scientific article; zbMATH DE number 3861490

From MaRDI portal
Publication:3329685





For E a l.c.s., the \(\mu\)-equivalence \(f\cong g\) for vector-valued functions \(\Omega \to E\) with respect to a measure \(\mu\) is defind. Given a strictly localizable measure space (\(\Omega\),\(\Sigma\),\(\mu)\), a lifting \(\rho\) on the vector space of the \(\mu\)-simple functions \(f:\Omega \to E\) with values in the l.c.s. E is defined. Under this hypothesis we have the following statement: Let f:\(\Omega \to E\) be a \(\mu\)-integrable function and F a closed subset of E such that 0\(\in F\) and \((\mu(A))^{-1}\int_{A}fd\mu \in F\) for every \(A\in \Sigma\) with \(0<\mu(A)<\infty.\) Then: (1) There exists a \(\mu\)-measurable function \(g\cong f\) such that \(\rho([g])=g\) (in the sense of Ionescu Tulcea) and g(t)\(\in F\) for every \(t\in \Omega.\) (2) For every neighbourhood U of 0 in E, \(f(t)\in F+U\) for a.e. \(t\in \Omega\). The essential range \(er_ A(f)\) for \(f:\Omega \to E\) and \(A\in \Sigma\) is also studied. [For part III see the review above.]











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3329685)