Nonparametric Estimation for Processes

  • M. M. Rao
Part of the Mathematics and Its Applications book series (MAIA, volume 508)


This final chapter is concerned with questions of (asymptotic) unbiasedness, and consistent nonparametric estimation of some functions, such as bispectral densities of a class of second order processes. After some necessary preliminaries, we discuss spectral properties of separable, especially harmonizable, processes of second order in Section 1. Then we consider an asymptotically unbiased estimator, and a related function, of the spectral distribution of the process in the next section. These are usually not consistent estimators when the process is not stationary. So we need to use another procedure, the so-called resampling method, wherein the covariance between samples falls off at a reasonable rate (made precise later). This is described in Section 3 and using such a procedure it is possible to obtain a consistent estimator of the bispectral density of strongly harmonizable processes, and this is given for such a class. Then Section 4 contains a slightly more general second order family for which some related results are discussed. In Section 5 a limit distribution of the (nonparametric) estimator defined above for a strongly harmonizable class is presented. Thus the conditions imposed are progressively more stringent, but then we get more refined results. Several new avenues and possible improvements of the results are pointed out, with related exercises in the last section, usually with sketches of proofs, as complements to the preceding work.


Covariance Function Spectral Function Limit Distribution Unbiased Estimator Nonparametric Estimation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Bibliographical notes

  1. [1]
    Soedjak, H. Asymptotic Properties of Bispectral Density Estimators of Harmonizable Processes,Ph. D. thesis, UCR Library, Riverside, CA, 1996.Google Scholar
  2. [1]
    Ibragimov, I. A., and Linnik, Ju. V. Independent and Stationary Sequences of Random Variables,Noordhoff Publishers, The Netherlands, 1971.Google Scholar
  3. [1]
    Kampé de Fériet, J., and Frenkiel, F. N. Estimation de la corrélation d’une fonction aléatoire non stationnaire,“ C. R. Acad. Sci., Paris bf 249 (1959), 348–351.Google Scholar
  4. [4]
    Parzen, E. Spectral analysis of asymptotically stationary time series,“ Bull. Inst. Internat. Statist.,29(2) (1962), 87–103.MathSciNetGoogle Scholar
  5. [2]
    Rozanov, Yu. A. Spectral analysis of abstract functions,“ Theor. Prob., Appl. 4 (1959), 271–287.Google Scholar
  6. [1]
    Hanin, L. G., and Schreiber, B. M. “Discrete spectrum of nonstationary stochastic processes on LCA groups,” J. Theor. Prob., 11 (1998), 1111–1133.MathSciNetzbMATHCrossRefGoogle Scholar
  7. [2]
    Swift, R. J. Some aspects of harmonizable processes and fields,“ In Real and Stochastic Analysis: Recent Advances,CRC Press, Boca Raton, FL, (1997), 303–365.Google Scholar
  8. [1]
    Nagabhushanam, K. “The primary process of a smoothing relation,” Ark. Mat. 1 (1951) 421–488.Google Scholar
  9. [1]
    Palais, R. S. Natural operations on differential forms,“ Trans. Am. Math. Soc.,92 (1959), 125–141.MathSciNetzbMATHCrossRefGoogle Scholar
  10. [1]
    Hirai, T., and Shimomura, H. Relations between unitary representations of Diffeomorphism groups and those of the infinite symmetric group or of related permutation groups,“ J. Math. Kyoyo Univ.,37 (1997), 261–316.Google Scholar
  11. [1]
    Grenander, U., and Miller, M. I. “Representations of knowledge in complex systems,” J. R. Statist. Soc. Ser B, 56 (1994), 549–603.MathSciNetzbMATHGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2000

Authors and Affiliations

  • M. M. Rao
    • 1
  1. 1.University of CaliforniaRiversideUSA

Personalised recommendations