On the rate of convergence and asymptotic expansions for U-statistics under alternatives

In this paper, we consider asymptotic expansions and the rate of convergence for the distribution function of asymptotically efficient U-statistics under alternatives in the one-sample problem. Section 1 is an introduction. Section S contains the theorem concerning the rate of convergence for U-statistics; in Sec. 3, we formulate sets of sufficient conditions under which Edgeworth-type asymptotic expansions for U-statistics under alternatives will be constructed (see Theorem 2). Finally, these theorems are proved in Sec. 4. © 2000 Kluwer Academic/Plenum Publishers.