Skip to main content
Log in

Mathematical and statistical details on the simulation of MARKOFF-Type stochastic processes on an electronic computer

  • Published:
Computing Aims and scope Submit manuscript

Summary

This paper deals with the generation of stationaryp th order linear autoregressive series (calledM p -series) on an electronic computer. The interrelations between the coefficients of autocorrelation are discussed and a device and a flow diagram are given for the generation ofM p -series which possess the autocorrelation coefficients ϱ1, ϱ2,...ϱ p . The conclusion is that there is anM p -series for a given set of values only if there is anM q -series for any subset ϱ1,...ϱ q withq=1,2,...q−1 and that, conversely, if there is anM p -series for given ϱ12,...ϱ p , there is also anM q -series with ϱ1,...ϱ q for 1≤q<p.

The series withp=1, 2, 3 are treated fully and numerical examples forp=1 andp=2 are given in Fig. 4.

Zusammenfassung

In diesem Aufsatz wird besprochen wie mit Hilfe einer elektronischen Rechenmaschine stationäre lineare autoregressive Reihen der Ordnungp (M p -Reihen genannt) konstruiert werden können. Nachdem die Beziehungen zwischen den Autokorrelationskoeffizienten abgeleitet worden sind, wird ein Schema und ein Flußdiagramm zur Erzeugung vonM p -Reihen gegeben, die vorgegebene Autokorrelationskoeffizienten ϱ1,...ϱ p besitzen. Das Ergebnis lautet: EineM p -Reihe für eine Gruppe von gegebenen Werten ϱ1,...ϱ p ist nur möglich, wenn eineM q -Reihe für jede Untergruppe ϱ1,...ϱ q mitq=1, 2, ...p−1 möglich ist. Wenn einmal eineM p -Reihe mit gegebenen ϱ1,...ϱ p existiert, dann existiert ebenfalls jedeM q -Reihe mit ϱ1,...ϱ q , wobei 1≤q<p ist.

Die Fällep=1, 2, 3 werden ausführlich behandelt, während die Abb. 4 numerische Beispiele fürp=1 undp=2 zeigt.

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. Kendall, M. G. andA. Stuart: The Advanced Theory of Statistics, Vol. III. 1st Ed., chapter 46. London: Charles Griffin and Company Limited. 1966.

    Google Scholar 

  2. Wise, J.: Stationarity Conditions for Stochastic Processes of the Autoregressive and Moving-Average Type. Biometrika43, 215 (1956).

    Google Scholar 

  3. Deming, W. E.: Statistical Adjustment of Data. 20th Ed. New York-London: J. Wiley and Sons. 1946.

    Google Scholar 

  4. Yule, G. U.: On a Method of Investigation Periodicities in Disturbed Series. Phil. Transactions226, 267 (1927).

    Google Scholar 

  5. Walker, G. T.: On Periodicities in Series of Related Terms. Proc. Roy. Soc.A 131, 518 (1931).

    Google Scholar 

  6. Brooks, C. E. P. andN. Carruthers: Handbook of Statistical Methods in Meteorology. 1st Ed., chapter 16. London: Her Majesty's Stationery Office. 1953.

    Google Scholar 

  7. Hastings, C.: Approximations for Digital Computers. 1st Ed., sheets 63 and 68. Princeton: Univ. Press. 1955.

    Google Scholar 

  8. Czepa, O.: Zur Realisierung stationärer Zufallsfolgen. Beitr. zur Geoph.75, 425 (1966).

    Google Scholar 

  9. Forsythe, G. E.: Monte Carlo Method, Chapter 12: Generation and Testing of Random Digits at the National Bureau of Standards, Los Angeles. National Bureau of Standards Applied Mathematics Series. 12 U.S. Dep. of Commerce. 1951.

Download references

Author information

Authors and Affiliations

Authors

Additional information

With 4 Figures

Rights and permissions

Reprints and permissions

About this article

Cite this article

Levert, C., van Galen, J. Mathematical and statistical details on the simulation of MARKOFF-Type stochastic processes on an electronic computer. Computing 3, 65–75 (1968). https://doi.org/10.1007/BF02238105

Download citation

  • Received:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF02238105

Keywords

Navigation