Summary
We study a nonclassical form of empirical df H nwhich is of U-statistic structure and extend to H nthe classical exponential probability inequalities and Glivenko-Cantelli convergence properties known for the usual empirical df. An important class of statistics is given byT(H n), where T(·) is a generalized form of L-functional. For such statisticswe prove almost sure convergence using an approach which separates the functional-analytic and stochastic components of the problem and handles the latter component by application of Glivenko-Cantelli type properties.Classical results for U-statistics and L-statistics are obtained as special cases without addition of unnecessary restrictions.Many important new types of statistics of current interest are covered as well by our result.
Article PDF
Similar content being viewed by others
References
Bickel, P.J., Lehmann, E.L.: Descriptive statistics for non-parametric models. IV. Spread. In: Jurečková, J. (ed.) Contributions to statistics. Hájek Memorial Volume, pp. 33–40. Prague: Academia 1979
Chung, K.L.: An estimate concerning the Kolmogorov limit distribution. Trans. Am. Math. Soc. 67, 36–50 (1949)
Dunford, N., Schwartz, J.T.: Linear operators, vol. I. New York: Wiley 1958
Dvoretzky, A., Kiefer, J., Wolfowitz, J.: Asymptotic minimax character of the sample distribution function and of the classical multinomial estimator. Ann. Math. Statist. 27, 642–669 (1956)
Gaenssler, P.: Empirical processes, Hayward, Calif.: Institute of Mathematical Statistics, Lecture Notes-Monograph Series, vol. 3 (1983)
Helmers, R.: Edgeworth expansions for linear combinations of order statistics. Mathematical Centre Tracts, vol. 105. Amsterdam: Mathematisch Centrum 1982
Helmers, R., Janssen, P., Serfling, R.: Glivenko-Cantelli properties of some generalized empirical df's and strong convergence of generalized L-statistics. Technical Report No. 460, Dept. of Math. Sciences, Johns Hopkins University, Baltimore (1985)
Hoeffding, W.: The strong law of large numbers for U-statistics. Univ. of North Carolina Institute of Statistics Mimeo Series, No. 302 (1961)
Hoeffding, W.: Probability inequalities for sums of bounded random variables. J. Am. Stat. Assoc. 58, 13–30 (1963)
Janssen, P., Serfling, R., Veraverbeke, N.: Asymptotic normality for a general class of statistical functions and application to measure of spread. Ann. Stat. 12, 1369–1379 (1984)
McConnell, T.R.: Two-parameter strong laws and maximal inequalities for U-statistics. Proc. of the Royal Soc. of Edinburgh 107A, 133–151 (1987)
Sen, P.K.: Almost sure convergence of generalized U-statistics. Ann. Probab. 5, 287–290 (1977)
Serfling, R.J.: Approximation theorems of mathematical statistics. New York: Wiley 1980
Serfling, R.J.: Generalized L-, M-and R-statistics. Ann. Stat. 12, 76–86 (1984)
Serfling, R.J.: A bahadur representation for quantiles of empirical df's of generalized U-statistic structure. Technical Report No. 453, Dept. of Math. Sciences, Johns Hopkins University, Baltimore (1985a)
Serfling, R.J.: A note on convergence of functions of random elements. Technical Report No. 459, Dept. of Math. Sciences, Johns Hopkins University, Baltimore (1985b)
van Zwet, W. R.: A strong law for linear functions of order statistics. Ann. Probab. 8, 986–990 (1980)
Wellner, J.A.: A Glivenko-Cantelli theorem and strong laws of large numbers for functions of order statistics. Ann. Stat. 5, 473–480 (1977). Correction, ibid. 6, 1394 (1977)
Author information
Authors and Affiliations
Additional information
Research supported by the U.S. Department of Navy under Office of Naval Research Contract No. N00014-79-C-0801 and by NATO under Research Grant No. 0034/87
Rights and permissions
About this article
Cite this article
Helmers, R., Janssen, P. & Serfling, R. Glivenko-Cantelli properties of some generalized empirical DF's and strong convergence of generalized L-statistics. Probab. Th. Rel. Fields 79, 75–93 (1988). https://doi.org/10.1007/BF00319105
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF00319105