An Infinite-Dimensional Approximation for Nearest Neighbor Goodness of Fit Tests
Schilling, Mark F.
Ann. Statist., Tome 11 (1983) no. 1, p. 13-24 / Harvested from Project Euclid
Let $X_1, \cdots, X_n$ be i.i.d. $\mathbb{R}^m$-valued observations from a bounded density $g(x)$ continuous on its support. Let $W_i = \exp\{- ng(X_i)V(R_i)\}, i = 1, \cdots, n$, where $V(R_i)$ is the volume of the nearest neighbor sphere around $X_i$, and let $w(x)$ be any bounded continuous weight function on $\mathbb{R}^m$. An infinite-dimensional approximation to the asymptotic form of the weighted empirical distribution function of the $W_i$'s is presented. The distributions of quadratic functionals of the limiting normalized weighted e.d.f. are found and tabulated for $m = \infty$ and $m = 1$ and compared with finite $m > 1$. Monte Carlo results are given for $n, m < \infty$.
Publié le : 1983-03-14
Classification:  Nearest neighbor,  infinite-dimensional approximation,  empirical distribution function,  quadratic functional,  62M99,  62G10,  62H15,  62E20
@article{1176346052,
     author = {Schilling, Mark F.},
     title = {An Infinite-Dimensional Approximation for Nearest Neighbor Goodness of Fit Tests},
     journal = {Ann. Statist.},
     volume = {11},
     number = {1},
     year = {1983},
     pages = { 13-24},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176346052}
}
Schilling, Mark F. An Infinite-Dimensional Approximation for Nearest Neighbor Goodness of Fit Tests. Ann. Statist., Tome 11 (1983) no. 1, pp.  13-24. http://gdmltest.u-ga.fr/item/1176346052/