New methods for bias correction at endpoints and boundaries. (English) Zbl 1014.62041

Hall, Peter; Park, Byeong U.

Ann. Stat. 30, No. 5, 1460-1479 (2002).

Summary: We suggest two new, translation-based methods for estimating and correcting for bias when estimating the edge of a distribution. The first uses an empirical translation applied to the argument of the kernel, in order to remove the main effects of the asymmetries that are inherent when constructing estimators at boundaries. Placing the translation inside the kernel is in marked contrast to traditional approaches, such as the use of high-order kernels, which are related to the jackknife and, in effect, apply the translation outside the kernel. Our approach has the advantage of producing bias estimators that, while enjoying a high order of accuracy, are guaranteed to respect the sign of bias.
Our second method is a new bootstrap technique. It involves translating an initial boundary estimate toward the body of the dataset, constructing repeated boundary estimates from data that lie below the respective translations, and employing averages of the resulting empirical bias approximations to estimate the bias of the original estimator. The first of the two methods is most appropriate in univariate cases, and is studied there; the second approach may be used to bias-correct estimates of boundaries of multivariate distributions, and is explored in the bivariate case.

MSC:

62G07	Density estimation
62G20	Asymptotic properties of nonparametric inference

Keywords:

bias estimation; bootstrap; curve estimation; free disposal hull estimator; frontier estimation; kernel methods; productivity analysis; translations

Software:

KernSmooth

References:

[1]	ATHREy A, K. B. (1987a). Bootstrap of the mean in the infinite variance case. Ann. Statist. 15 724- 731. · Zbl 0628.62042 · doi:10.1214/aos/1176350371
[2]	ATHREy A, K. B. (1987b). Bootstrap of the mean in the infinite variance case. In Proceedings of the First World Congress of Bernoulli Society 2 95-98. VNU Scientific Press, Utrecht.
[3]	BICKEL, P., GÖTZE, F. and VAN ZWET, W. R. (1997). Resampling fewer than n observations: Gains, losses, and remedies for losses. Statist. Sinica 7 1-31. · Zbl 0927.62043
[4]	BLOCH, D. A. and GASTWIRTH, J. L. (1968). On a simple estimate of the reciprocal of the density function. Ann. Math. Statist. 39 1083-1085. · Zbl 0245.62043 · doi:10.1214/aoms/1177698342
[5]	BOFINGER, E. (1975). Estimation of a density function using order statistics. Austral. J. Statist. 17 1-7. · Zbl 0346.62038 · doi:10.1111/j.1467-842X.1975.tb01366.x
[6]	CHENG, C. (1995). Uniform consistency of generalized kernel estimators of quantile density. Ann. Statist. 23 2285-2291. · Zbl 0853.62031 · doi:10.1214/aos/1034713657
[7]	CHENG, C. and PARZEN, E. (1997). Unified estimators of smooth quantile and quantile density functions. J. Statist. Plann. Inference 59 291-307. · Zbl 0900.62209 · doi:10.1016/S0378-3758(96)00110-3
[8]	CHEVALIER, J. (1976). Estimation du support et du contenu du support d’une loi de probabilité. Ann. Inst. H. Poincaré Probab. Statist. 12 339-364. · Zbl 0372.62036
[9]	CHOI, E. and HALL, P. (1999). Data sharpening as a prelude to density estimation. Biometrika 86 941-947. JSTOR: · Zbl 0942.62038 · doi:10.1093/biomet/86.4.941
[10]	CSÖRG O, M. (1983). Quantile Processes with Statistical Applications. SIAM, Philadelphia. · Zbl 0518.62043
[11]	DEPRINS, D., SIMAR, L. and TULKENS, H. (1984). Measuring labor efficiency in post offices. In The Performance of Public Enterprises: Concepts and Measurements (M. Marchand, P. Pestieau and H. Tulkens, eds.) 243-267. North-Holland, Amsterdam.
[12]	FALK, M. (1986). On the estimation of the quantile density function. Statist. Probab. Lett. 4 69-73. · Zbl 0585.62076 · doi:10.1016/0167-7152(86)90020-9
[13]	GIJBELS, I., MAMMEN, E., PARK, B. U. and SIMAR, L. (1999). On estimation of monotone and concave frontier functions. J. Amer. Statist. Assoc. 94 220-228. JSTOR: · Zbl 1043.62105 · doi:10.2307/2669696
[14]	HALL, P. (1989). On convergence rates in nonparametric problems. Internat. Statist. Rev. 57 45-58. · Zbl 0707.62091 · doi:10.2307/1403583
[15]	HALL, P. (1990). Asy mptotic properties of the bootstrap for heavy-tailed distributions. Ann. Probab. 18 1342-1360. · Zbl 0714.62035 · doi:10.1214/aop/1176990748
[16]	HALL, P., PARK, B. U. and STERN, S. E. (1998). On poly nomial estimators of frontiers and boundaries. J. Multivariate Anal. 66 71-98. · Zbl 1127.62358 · doi:10.1006/jmva.1998.1738
[17]	HALL, P., PARK, B. U. and TURLACH, B. A. (1998). Rolling-ball method for estimating the boundary of the support of a point-process intensity. Ann. Inst. H. Poincaré Probab. Statist. · Zbl 1011.62035 · doi:10.1016/S0246-0203(02)01132-9
[18]	HÄRDLE, W., PARK, B. U. and TSy BAKOV, A. B. (1995). Estimation of nonsharp support boundaries. J. Multivariate Anal. 55 205-218. · Zbl 0863.62030
[19]	JONES, M. C. (1992). Estimating densities, quantiles, quantile densities and density quantiles. Ann. Inst. Statist. Math. 44 721-727. · Zbl 0772.62022 · doi:10.1007/BF00053400
[20]	JONES, M. C. and SIGNORINI, D. F. (1997). A comparison of higher-order bias kernel density estimators. J. Amer. Statist. Assoc. 92 1063-1073. JSTOR: · Zbl 0888.62035 · doi:10.2307/2965571
[21]	KNEIP, A., PARK, B. U. and SIMAR, L. (1998). A note on the convergence of nonparametric DEA estimators for production efficiency scores. Econom. Theory 14 783-793. JSTOR: · doi:10.1017/S0266466698146042
[22]	KNIGHT, K. (1989). On the bootstrap of the sample mean in the infinite variance case. Ann. Statist. 17 1168-1175. · Zbl 0687.62017 · doi:10.1214/aos/1176347262
[23]	KOROSTELEV, A. P., SIMAR, L. and TSy BAKOV, A. B. (1995). Efficient estimation of monotone boundaries. Ann. Statist. 23 476-489. · Zbl 0829.62043 · doi:10.1214/aos/1176324531
[24]	KOROSTELEV, A. P. and TSy BAKOV, A. B. (1993). Minimax Theory of Image Reconstruction. Springer, New York. · Zbl 0833.62039
[25]	LINTON, O. and NIELSEN, J. P. (1994). A multiplicative bias reduction method for nonparametric regression. Statist. Probab. Lett. 19 181-187. · Zbl 0791.62043 · doi:10.1016/0167-7152(94)90102-3
[26]	MAMMEN, E. and TSy BAKOV, A. B. (1995). Asy mptotically minimax recovery of sets with smooth boundaries. Ann. Statist. 23 502-524. · Zbl 0834.62038 · doi:10.1214/aos/1176324533
[27]	PARK, B. U., SIMAR, L. and WEINER, C. (2000). The FDH estimator for productivity efficiency scores: Asy mptotic properties. Econom. Theory 16 855-877. JSTOR: · Zbl 0967.62102 · doi:10.1017/S0266466600166034
[28]	POLITIS, D., ROMANO, J. P. and WOLF, M. (1999). Subsampling. Springer, New York. · Zbl 0943.60003 · doi:10.1016/S0378-3758(98)00174-8
[29]	REISS, R.-D. (1978). Approximate distribution of the maximum deviation of histograms. Metrika 25 9-26. · Zbl 0375.62018 · doi:10.1007/BF02204347
[30]	RIPLEY, B. D. and RASSON, J. P. (1977). Finding the edge of a Poisson forest. J. Appl. Probab. 14 483-491. JSTOR: · Zbl 0373.62058 · doi:10.2307/3213451
[31]	SAMIUDDIN, M. and EL-SAy YAD, G. M. (1990). On nonparametric kernel density estimates. Biometrika 77 865-874. JSTOR: · Zbl 0712.62033 · doi:10.1093/biomet/77.4.865
[32]	SIDDIQUI, M. M. (1960). Distribution of quantiles in samples from a bivariate population. Journal of Research of the National Bureau Standards 64B 145-150. · Zbl 0096.13402
[33]	SIMAR, L. and WILSON, P. W. (1998). A general methodology for bootstrapping in nonparametric frontier models. Research Report DP 9811, Institut de Statistique, Univ. Catholique de Louvain, Belgium. · Zbl 1022.62125
[34]	WAND, M. P. and JONES, M. C. (1995). Kernel Smoothing. Chapman and Hall, London. · Zbl 0854.62043
[35]	WELSH, A. H. (1988). Asy mptotically efficient estimation of the sparsity function at a point. Statist. Sinica 6 427-432. · Zbl 0637.62043 · doi:10.1016/0167-7152(88)90103-4
[36]	CANBERRA, ACT 0200 AUSTRALIA DEPARTMENT OF STATISTICS SEOUL NATIONAL UNIVERSITY SEOUL 151-747 KOREA E-MAIL: [e-mail]