Summary
A parameter which may be represented as a functionalT(F) of a distribution functionF may be estimated by the “statistical function”T(Fn), whereFn is the empirical distribution function. Recently, Boos and Serfling (1979, Florida State University Statistics Report No. M 499) obtained sufficient conditions for the Berry-Esseen theorem to hold forT(Fn)-T(F) and applied the results to derive rates of convergence inL∞ forL-estimates. The present note complements their work by obtaining theLp-rates of convergence, 1≦p<∞ forT(Fn)-T(F) and its application toL-estimates.
Similar content being viewed by others
References
Ahmad, I. A. (1981). On some asymptotic properties ofU-statistics,Scand. J. Statist.,8, 175–182.
Boos, D. D. and Serfling, R. J. (1979). On Berry-Esseen rates for statistical functions with application toL-estimates, Florida State University Statistics Report No. M499.
Callaert, H. and Janssen, P. (1978). The Berry-Esseen theorem forU-statistics,Ann. Statist.,6, 417–421.
Heyde, C. C. (1975). A non-uniform bound on convergence to normality,Ann. Prob.,3, 903–907.
Ibragimov, I. A. (1966). On the accuracy of Gaussian approximation to distribution function of sums of independent variables,Theory Prob. Appl.,11, 559–579.
Von Mises, R. (1947). On the asymptotic distributions of differentiable statistical functions,Ann. Math. Statist.,18, 309–348.
Author information
Authors and Affiliations
About this article
Cite this article
Ahmad, I.A. OnLp-convergence rates for statistical functions with application toL-estimates. Ann Inst Stat Math 35, 401–406 (1983). https://doi.org/10.1007/BF02480996
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/BF02480996