A $U$-process is a collection of $U$-statistics indexed by a family of symmetric kernels. In this paper, two functional limit theorems are obtained for sequences of standardized $U$-processes. In one case the limit process is Gaussian; in the other, the limit process has finite dimensional distributions of infinite weighted sums of $\chi^2$ random variables. Goodness-of-fit statistics provide examples.
@article{1176991691,
author = {Nolan, Deborah and Pollard, David},
title = {Functional Limit Theorems for $U$-Processes},
journal = {Ann. Probab.},
volume = {16},
number = {4},
year = {1988},
pages = { 1291-1298},
language = {en},
url = {http://dml.mathdoc.fr/item/1176991691}
}
Nolan, Deborah; Pollard, David. Functional Limit Theorems for $U$-Processes. Ann. Probab., Tome 16 (1988) no. 4, pp. 1291-1298. http://gdmltest.u-ga.fr/item/1176991691/