Logarithmic convexity of fixed points of stochastic kernel operators (Q2424861)

In this article, the author demonstrates results on fixed points of stochastic kernel operators which are about logarithmic convexity. Those results are awaited to play a major future role in economics, e.g., for applications on strategic market games. The achieved results extend and generalize already given finite-dimensional results. The author succeeded to extend his results for cases of \(n\times n\) matrices. These results were inspired by application to strategic market games. In fact, along with Kakutani's fixed point theorem, they mean a crucial advancement for existence proof of Nash equilibria in an exchange economy within complete markets. Herewith, the new results can become a main ingredient of a future infinite-dimensional generalization of that economic model. The paper is organized as follows. In Section 1, the author recalls some basic definitions and facts, needed in the later proofs of the main results, given in Section 2. In Section 3, he applies his results on finite or infinite non-negative matrices on which weighted operators on sequence spaces are defined, and he explains how his results apply to strategic market games. Furthermore, the author works out possible applications to the Arrow-Debreu model, the open Leontief model and to the Google page-rank model of Internet usage. Here, it might be enough to restrict the results on (infinite-dimensional) non-negative matrices, however, kernel (integral) operators are very important and of central importance in many applications of differential and integro-differential equations, problems of physics (especially, thermodynamics), modern technologies, statistics, and economical models. Hence, the author preferred to offer his results in this broader mathematical setting.