.   CrossRef
  2. T. Cipra: Robust exponential smoothing. J. Forecasting 11 (1992), 1, 57-69.   DOI:10.1002/for.3980110106
  3. T. Cipra and R. Romera: Recursive time series methods in $L_1$-norm. In: $L_1$-Statistical Analysis and Related Methods (Y. Dodge ed.), pp. 233-243, 1992.   DOI:10.2307/2289888
  4. S. Gelper, R. Fried and Ch. Croux: Robust forecasting with exponential and Holt-Winters smoothing. J. Forecasting 29 (2010), 3, 285-300.   DOI:10.2139/ssrn.1089403
  5. T. Hanzák and T. Cipra: Exponential smoothing for time series with outliers. Kybernetika 47 (2011), 2, 165-178.   CrossRef
  6. Ch. C. Holt: Forecasting seasonals and trends by exponentially weighted moving averages. Int. J. Forecasting 20 (2004), 5-10.   DOI:10.1016/j.ijforecast.2003.09.015
  7. R. J. Hyndman, A. B. Koehler, J. K. Ord and R. D. Snyder: Forecasting with Exponential Smoothing. Springer, Berlin - Heidelberg 2008.   DOI:10.1007/978-3-540-71918-2
  8. J. Jurečková and J. Picek: Robust Statistical Methods with R. Chapman and Hall/CRC, 2005.   DOI:10.1201/9781420035131
  9. R. Koenker and G. Bassett: Regression quantiles. Econometrica 46 (1978), 1, 33-50.   DOI:10.2307/1913643
  10. A. Koubková: Forecasting seasonals and trends by exponentially weighted moving averages. In: COMPSTAT 2004 Proceedings (J. Antoch, ed.), Springer Verlag 2004, pp. 1345-1352.   DOI:10.1007/978-3-7908-2656-2
  11. R. A. Maronna, D. R. Martin and V. J. Yohai: Robust Statistics: Theory and Methods. John Wiley and Sons Ltd, Chichester 2006.   DOI:10.1002/0470010940
  12. G. H. Moore and W. A. Wallis: Time series significance tests based on signs of differences. J. Amer. Statist. Assoc. 38 (1943), 222, 153-164.   DOI:10.1080/01621459.1943.10501791
  13. M. Papageorgiou, A. Kotsialos and A. Poulimenos: Long-term sales forecasting using holt-winters and neural network methods. J. Forecasting 25 (2005), 5, 353-368.   DOI:10.1002/for.943
  14. R. Romera and T. Cipra: On practical implementation of robust Kalman filtering. Commun. Statist. - Simulation and Computation 24 (1995), 2, 461-488.   DOI:10.1080/03610919508813252
  15. F. Wilcoxon: Individual comparisons by ranking methods. Biometrics Bull. 1 (1945), 6, 80-83.   DOI:10.2307/3001968
  16. J. Wolfowitz: Asymptotic distribution of runs up and down. Ann. Math. Statist. 15 (1944), 2, 163-172.   DOI:10.1214/aoms/1177731281