Statistical Inference for Stochastic Processes

, Volume 8, Issue 2, pp 151–184 | Cite as

On the Non-parametric Prediction of Conditionally Stationary Sequences

Abstract

We prove the strong consistency of estimators of the conditional distribution function and conditional expectation of a future observation of a discrete time stochastic process given a fixed number of past observations. The results apply to conditionally stationary processes (a class of processes including Markov and stationary processes) satisfying a strong mixing condition, and they extend and bring together the work of several authors in the area of non-parametric estimation. One of our goals is to provide further justification for the growing practical application of non-parametric estimators in non-stationary time series and in other `non-i.i.d.' settings. Some arguments as to why such estimators should work very generally in practice, often in a nearly `optimal' way, are given. Two numerical illustrations are included, one with simulated data and the other with oceanographic data.

Keywords

non-parametric prediction conditional distribution function conditional expectation time series data analysis 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Ango Nze, P., Doukhan, P 2002

    Weak dependence: models and applications

    Dehling, H.Mikösh, T.Sorensen, M. eds. Empirical Process Techniques for Dependent Data.BirkhäuserBasel117136
    Google Scholar
  2. Ash, R., Dol´eans-Dade, C 2000Probability and Measure Theory, 2nd editionAcademic PressNew YorkGoogle Scholar
  3. Bosq, D 1997Nonparametric Statistics for Stochastic Processes, 2nd edition, Lecture Notes in Statistics 110Springer-VerlagNew YorkGoogle Scholar
  4. Brockwell, P., Davis, R 1987Time Series: Theory and MethodsSpringer-VerlagNew YorkGoogle Scholar
  5. Caires, S. and Sterl, A.: Validation of ocean wind and wave data using triple collocation, J. Geophys. Res. 108(C3) (2003), 43.1-43.16, doi:10.1029/2002JC001491.Google Scholar
  6. Carbon, M 1983Inégalité de Bernstein pour les processus fortement m´elangeants, non n´ecessairement stationairesApplications, C. R. Acad. Sci. Paris, I, t.297303306MATHMathSciNetGoogle Scholar
  7. Collomb, G 1984de convergence presque compl‘ete du pr´edicteur ‘anoyau, ZWahrscheinlichkeitstheorie verw. Gebiete66441460CrossRefMATHMathSciNetGoogle Scholar
  8. Collomb, G 1985Nonparametric regression: an up-to-date bibliographyStatistics16309324MATHMathSciNetGoogle Scholar
  9. Devroye, L 1981On the almost everywhere convergence of nonparametric regression function estimatesAnn. Stat.913101319MATHMathSciNetGoogle Scholar
  10. Grenander, U, Szegö, G 1958Toeplitz Forms and Their ApplicationsUniversity of California PressCaliforniaGoogle Scholar
  11. Gy örfi, L, Härdle, W, Sarda, P, Vieu, P 1989Nonparametric Curve Estimation from Time Series, Lecture Notes in Statistics 60Springer-VerlagNew YorkGoogle Scholar
  12. Härdle, W 1989Applied Nonparametric RegressionCambridge University PressCambridgeGoogle Scholar
  13. Ibragimov, I., Rozanov, Y 1978Gaussian Random ProcessesSpringer-VerlagNew YorkGoogle Scholar
  14. Knopp, K 1928Theory and Application of Infinite SeriesBlackie and SonLondon GlasgowGoogle Scholar
  15. Nadaraya, E 1964On estimating regressionTheor. Probab. Appl.9141142CrossRefGoogle Scholar
  16. Roussas, G 1969Nonparametric estimation of the transition distribution function of a Markov process, AnnMath. Stat.4013861400MATHMathSciNetGoogle Scholar
  17. Roussas, G 1990Nonparametric regression estimation under mixing conditions StochastProcess. Appl.36107116CrossRefMATHMathSciNetGoogle Scholar
  18. Roussas, G. 1991

    Estimation of transition distribution function and its quantiles in Markov processes: strong consistency and asymptotic normality

    Roussas, G. eds. Nonparametric Funcional Estimation and Related Topics Vol 335.KluwerDordrecht443462
    Google Scholar
  19. Roussas, G. (ed): Nonparametric Funcional Estimation and Related Topics, Vol. 335, Kluwer, Dordrecht, 1991b.Google Scholar
  20. Stute, W 1986On almost sure convergence of conditional empirical distribution functionsAnn. Probab.14891901MATHMathSciNetGoogle Scholar
  21. Tucker, H, Graduate, A 1967Course in ProbabilityAcademic PressNew York LondonGoogle Scholar
  22. Watson, G. 1964Smooth regression analysisSankhya26359372MATHGoogle Scholar
  23. Yakowitz, S 1979Nonparametric estimation of Markov transition functions AnnStat.7671679MATHMathSciNetGoogle Scholar
  24. Yakowitz, S 1985Nonparametric density estimation, prediction and regression for Markov sequencesAm. Statist. Soc.80215221MATHMathSciNetGoogle Scholar

Copyright information

© Springer 2005

Authors and Affiliations

  1. 1.KNMIRoyal Netherlands Meteorological InstituteDe BiltThe Netherlands
  2. 2.CWICentrum voor Wiskunde en InformaticaAmsterdamThe Netherlands

Personalised recommendations