Skip to main content
Log in

Several forecast models applied to a specific economic time series

Verschiedene Vorhersagemodelle angewandt auf eine spezifische ökonomische Zeitreihe

Plusieurs modèles prédictifs appliqués à une série temporelle spécifique

Разные модели прогноза, примененные в специфическом Экономическом временном ряде

  • Published:
Statistische Hefte Aims and scope Submit manuscript

Zusammenfassung

Die Modell-Identifikation wird für verschiedene Vorhersagemethoden angewandt auf eine spezifische ökonomische Zeitreihe.

Die Methoden sind: 1) Die Methode der exponentiellen Ausgleichung von Brown; 2) Die Wintersche Methode der exponentiellen Ausgleichung für multiplikative periodische Zeitreihen; 3) Die "characteristic modes" Methode.

Bei allen Vorhersagemethoden ist der "Cusum-Test" angewendet worden, um zu entscheiden, ob die Modellparameter aufs neue berechnet werden müssen.

Zwei verschiedene Reihen von Vorhersagen sind zusammengestellt worden zu einer kombinierten Reihe. Es gibt Gründe anzunehmen, daß eine Kombination von den Vorhersagen von 1) und 3) im allgemeinen eine niedrigere "M.A.D." gibt als jede der separaten Vorhersagen.

Summary

Model identification for several forecast procedures is applied to a specific economic time series. The procedures are 1) Brown’s method of exponential smoothing; 2) Winters’ method of exponential smoothing for multiplicative periodic time series; 3) the method of characteristic modes. For all the forecast models we have used a control procedure (cumulative sum test) to decide whether the model parameters have to be recalculated. Two separate sets of forecasts have been combined to form a composite set of forecasts. There is an indication that a combination of forecasts from the procedures 1) and 3) yield in general a lower M.A.D. than either of the original forecasts.

Résumé

Nous avons appliqué trois méthodes prédictives à une série temporelle spécifique.

Pour tous les modèles prédictifs nous avons usé le test "cusum" pour décider si il est nécessaire de récalculer les estimateurs du modèle. Deux collections de prédictions différentes sont combinées à une seule collection mixte. Il y a l’indication qu’une combination de prédictions donne en géneral un résultat avec une "déviation moyenne absolue" moins grande que chaqu’une des prédictions originelles.

Резюме

Йдентификация модели для разных метод прогноза применяется в специфическом Экономическом временном ряде.

Это следуюшие методы: I) Метод Экспонентного выравнивания Брауна; 2) Метод Экспонентного выравнивания для мультипликативных периодических временных рядов (Винтере); 3) Метод "характерных моде".

Во всех методах прогноза применялся "Кусум-Тест", чтобы рещить, нужно ли вновь вычислить параметры моделей. Два разных ряда прогноз сопоставились в один комбинированный ряд. Ймеется причины предполагать, что комбинация прогноз из метод I) и 3) дает в обшем низщий "М.А.Д." чем каждый отдельный прогноз.

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.

7. References

  1. Anderson, T.W. (1971), The statistical analysis of time series; John Wiley & Sons, New York.

    MATH  Google Scholar 

  2. Bates, J.M. and C.W.J. Granger, "The combination of forecasts". Operational Research Quarterly,20, 4, pp. 451–468.

  3. Brown, R.G. (1963), Smoothing, forecasting and prediction of discrete time series; Prentice Hall, Englewood Cliffs.

    Google Scholar 

  4. Christ, C.R. (1966), Econometric models and methods; John Wiley & Sons, New York.

    Google Scholar 

  5. Dobben de Bruyn, C.v. (1968), Theory and practice of cumulated sum tests; Griffin’s statistical monographs and courses, nr. 24.

  6. Durbin, J. (1960), "The fitting of time-series models"; Revue de l’institute international de statistique,30, 3.

    Google Scholar 

  7. Farmer, E.D. (1964), "A method of prediction for nonstationary processes and its application to the problem of load estimation"; Proceedings of the 2nd Congress of the Itern. Fed. Automatic Control, Basle, 1963, Butterworth, London.

    Google Scholar 

  8. Freeman, H. (1964), Discrete time systems; John Wiley & Sons, New York.

    Google Scholar 

  9. Goldberger, A.S. (1964), Econometric theory; John Wiley & Sons, New York.

    MATH  Google Scholar 

  10. Granger, C.W.J. (1964), Spectral analysis of economic time series; Princeton University Press.

  11. Graybill, F.A. (1961), An introduction to linear statistical models; McGraw-Hill Book Company, New York.

    MATH  Google Scholar 

  12. Grenander, U. (1950), "Stochastic processes and statistical inference"; Arkiv för Matematik,1, 17.

    Article  MathSciNet  Google Scholar 

  13. Hartley, H.O. (1949), "Test for significance in harmonic analysis"; Biometrika,36.

  14. Heuts, R.M.J. (1973), "A new test statistic for searching hidden periodicities in time series and the derivation and numerical calculation of its power function"; Statistische Hefte,14, 1.

    Article  MathSciNet  Google Scholar 

  15. Heuts, R.M.J. and P.J. Rens (1971), Model building and time series forecasting, with some applications; R.C.-notitie nr. 10, Tilburg School of Economics.

  16. Heuts, R.M.J. and P.J. Rens (1973), Several forecast models applied to a specific economic time series with application of control theory. R.C.-notitie nr. 15, Tilburg School of Economics.

  17. Johnston, J. (1963), Econometric methods; McGraw-Hill Book Company, New York.

    Google Scholar 

  18. Karhunen, K. (1952), "Über ein Extrapolationproblem in dem Hilbertschen Raum"; Second Scandanavian Mathem. Congress.

  19. Koerts, J. and A.P.J. Abrahamse (1969), On the theory and application of the general linear model; Rotterdam University Press.

  20. Malinvaud, E. (1970), Statistical methods of econometrics; North Holland Publ. Company, Amsterdam.

    MATH  Google Scholar 

  21. Meyer, R.F. (1963), "An adaptive method for routine short term forecasting"; presented at IFORS, Oslo, July.

  22. Wilkinson, J.H. (1965), The algebraic eigenvalue problem; Clarendon Press, Oxford.

    MATH  Google Scholar 

  23. Winkel, E.G.F. van and D.J. Fraser (1970), Tijdreeksvoorspellingen en hun bewaking; N.V. Samson, Alphen a/d Rijn.

  24. Winters, R. (1960), "Forecasting sales by exponentially weighted moving averages"; Management Science,6, 3.

    Article  MathSciNet  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

About this article

Cite this article

Heuts, R.M.J., Rens, P.J. Several forecast models applied to a specific economic time series. Statistische Hefte 16, 157–187 (1975). https://doi.org/10.1007/BF02922998

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF02922998

Keywords

Navigation