Skip to main content
Log in

Evaluation of forecasts in AR models with outliers

  • Theoretical Papers
  • Published:
Operations-Research-Spektrum Aims and scope Submit manuscript

Abstract

The ex-post evaluation of forecasts by the realized forecast errors is an important tool in choosing an adequate model to represent the analyzed time series data and also in comparing competing forecast methods. Measures like the mean squared error MSE and the mean absolute error MAE are frequently used for this purpose. Especially when analyzing data with outliers MSE and MAE may produce misleading results. This paper presents robustified versions MRSE and MRAE of MSE and MAE, respectively. They are much better suited to identify those forecasts which are based on the parameters of the underlying model. This feature is illustrated by a simulation study.

Zusammenfassung

Die Ex-post-Beurteilung von Vorhersagen anhand der realisierten Prognosefehler ist ein wichtiges Hilfsmittel bei der vergleichenden Bewertung konkurrierender Prognoseverfahren oder auch bei der Modellauswahl im Rahmen der Analyse einer Zeitreihe. Hierzu werden häufig der mittlere quadratische Fehler MSE, der mittlere absolute Fehler MAE und ähnliche Maße herangezogen. Gerade bei ausreißerbehafteten Daten zeigt es sich, daß die Verwendung von MSE und MAE zu irreführenden Resultaten führen kann. In dieser Arbeit werden robustifizierte Versionen des MSE und MAE -MRSE bzw. MRAE—vorgestellt, die in einer Simulationsstudie insbesondere bei Daten mit Ausreißern zu einer deutlich sichereren Identifikation der Vorhersage anhand der zugrundeliegenden Modellparameter führten.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. Box GEP, Jenkins GM (1976) Time Series Analysis: Forecasting and Control. Revised Edition. Prentice Hall, Englewood Cliffs, New Jersey

    Google Scholar 

  2. Elsebach R (1993) Prognosen bei AR-Prozessen mit Ausreißern. Dissertation. Universität der Bundeswehr Hamburg

  3. Huang D (1990) Levinson-Type Recursive Algorithms for Least-Squares Autoregression. Journal of Time Series Analysis 11(4): 295–315

    Article  Google Scholar 

  4. Huber PJ (1964) Robust Estimation of a Location Parameter. Annals of Mathematical Statistics 35: 73–101

    Article  Google Scholar 

  5. Makridakis S et al. (1982) The Accuracy of Extrapolation (Time Series) Methods: Results of a Forecasting Competition. Journal of Forecasting 1: 111–153

    Article  Google Scholar 

  6. Martin RD (1979) Robust Estimation for Time Series Autoregression. In: Launer, Wilkinson (eds) Robustness in Statistics. Academic Press, New York, pp 147–176

    Chapter  Google Scholar 

  7. Martin RD (1981) Robust Methods for Time Series. In: Findley DF (ed). Applied Time Series Analysis II. Academic Press, New York, pp 683–759

    Chapter  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Additional information

This article was processed using Springer-Verlag TEX OR Spektrum macro package 1.0 and the AMS fonts, developed by the American Mathematical Society.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Elsebach, R. Evaluation of forecasts in AR models with outliers. OR Spektrum 16, 41–45 (1994). https://doi.org/10.1007/BF01719702

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF01719702

Schlüsselwörter

Key words

Navigation