Kernel and Nearest-Neighbor Estimation of a Conditional Quantile
Bhattacharya, P. K. ; Gangopadhyay, Ashis K.
Ann. Statist., Tome 18 (1990) no. 1, p. 1400-1415 / Harvested from Project Euclid
Let $(X_1, Z_1), (X_2, Z_2), \ldots, (X_n, Z_n)$ be iid as $(X, Z), Z$ taking values in $R^1$, and for $0 < p < 1$, let $\xi_p(x)$ denote the conditional $p$-quantile of $Z$ given $X = x,$ i.e., $P(Z \leq \xi_p(x)\mid X = x) = p$. In this paper, kernel and nearest-neighbor estimators of $\xi_p(x)$ are proposed. In order to study the asymptotics of these estimates, Bahadur-type representations of the sample conditional quantiles are obtained. These representations are used to examine the important issue of choosing the smoothing parameter by a local approach (for a fixed $x$) based on weak convergence of these estimators with varying $k$ in the $k$-nearest-neighbor method and with varying $h$ in the kernel method with bandwidth $h$. These weak convergence results lead to asymptotic linear models which motivate certain estimators.
Publié le : 1990-09-14
Classification:  Conditional quantile,  kernel estimator,  nearest-neighbor estimator,  Bahadur representation,  weak convergence,  Browian motion,  asymptotic linear model,  order statistics,  induced order statistics,  62G05,  62J02,  62G20,  62G30,  60F17
@article{1176347757,
     author = {Bhattacharya, P. K. and Gangopadhyay, Ashis K.},
     title = {Kernel and Nearest-Neighbor Estimation of a Conditional Quantile},
     journal = {Ann. Statist.},
     volume = {18},
     number = {1},
     year = {1990},
     pages = { 1400-1415},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176347757}
}
Bhattacharya, P. K.; Gangopadhyay, Ashis K. Kernel and Nearest-Neighbor Estimation of a Conditional Quantile. Ann. Statist., Tome 18 (1990) no. 1, pp.  1400-1415. http://gdmltest.u-ga.fr/item/1176347757/