Best Possible Constant for Bandwidth Selection
Fan, Jianqing ; Marron, James S.
Ann. Statist., Tome 20 (1992) no. 1, p. 2057-2070 / Harvested from Project Euclid
For the data based choice of the bandwidth of a kernel density estimator, several methods have recently been proposed which have a very fast asymptotic rate of convergence to the optimal bandwidth. In particular the relative rate of convergence is the square root of the sample size, which is known to be the best possible. The point of this paper is to show how semiparametric arguments can be employed to calculate the best possible constant coefficient, that is, an analog of the usual Fisher information, in this convergence. This establishes an important benchmark as to how well bandwidth selection methods can ever hope to perform. It is seen that some existing methods attain the bound, others do not.
Publié le : 1992-12-14
Classification:  Bandwidth selection,  efficient bounds,  kernel density estimator,  semiparametric methods,  62G07,  62B10,  62C20
@article{1176348902,
     author = {Fan, Jianqing and Marron, James S.},
     title = {Best Possible Constant for Bandwidth Selection},
     journal = {Ann. Statist.},
     volume = {20},
     number = {1},
     year = {1992},
     pages = { 2057-2070},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176348902}
}
Fan, Jianqing; Marron, James S. Best Possible Constant for Bandwidth Selection. Ann. Statist., Tome 20 (1992) no. 1, pp.  2057-2070. http://gdmltest.u-ga.fr/item/1176348902/