Previous |  Up |  Next

Article

Title: Exponential smoothing based on L-estimation (English)
Author: Bejda, Přemysl
Author: Cipra, Tomáš
Language: English
Journal: Kybernetika
ISSN: 0023-5954 (print)
ISSN: 1805-949X (online)
Volume: 51
Issue: 6
Year: 2015
Pages: 973-993
Summary lang: English
.
Category: math
.
Summary: 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. (English)
Keyword: change point
Keyword: exponential smoothing
Keyword: quantiles
Keyword: robust methods
Keyword: sign test
MSC: 62M10
MSC: 62M20
MSC: 62N02
idZBL: Zbl 06537791
idMR: MR3453681
DOI: 10.14736/kyb-2015-6-0973
.
Date available: 2016-01-21T18:20:33Z
Last updated: 2018-01-10
Stable URL: http://hdl.handle.net/10338.dmlcz/144820
.
Reference: [1] Brown, R. G.: Smoothing, Forecasting and Prediction of Descrete Time Series..Prentice-Hall, Englewood Cliffs, NJ 1962.
Reference: [2] Cipra, T.: Robust exponential smoothing..J. Forecasting 11 (1992), 1, 57-69. 10.1002/for.3980110106
Reference: [3] Cipra, T., Romera, R.: Recursive time series methods in $L_1$-norm..In: $L_1$-Statistical Analysis and Related Methods (Y. Dodge ed.), pp. 233-243, 1992. MR 1214835, 10.2307/2289888
Reference: [4] Gelper, S., Fried, R., Croux, Ch.: Robust forecasting with exponential and Holt-Winters smoothing..J. Forecasting 29 (2010), 3, 285-300. Zbl 1203.62164, MR 2752114, 10.2139/ssrn.1089403
Reference: [5] Hanzák, T., Cipra, T.: Exponential smoothing for time series with outliers..Kybernetika 47 (2011), 2, 165-178. Zbl 1220.62114, MR 2828571
Reference: [6] Holt, Ch. C.: Forecasting seasonals and trends by exponentially weighted moving averages..Int. J. Forecasting 20 (2004), 5-10. 10.1016/j.ijforecast.2003.09.015
Reference: [7] Hyndman, R. J., Koehler, A. B., Ord, J. K., Snyder, R. D.: Forecasting with Exponential Smoothing..Springer, Berlin - Heidelberg 2008. Zbl 1211.62165, 10.1007/978-3-540-71918-2
Reference: [8] Jurečková, J., Picek, J.: Robust Statistical Methods with R..Chapman and Hall/CRC, 2005. Zbl 1097.62020, 10.1201/9781420035131
Reference: [9] Koenker, R., Bassett, G.: Regression quantiles..Econometrica 46 (1978), 1, 33-50. Zbl 0482.62023, MR 0474644, 10.2307/1913643
Reference: [10] Koubková, A.: Forecasting seasonals and trends by exponentially weighted moving averages..In: COMPSTAT 2004 Proceedings (J. Antoch, ed.), Springer Verlag 2004, pp. 1345-1352. 10.1007/978-3-7908-2656-2
Reference: [11] Maronna, R. A., Martin, D. R., Yohai, V. J.: Robust Statistics: Theory and Methods..John Wiley and Sons Ltd, Chichester 2006. Zbl 1094.62040, MR 2238141, 10.1002/0470010940
Reference: [12] Moore, G. H., Wallis, W. A.: Time series significance tests based on signs of differences..J. Amer. Statist. Assoc. 38 (1943), 222, 153-164. Zbl 0063.04084, MR 0008323, 10.1080/01621459.1943.10501791
Reference: [13] Papageorgiou, M., Kotsialos, A., Poulimenos, A.: Long-term sales forecasting using holt-winters and neural network methods..J. Forecasting 25 (2005), 5, 353-368. MR 2190371, 10.1002/for.943
Reference: [14] Romera, R., Cipra, T.: On practical implementation of robust Kalman filtering..Commun. Statist. - Simulation and Computation 24 (1995), 2, 461-488. Zbl 0850.62688, MR 1333047, 10.1080/03610919508813252
Reference: [15] Wilcoxon, F.: Individual comparisons by ranking methods..Biometrics Bull. 1 (1945), 6, 80-83. 10.2307/3001968
Reference: [16] Wolfowitz, J.: Asymptotic distribution of runs up and down..Ann. Math. Statist. 15 (1944), 2, 163-172. Zbl 0063.08310, MR 0010362, 10.1214/aoms/1177731281
.

Files

Files Size Format View
Kybernetika_51-2015-6_5.pdf 615.8Kb application/pdf View/Open
Back to standard record
Partner of
EuDML logo