Some existence theorems for functional equations arising in dynamic programming. II (Q1114599)

[For part I by the first author and \textit{S. Mitra} see ibid. 98, 348-362 (1984; Zbl 0533.90091).] The authors continue the study of existence theorems for functional equations. In the beginning two fixed point theorems are proved and these are then used to establish various existence theorems for functional equations. Let X and Y be Banach spaces. Let \(S\subset X\) be the state space, \(D\subset Y\) be the decision space and E denote the set of all real-valued mappings on S which are bounded on bounded subsets of S. Let G: \(S\times D\times E\to R\), \(h_ i:S\times D\to R\), \(T_ i:S\times D\to S\) \((i=1,...,N)\) and g: \(S\times D\to R\). Then the authors prove the existence of unique solutions to various functional equations under various assumptions e.g., (i) under certain assumptions (see the paper for details) the functional equation \[ {\mathcal J}(x)=\inf_{y\in D}G(x,y,{\mathcal J}) \] possesses a unique solution which is bounded on bounded subsets of S; (ii) under certain assumptions (see the paper for details) the functional equation \[ {\mathcal J}(x)=_{y\in D}[g(x,y)+\sum^{N}_{i=1}h_ i(x,y){\mathcal J}(T_ i(x,y)]\quad (x\in S) \] possesses a unique solution which is bounded on bounded subsets of S.