Range convergence monotonicity for vector measures and range monotonicity of the mass (Q2184197)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Range convergence monotonicity for vector measures and range monotonicity of the mass
scientific article

    Statements

    Range convergence monotonicity for vector measures and range monotonicity of the mass (English)
    0 references
    0 references
    0 references
    27 May 2020
    0 references
    In [Proc. Am. Math. Soc. 111, No. 1, 205--214 (1991; Zbl 0727.28005)], \textit{L. Rodríguez-Piazza} showed that, if the closed convex hulls of the ranges of two finitely additive vector measures with values in a normed space coincide, then their total variations also coincide. In the article under review, the authors consider sequences of finitely additive vector measures and the relationship between the sequences of their ranges and the sequences of their total variations. Let \(X\) be a locally compact topological space and let \(\mathcal{B}\) be the \(\sigma\)-algebra of Borel sets in \(X\). If \((\mu_n)\) is a sequence of vector measures from \(X\) to a vector space \(V\) and \(\mu\) is a vector measure from \(X\) to \(V\), the sequence \((\mu_n)\) \textit{converges widely} to \(\mu\) if, for every continuous real-valued function on \(X\) with compact support, \(\displaystyle\lim_{n\to\infty} \int_X \phi\, d\mu_n = \int_X \phi\,d\mu\)\,. (This is often referred to as weak or weak\(^*\) convergence of the sequence of measures.) The authors define the \textit{range} \(\text{rg}\,\mu(X)\) of a vector measure \(\mu\) to be the closed convex hull of the set \(\{ \mu(A): A\in \mathcal{B}\}\). The authors prove the following result: Let \(X\) be a locally compact separable metric space endowed with its Borel \(\sigma\)-algebra and let \(V\) be a finite-dimensional space with seminorm \(|\cdot|_V\). If \((\mu_n)\) is a sequence of Borel vector measures from \(X\) to \(V\) that converges widely to a Borel vector measure \(\mu\) and if the sequence of ranges \((\text{rg}\,\mu_n(X))\) converges to \(\text{rg}\,\mu(X)\) in Hausdorff distance, then the sequence of total variations \((|\mu_n|_V(X))\) converges to \(|\mu|_V(X)\). Although the converse of this theorem does not hold in general, the authors prove that if a finite-dimensional normed linear space \((V,|\cdot|_V)\) is strictly convex and if \((\mu_n)\) is a sequence of Borel vector measures from \(X\) to \(V\) converging widely to a vector measure \(\mu\), then the sequence \((\text{rg}\,\mu_n(X))\) of ranges converges to \(\text{rg}\,\mu(X)\) in Hausdorff distance if and only if the sequence \((|\mu_n|(X))\) of total variations converges to \(|\mu|(X)\). Other results connecting the monotonicity of ranges of two vector measures with their total variations are also considered.
    0 references
    0 references
    wide convergence of vector measures
    0 references
    total variation
    0 references
    strict convexity
    0 references
    zonoids
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references