NONCONVEXITY PHENOMENON ON ITÔ'S INTEGRALS AND ON STOCHASTIC ATTAINABLE SETS
shanjian tang
DOI: https://doi.org/10.1142/9789812790552_0011
2007-01-01
Abstract:It is well-known that the vector-valued measure defined by Lebesgue’s integrals for lR− valued functions have closed and convex ranges. The result was fist proved by Lyapunov 8 in 1940, and its many other proofs can be found in the literature ( see Lindenstrauss , Yorke , and Tardella ). The finiteness of the space where the associated functions take their values turns out to be crucial. In 1946, Lyapunov 9 pointed out that the assertion may fail to hold for Bochner’s integrals. In 1969, however, Uhl 13 proved that the closure of the range of the vector-valued measure defined via Bochner’s integrals is convex. Consider a complete probability space (Ω,F ,Ft, P ) with the filtration {Ft; t ∈ [0, 1]} satisfying the so-called “usual conditions”. Let L(Ω,F , P ; lR) denote the space of lR-valued square integrable random variables defined on (Ω,F , P ). Let {W (t); t ∈ [0, 1]} be a one-dimensional (Ft)−adapted standard Brownian motion, and let LF(0, 1; lR ) denote the space of lR-valued (Ft)-adapted processes f(·) such that