In this paper we consider a Hamiltonian H on P 2 (R 2d ), the set of probability measures with finite quadratic moments on the phase space R 2d = R d × R d , which is a metric space when endowed with the Wasserstein distance W 2 . We study the initial value problem dµ t /dt +∇•(J d v t µ t ) = 0, wh
Embedding the Abstract Wiener Space in a Probability Space
✍ Scribed by A.S. Üstünel; M. Zakai
- Publisher
- Elsevier Science
- Year
- 2000
- Tongue
- English
- Weight
- 160 KB
- Volume
- 171
- Category
- Article
- ISSN
- 0022-1236
No coin nor oath required. For personal study only.
✦ Synopsis
In the first part, of this paper it is pointed out that for certain applications of the stochastic calculus of variations it is useful to replace the classical domain of definition the Wiener space with a general probability space in which the Wiener space is embedded. This yields a certain conditional Malliavin calculus'' and is applicable to signal'' and ``noise'' problems. In a somewhat analogous way, it is pointed out in the second part of the paper that formulating the Ito^calculus in a setup of an abstract Wiener space embedded in a general probability space endowed with a filtration has certain useful applications. In particular it enables the formulation and derivation of a dimension-free form of the Girsanov theorem as well as a dimension free form of the representation of L p -Wiener functionals as Ito^integrals.
📜 SIMILAR VOLUMES
Stochastic oscillatory integrals with quadratic phase function are studied in the case where amplitude functions are multiple Wiener integrals. We show that a principle of stationary phase holds when kernels of multiple Wiener integrals are of trace class. On the way, we establish a new criterion fo