Robust Forecasting with Exponential and Holt-Winters Smoothing

Robust versions of the exponential and Holt-Winters smoothing method for forecasting are presented. They are suitable for forecasting univariate time series in the presence of outliers. The robust exponential and Holt-Winters smoothing methods are presented as recursive updating schemes that apply the standard technique to pre-cleaned data. Both the update equation and the selection of the smoothing parameters are robustified. A simulation study compares the robust and classical forecasts. The presented method is found to have good forecast performance for time series with and without outliers, as well as for fat-tailed time series and under model misspecification. The method is illustrated using real data incorporating trend and seasonal effects. Copyright © 2009 John Wiley & Sons, Ltd.

[1]  Rosario Romera,et al.  On practical implementation of robust Kalman filtering , 1995 .

[2]  Nancy J. Kirkendall.  Monitoring for outliers and level shifts in kalman filter implementations of exponential smoothing , 1992 .

[3]  Rob J Hyndman,et al.  Prediction intervals for exponential smoothing using two new classes of state space models 30 January 2003 , 2003 .

[4]  Peter R. Winters,et al.  Forecasting Sales by Exponentially Weighted Moving Averages , 1960 .

[5]  B. Ripley,et al.  Robust Statistics , 2018, Encyclopedia of Mathematical Geosciences.

[6]  Steven C. Wheelwright,et al.  Forecasting: Methods and Applications, 3rd Ed , 1997 .

[7]  Christophe Croux,et al.  Robust Forecasting with Exponential and Holt-Winters Smoothing , 2007 .

[8]  V. Barnett,et al.  Applied Linear Statistical Models , 1975 .

[9]  B. Bowerman,et al.  Forecasting, time series, and regression : an applied approach , 2005 .

[10]  Tomáš Cipra.  Robust exponential smoothing , 1992 .

[11]  A. Siegel.  Robust regression using repeated medians , 1982 .

[12]  R. Martin,et al.  Robust bayesian estimation for the linear model and robustifying the Kalman filter , 1977 .

[13]  James W. Taylor,et al.  Forecasting daily supermarket sales using exponentially weighted quantile regression , 2007, Eur. J. Oper. Res..

[14]  W. D. Ray.  Applied Linear Statistical Models (3rd Edition) , 1991 .

[15]  Ursula Gather,et al.  Robust signal extraction for on-line monitoring data , 2004 .

[16]  Markos Papageorgiou,et al.  HOLT-WINTERS AND NEURAL-NETWORK METHODS FOR MEDIUM-TERM SALES FORECASTING , 2005 .

[17]  Michael H. Kutner.  Applied Linear Statistical Models , 1974 .

[18]  D. Thomson,et al.  Robust-resistant spectrum estimation , 1982, Proceedings of the IEEE.

[19]  C. Holt.  Author's retrospective on ‘Forecasting seasonals and trends by exponentially weighted moving averages’ , 2004 .

[20]  Everette S. Gardner,et al.  Exponential smoothing: The state of the art , 1985 .

[21]  Steven C. Wheelwright,et al.  Forecasting methods and applications. , 1979 .

[22]  T. Cipra,et al.  Holt-Winters method with missing observations , 1995 .

[23]  James W. Taylor.  Smooth transition exponential smoothing , 2004 .

[24]  Rosario Romera,et al.  Kalman filter with outliers and missing observations , 1997 .

[25]  Jr. Everette S. Gardner,et al.  Rule-Based Forecasting Vs. Damped-Trend Exponential Smoothing , 1999 .

[26]  V. Yohai,et al.  Robust Statistics: Theory and Methods , 2006 .

[27]  Markos Papageorgiou,et al.  Long‐term sales forecasting using holt–winters and neural network methods , 2005 .

[28]  E. S. Gardner.  EXPONENTIAL SMOOTHING: THE STATE OF THE ART, PART II , 2006 .

[29]  V. Yohai,et al.  High Breakdown-Point Estimates of Regression by Means of the Minimization of an Efficient Scale , 1988 .

[30]  Ronald K. Pearson,et al.  Mining imperfect data - dealing with contamination and incomplete records , 2005 .

[31]  Richard A. Davis,et al.  Introduction to time series and forecasting , 1998 .

[32]  Roland Fried,et al.  Robust filtering of time series with trends , 2004 .

[33]  Steven C. Wheelwright,et al.  Forecasting: Methods and Applications, 3rd Edition , 1998 .