In this paper we derive almost sure convergence of kernel-type conditional \(U\)-statistics in \(p\) th mean under mild conditions on the smoothing parameter. An application to discrimination is discussed in detail. (C) 1994 Academic Press. Inc.
Lp-Continuity of Conditional Expectations
โ Scribed by Alberto Alonso; Fernando Brambila-Paz
- Publisher
- Elsevier Science
- Year
- 1998
- Tongue
- English
- Weight
- 213 KB
- Volume
- 221
- Category
- Article
- ISSN
- 0022-247X
No coin nor oath required. For personal study only.
โฆ Synopsis
A necessary and sufficient condition on a sequence n nโN of ฯ-subalgebras that assures L p -convergence of the conditional expectations is given. This result generalizes the L p -martingales, the Fetter and the Boylan (equiconvergence) theorems.
๐ SIMILAR VOLUMES
We prove that the martingale convergence theorem for generalized conditional expectations in von Neumann algebras holds in the weak topology without restrictions. The situation is therefore different fom the strong topology case, where there are restrictive conditions which distinguish between incre
## Abstract For an essentially bounded closedโconvexโnonempty setโvalued random variable, it is shown that the conditional expectation is characterized by its integrals over the sets of the associated __ฯ__ โfield. (ยฉ 2006 WILEYโVCH Verlag GmbH & Co. KGaA, Weinheim)
The continuity equation with the random velocity field in the d-dimensional space is studied and the existence of the corresponding family of random evolution operators is proved. In case of the random velocities defined by means of the Gaussian random fields the expectations of the evolution operat