A Note on the $L_1$ Consistency of Variable Kernel Estimates
Devroye, Luc
Ann. Statist., Tome 13 (1985) no. 1, p. 1041-1049 / Harvested from Project Euclid
A sample $X_1, \cdots, X_n$ of i.i.d. $R^d$-valued random vectors with common density $f$ is used to construct the density estimate $f_n(x) = (1/n) \sum^n_{i = 1} H^{-d}_{ni}K((x - X_i)/H_{ni}),$ where $K$ is a given density on $R^d$, and the $H_{ni}$'s are positive functions of $n, i$ and $X_1, \cdots, X_n$ (but not of $x$). The $H_{ni}$'s can be thought of as locally adapted smoothing parameters. We give sufficient conditions for the weak convergence to 0 of $\int |f_n - f|$ for all $f$. This is illustrated for the estimate of Breiman, Meisel and Purcell (1977).
Publié le : 1985-09-14
Classification:  Nonparametric density estimation,  consistency,  variable kernel estimate,  nearest neighbor,  embedding,  60F15,  62G05
@article{1176349655,
     author = {Devroye, Luc},
     title = {A Note on the $L\_1$ Consistency of Variable Kernel Estimates},
     journal = {Ann. Statist.},
     volume = {13},
     number = {1},
     year = {1985},
     pages = { 1041-1049},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176349655}
}
Devroye, Luc. A Note on the $L_1$ Consistency of Variable Kernel Estimates. Ann. Statist., Tome 13 (1985) no. 1, pp.  1041-1049. http://gdmltest.u-ga.fr/item/1176349655/