Values and coalition configurations (Q2264100)

The authors study a new class of values for TU games \((N,v)\), called the \textit{weighted bounded configuration values}. These values \(\Phi^{\omega,{\mathcal B}}(v)\) are defined (in a special way) with the help of two parameters, \({\mathcal B}\) and \(\omega\), where \({\mathcal B} = \{{\mathcal B}_1,\ldots, {\mathcal B}_m\}\) is a family of different subsets of \(N\) such that \(\cup_{i=1}^m {\mathcal B}_i=N\) and \(\omega\) is a vector with positive components indexed by the pairs \((i,{\mathcal B}_q)\) with \(i\in N\) and \({\mathcal B}_q\in {\mathcal B}\) satisfying \(i\in {\mathcal B}_q\). Several axiomatizations of the value \(\Phi^{\omega,{\mathcal B}}(v)\) are given with the help of such standard properties like efficiency, linearity, partnership additivity and null player out. The weighted bounded configuration values generalize (in some way) the Shapley value, the Oven value and the weighted coalition structure values earlier studied by \textit{A. Levy} and \textit{R. P. McLean} [Games Econ. Behav. 1, No. 3, 234--249 (1989; Zbl 0753.90079)].