scientific article; zbMATH DE number 3849116

From MaRDI portal
Publication:3318669





[For the entire collection see Zbl 0519.00010.] Let S be a denumerable set, R the real line, \({\mathcal B}\) the Borel \(\sigma\)-algebra of R and t(s,x) a \({\mathcal B}\)-measurable functional in x for each \(s\in S\). Let \(\xi\) be a random variable on a given probability space. The authors introduce the following three optimality criteria: 1) \(\min_{s}t(s,E\xi)\); 2) \(\min_{s}Et(s,\xi)\); 3) \(\max_{s}P_ s\), where \(P_ s=P\{\omega:\quad \arg \quad \min \quad t(l,\xi(\omega))=s\}\) and \(\arg \quad \min \quad f(x)\) of the function f(x), \(x\in X\), is a point \(x^ 0\in X\) such the \(f(x^ 0)=\min_{x}f(x),\quad x\in X.\) The elements of S are called strategies. Denote by \(\gamma\), \(\mu\) and \(\pi\) the optimal strategies corresponding to the criteria 1), 2) and 3), respectively. The aim of the present paper is to study the relations between the strategies \(\gamma\), \(\mu\) and \(\pi\). The authors prove a few statements showing that under some conditions are optimal strategies coincide, and under other conditions, one of the strategies dominates, in an appropriate sense, the other strategies. Useful remarks and illustrative examples are given, too.











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 Q3318669)