Computational aspects of robust Holt-Winters smoothing based on $M$-estimation

Croux, Christophe, Gelper, Sarah, and Fried, Roland. "Computational aspects of robust Holt-Winters smoothing based on $M$-estimation." Applications of Mathematics 53.3 (2008): 163-176. <http://eudml.org/doc/37776>.

@article{Croux2008,
abstract = {To obtain a robust version of exponential and Holt-Winters smoothing the idea of $M$-estimation can be used. The difficulty is the formulation of an easy-to-use recursive formula for its computation. A first attempt was made by Cipra (Robust exponential smoothing, J. Forecast. 11 (1992), 57–69). The recursive formulation presented there, however, is unstable. In this paper, a new recursive computing scheme is proposed. A simulation study illustrates that the new recursions result in smaller forecast errors on average. The forecast performance is further improved upon by using auxiliary robust starting values and robust scale estimates.},
author = {Croux, Christophe, Gelper, Sarah, Fried, Roland},
journal = {Applications of Mathematics},
keywords = {Holt-Winters smoothing; robust methods; time series; Holt-Winters smoothing; robust methods; time series},
language = {eng},
number = {3},
pages = {163-176},
publisher = {Institute of Mathematics, Academy of Sciences of the Czech Republic},
title = {Computational aspects of robust Holt-Winters smoothing based on $M$-estimation},
url = {http://eudml.org/doc/37776},
volume = {53},
year = {2008},
}

TY - JOUR
AU - Croux, Christophe
AU - Gelper, Sarah
AU - Fried, Roland
TI - Computational aspects of robust Holt-Winters smoothing based on $M$-estimation
JO - Applications of Mathematics
PY - 2008
PB - Institute of Mathematics, Academy of Sciences of the Czech Republic
VL - 53
IS - 3
SP - 163
EP - 176
AB - To obtain a robust version of exponential and Holt-Winters smoothing the idea of $M$-estimation can be used. The difficulty is the formulation of an easy-to-use recursive formula for its computation. A first attempt was made by Cipra (Robust exponential smoothing, J. Forecast. 11 (1992), 57–69). The recursive formulation presented there, however, is unstable. In this paper, a new recursive computing scheme is proposed. A simulation study illustrates that the new recursions result in smaller forecast errors on average. The forecast performance is further improved upon by using auxiliary robust starting values and robust scale estimates.
LA - eng
KW - Holt-Winters smoothing; robust methods; time series; Holt-Winters smoothing; robust methods; time series
UR - http://eudml.org/doc/37776
ER -