Abstract
Using the kernel estimator of the pth quantile of a distribution brings about an improvement in comparison to the sample quantile estimator. The size and order of this improvement is revealed when studying the Edgeworth expansion of the kernel estimator. Using one more term beyond the normal approximation significantly improves the accuracy for small to moderate samples. The investigation is non- standard since the influence function of the resulting L-statistic explicitly depends on the sample size. We obtain the expansion, justify its validity and demonstrate the numerical gains in using it.
Similar content being viewed by others
References
Bickel P., Götze F., van Zwet W. (1986) The Edgeworth expansion for U-statistics of degree two. The Annals of Statistics 14: 1463–1484
Borovkov K., Dehling H., Pfeifer D. (1996) On asymptotic behavior of weighted sample quantiles. Mathematical Methods of Statistics 5: 173–186
Callaert H., Janssen P., Veraverbeke N. (1980) An Edgeworth expansion for U-statistics. The Annals of Statistics 8: 299–312
Dharmadhikari S., Fabian V., Jogdeo K. (1968) Bounds on the moments of martingales. The Annals of Mathematical Statistics 39: 1719–1723
DiDonato A., Morris A. (1987) ALGORITHM 654: FORTRAN subroutines for computing the incomplete gamma function ratios and their inverse. Association for Computing Machinery Transactions on Mathematical Software 13(3): 318–319
Falk M. (1984) Relative deficiency of kernel type estimators of quantiles. The Annals of Statistics 12(1): 261–268
Falk M. (1985) Asymptotic normality of the kernel quantile estimator. The Annals of Statistics 13(1): 428–433
Feller W. (1971) An introduction to probability and its applications (Vol. 2). Wiley, New York
Helmers R. (1982) Edgeworth expansions for linear combinations of order statistics. Mathematical Centre Tracts (Vol. 105). Amsterdam, Mathematisch Centrum
Jing B.-Y., Wang Q. (2003) Edgeworth expansions for U-statistics under minimal conditions. The Annals of Statistics 31(4): 1376–1391
Kozek A. (2005) How to vombine M-estimators to estimate quantiles and a score function. Sankhya 67(2): 277–294
Lai T., Wang J. (1993) Edgeworth expansions for symmetric statistics with applications to bootstrap methods. Statistica Sinica 3: 517–542
Lorenz M. (1905) Methods of measuring concentration of wealth. Journal of the American Statistical Association 9: 209–219
Maesono Y. (1987) Edgeworth expansion for one-sample U-statistics. Bulletin of Informatics and Cybernetics 22: 189–197
Ogryczak W., Ruszczyński A. (2002) Dual stochastic dominance and related mean-risk models. SIAM Journal of Optimization 13(1): 60–78
Serfling R. (1980) Approximation theorems of mathematical statistics. Wiley, New York
Sheather S., Marron J. (1990) Kernel quantile estimators. Journal of the American Statistical Association 85(410): 410–416
Shorack G., Wellner J. (1986) Empirical processes with applications to statistics. Wiley, New York
Xiang X. (1995a) A Berry-Esseen theorem for the kernel quantile estimator with application to studying the deficiency of qunatile estimators. Annals of the Institute of Statistical Mathematics 47(2): 237–251
Xiang X. (1995b) Defficiency of the sample quantile estimator with respect to kernel quantile estimators for censored data. The Annals of Statistics 23(3): 836–854
Xiang X., Vos P. (1997) Quantile estimators and covering probabilities. Journal of Nonparametric Statistics 7: 349–363
Author information
Authors and Affiliations
Corresponding author
About this article
Cite this article
Maesono, Y., Penev, S. Edgeworth expansion for the kernel quantile estimator. Ann Inst Stat Math 63, 617–644 (2011). https://doi.org/10.1007/s10463-009-0241-5
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10463-009-0241-5