Robust Forecasting of Non-Stationary Time Series

C. Croux, R. Fried, I. Gijbels, K. Mahieu

Research output: Working paperDiscussion paperOther research output

247 Downloads (Pure)

Abstract

This paper proposes a robust forecasting method for non-stationary time series. The time series is modelled using non-parametric heteroscedastic regression, and fitted by a localized MM-estimator, combining high robustness and large efficiency. The proposed method is shown to produce reliable forecasts in the presence of outliers, non-linearity, and heteroscedasticity. In the absence of outliers, the forecasts are only slightly less precise than those based on a localized Least Squares estimator. An additional advantage of the MM-estimator is that it provides a robust estimate of the local variability of the time series.
Original languageEnglish
Place of PublicationTilburg
PublisherEconometrics
Number of pages17
Volume2010-105
Publication statusPublished - 2010

Publication series

NameCentER Discussion Paper
Volume2010-105

Fingerprint

time series
outlier
forecasting method
nonlinearity
forecast
method

Keywords

  • Heteroscedasticity
  • Non-parametric regression
  • Prediction
  • Outliers
  • Robustness

Cite this

Croux, C., Fried, R., Gijbels, I., & Mahieu, K. (2010). Robust Forecasting of Non-Stationary Time Series. (CentER Discussion Paper; Vol. 2010-105). Tilburg: Econometrics.
Croux, C. ; Fried, R. ; Gijbels, I. ; Mahieu, K. / Robust Forecasting of Non-Stationary Time Series. Tilburg : Econometrics, 2010. (CentER Discussion Paper).
@techreport{94542b5e43194f5abc35258e532d573c,
title = "Robust Forecasting of Non-Stationary Time Series",
abstract = "This paper proposes a robust forecasting method for non-stationary time series. The time series is modelled using non-parametric heteroscedastic regression, and fitted by a localized MM-estimator, combining high robustness and large efficiency. The proposed method is shown to produce reliable forecasts in the presence of outliers, non-linearity, and heteroscedasticity. In the absence of outliers, the forecasts are only slightly less precise than those based on a localized Least Squares estimator. An additional advantage of the MM-estimator is that it provides a robust estimate of the local variability of the time series.",
keywords = "Heteroscedasticity, Non-parametric regression, Prediction, Outliers, Robustness",
author = "C. Croux and R. Fried and I. Gijbels and K. Mahieu",
note = "Pagination: 17",
year = "2010",
language = "English",
volume = "2010-105",
series = "CentER Discussion Paper",
publisher = "Econometrics",
type = "WorkingPaper",
institution = "Econometrics",

}

Croux, C, Fried, R, Gijbels, I & Mahieu, K 2010 'Robust Forecasting of Non-Stationary Time Series' CentER Discussion Paper, vol. 2010-105, Econometrics, Tilburg.

Robust Forecasting of Non-Stationary Time Series. / Croux, C.; Fried, R.; Gijbels, I.; Mahieu, K.

Tilburg : Econometrics, 2010. (CentER Discussion Paper; Vol. 2010-105).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Robust Forecasting of Non-Stationary Time Series

AU - Croux, C.

AU - Fried, R.

AU - Gijbels, I.

AU - Mahieu, K.

N1 - Pagination: 17

PY - 2010

Y1 - 2010

N2 - This paper proposes a robust forecasting method for non-stationary time series. The time series is modelled using non-parametric heteroscedastic regression, and fitted by a localized MM-estimator, combining high robustness and large efficiency. The proposed method is shown to produce reliable forecasts in the presence of outliers, non-linearity, and heteroscedasticity. In the absence of outliers, the forecasts are only slightly less precise than those based on a localized Least Squares estimator. An additional advantage of the MM-estimator is that it provides a robust estimate of the local variability of the time series.

AB - This paper proposes a robust forecasting method for non-stationary time series. The time series is modelled using non-parametric heteroscedastic regression, and fitted by a localized MM-estimator, combining high robustness and large efficiency. The proposed method is shown to produce reliable forecasts in the presence of outliers, non-linearity, and heteroscedasticity. In the absence of outliers, the forecasts are only slightly less precise than those based on a localized Least Squares estimator. An additional advantage of the MM-estimator is that it provides a robust estimate of the local variability of the time series.

KW - Heteroscedasticity

KW - Non-parametric regression

KW - Prediction

KW - Outliers

KW - Robustness

M3 - Discussion paper

VL - 2010-105

T3 - CentER Discussion Paper

BT - Robust Forecasting of Non-Stationary Time Series

PB - Econometrics

CY - Tilburg

ER -

Croux C, Fried R, Gijbels I, Mahieu K. Robust Forecasting of Non-Stationary Time Series. Tilburg: Econometrics. 2010. (CentER Discussion Paper).