scientific article; zbMATH DE number 3851480

From MaRDI portal
Publication:3320664





Let U be a domain in the complex plane \(\mathbb{C}\) and \(F_ c(\mathbb{C})\) be the family of all non-empty compact subsets in \(\mathbb{C}\), with the usual topology. An upper semi-continuous map \(f: U\to F_ c(\mathbb{C})\) is called an analytic set-valued function, in the sense of K. Oka, iff the set of all points \((u,z)\in U\times \mathbb{C}\) satisfying \(z\not\in f(u)\) is a domain of holomorphy in \(\mathbb{C}^ 2\). Let \(h_ d\) (resp. \(h_ c)\) denote the Hausdorff distance between subsets of a bounded domain \(D\) in \(\mathbb{C}\) defined in terms of the euclidean (resp. Carathéodory) distance \(d\) (resp. \(c\)): for \(H^ j\subset D (j=1,2)\), \[ h_ d(H^ 1,H^ 2)=\max \{\sup \{d(z,H^ 2):\;z\in H^ 1\}, \] \[ \sup \{d(H^ 1,z):\;z\in H^ 2\}\}, \] where \(d(z,H^ j)=d(H^ j,z)=\inf \{| z-u| =d(z,u);\quad u\in H^ j\}.\) Similarly, for \(d_ c(H^ 1,H^ 2)\), using \(c\) instead of \(d\). The author investigates the question whether the functions \(u\to h_ d(f(u),g(u))\) (resp. \(u\to h_ c(f(u),g(u)))\) is subharmonic on \(U\) for analytic set-valued functions \(f,g:U\to F_ c(\mathbb{C}),\) the question being known to be true for holomorphic functions f, g. The author gives a negative answer to this question by using a result of \textit{Z. Słodkowski} [Math. Ann. 256, 363-386 (1981; Zbl 0452.46028)]. This answer is contrary to a statement added in proof to the author's previous paper [Adv. Math. 47, 50-73 (1983; Zbl 0517.31006)].











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 Q3320664)