Algebra structure for L^p of a vector measure

(L^1)-algebras on groups, semigroups, etc. (43A20) Spaces of measurable functions ((L^p)-spaces, Orlicz spaces, Köthe function spaces, Lorentz spaces, rearrangement invariant spaces, ideal spaces, etc.) (46E30) Vector-valued measures and integration (46G10)

Review text

Let \(G\) be a compact abelian group and consider its Borel \(\sigma\)-algebra \(\mathcal{B}(G)\). Let \(\nu: \mathcal{B}(G) \to X\) be a vector measure defined on a Banach space \(X\). Let \( 1 \leq p < \infty\). This paper presents the theory of vector measures satisfying a translation invariance type property that allows to provide an algebra structure on the space \(L^p(\nu)\) of \(p\)-integrable functions with respect to the vector measure \(\nu\). A vector measure is said to satisfy the norm integral translation invariant property (n.i.t.i.)\ if, for every translation \(\tau_a\) defined in the compact group \(G\), the inequality \(\| \int_G \tau_a \varphi \, d \nu \| \leq \| \int_G \varphi \, d\nu\|\) holds for every simple function \(\varphi\). In particular, this property implies that both the space \(L^p(\nu)\) and the space of weakly \(p\)-integrable functions \(L^p_w(\nu)\) are translation invariant (Corollary 3.7). In Section~4, under the n.i.t.i.\ assumption for \(\nu\), it is shown that the convolution product is well-defined in \(L^p(\nu)\). In order to do this, it is proven that the space \(L^1_w(\nu)\) is embedded in \(L^1(G)\) and so, since \(L^p_w(\nu)\) is always included in \(L^1_w(\nu)\), the convolution product is given in \(L^p(\nu)\) and in \(L^p_w(\nu)\). The proof of the result that allows the construction (Theorem 4.1) is modelled around a clever argument based on a fixed point theorem for families of operators (Markov-Kakutani Theorem). This is the key result that establishes the link between Rybakov control measures of the measure \(\nu\) and the Haar measure associated to the compact group \(G\). It must be pointed out that the use of this kind of argument is new in this setting, and consequently it provides new tools for the analysis of the structure of spaces \(L^p(\nu)\) of a vector measure \(\nu\), in particular when it is defined on a compact abelian group. Relevant examples of vector measures satisfying the n.i.t.i.\ property are explained in the last part of the paper (Section~5); in particular, the vector measures associated to the Fourier transform and the convolution operators satisfy this property. In the opinion of the reviewer, the techniques and results presented in this paper provide a new point of view for the applications of the spaces \(L^p(\nu)\) of a vector measure in Harmonic Analysis.

Recommendations

Cites work

Cited in

(11)

This page was built for publication: Algebra structure for \(L^p\) of a vector measure

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2389263)