Optimal bandwidth choice for density-weighted averages. (English) Zbl 0864.62022
Summary: This paper characterizes the optimal bandwidth value for estimating density-weighted averages, statistics that arise in semiparametric estimation methods for index models and models of selected samples based on nonparametric kernel estimators. The optimal bandwidth is derived by minimizing the leading terms of mean squared error of the density-weighted average. The optimal bandwidth formulation is developed by comparison to the optimal pointwise bandwidth of a naturally associated nonparametric estimation problem, highlighting the role of sample size and the structure of nonparametric estimation bias.
The methods are illustrated by estimators of average density, density-weighted average derivatives and conditional covariances, and bandwidth values are calculated for normal designs. A simple ‘plug-in’ estimator for the optimal bandwidth is proposed. Finally, the optimal bandwidth for estimating ratios of density-weighted averages is derived, showing that the earlier optimal formulae can be implemented directly using naturally defined ‘residual’ values.
The methods are illustrated by estimators of average density, density-weighted average derivatives and conditional covariances, and bandwidth values are calculated for normal designs. A simple ‘plug-in’ estimator for the optimal bandwidth is proposed. Finally, the optimal bandwidth for estimating ratios of density-weighted averages is derived, showing that the earlier optimal formulae can be implemented directly using naturally defined ‘residual’ values.
MSC:
| 62G07 | Density estimation |
| 62P20 | Applications of statistics to economics |
| 62G05 | Nonparametric estimation |
Keywords:
plug-in estimator; smoothing; optimal bandwidth; density-weighted averages; semiparametric estimation methods; index models; models of selected samples; kernel estimators; mean squared error; bias; average density; derivative; conditional covariancesReferences:
| [1] | Andrews, D. W. K.: Asymptotics for semiparametric econometric models: I. Estimation. (1989) |
| [2] | Bickel, P. J.; Ritov, Y.: Estimating integrated squared density derivatives: sharp best order of convergence estimates. Sankhyã 50 A, 381-393 (1988) · Zbl 0676.62037 |
| [3] | Gasser, T.; Kneip, A.; Kohler, W.: A flexible and fast method for automatic smoothing. Journal of the American statistical association 86, 643-653 (1991) · Zbl 0733.62047 |
| [4] | Goldstein, L.; Messer, K.: Optimal plug-in estimators for nonparametric functional estimation. Technical report no. 277 (1990) · Zbl 0763.62023 |
| [5] | Hall, P.; Marron, J. S.: Estimation of integrated squared density derivatives. Statistics and probability letters 6, 109-115 (1987) · Zbl 0628.62029 |
| [6] | Härdle, W.: Applied nonparametric regression. Econometric society monographs (1991) · Zbl 0714.62030 |
| [7] | Härdle, W.; Stoker, T. M.: Investigating smooth multiple regression by the method of average derivatives. Journal of the American statistical association 84, 986-995 (1989) · Zbl 0703.62052 |
| [8] | Härdle, W.; Tsybakov, A. B.: How sensitive are average derivatives?. Journal of econometrics 58, 31-48 (1993) · Zbl 0772.62021 |
| [9] | Härdle, W.; Hall, P.; Marron, S.: How far are automatically chosen regression smoothing parameters from their optimum?. Journal of the American statistical association 83, 86-95 (1988) · Zbl 0644.62048 |
| [10] | Härdle, W.; Hart, J.; Marron, J. S.; Tsybakov, A. B.: Bandwidth choice for average derivative estimation. Journal of the American statistical association 87, 227-233 (1992) · Zbl 0781.62044 |
| [11] | Hastie, T. J.; Tibshirani, R. J.: Generalized additive models. (1990) · Zbl 0747.62061 |
| [12] | Hoeffding, W.: A class of statistics with asymptotically normal distribution. Annals of mathematical statistics 19, 293-325 (1948) · Zbl 0032.04101 |
| [13] | Jaeckel, L. A.: Estimating regression coefficients by minimizing the dispersion of the residuals. Annals of mathematical statistics 43, 1449-1458 (1972) · Zbl 0277.62049 |
| [14] | Jones, M. C.; Sheather, S. J.: Using non-stochastic terms to advantage in kernel-based estimation of integrated squared density derivatives. Statistics and probability letters 11, 511-514 (1991) · Zbl 0724.62040 |
| [15] | Jurečková, J.: Nonparametric estimate of regression coefficients. Annals of mathematical statistics 42, 1328-1338 (1971) · Zbl 0225.62052 |
| [16] | Newey, W. K.: The asymptotic variance of semiparametric estimators. Econometrica 62, 1349-1382 (1994) · Zbl 0816.62034 |
| [17] | Nychka, D.: Choosing a range for the amount of smoothing in nonparametric regression. Journal of the American statistical association 86, 653-665 (1991) |
| [18] | Powell, J. L.: Semiparametric estimation of bivariate latent variables models. Draft (1987) |
| [19] | Powell, J. L.; Stock, J. H.; Stoker, T. M.: Semiparametric estimation of index coefficients. Econometrica 57, 1403-1430 (1989) · Zbl 0683.62070 |
| [20] | Robinson, P. M.: Root-N-consistent semiparametric regression. Econometrica 56, 931-954 (1988) · Zbl 0647.62100 |
| [21] | Serfling, R. J.: Approximation theorems of mathematical statistics. (1980) · Zbl 0538.62002 |
| [22] | Stoker, T. M.: Lectures on semiparametric econometrics, CORE lecture series. (1992) |
| [23] | Stoker, T. M.: Smoothing bias in the estimation of density derivatives. Journal of the American statistical association 88, 855-863 (1993) · Zbl 0780.62039 |
| [24] | Stoker, T. M.: Smoothing bias in the measurement of marginal effects. Journal of econometrics (1993) · Zbl 0843.62048 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
