Equivalence of decentralized stochastic dynamic decision systems via Girsanov's measure transformation

In this paper we present two methods for decentralized optimization of continuous and discrete-time stochastic dynamic decision systems, with multiple decision makers having nonclassical information structures. For both methods we apply Girsanov's change of measure to transform such problems into equivalent optimization problems, under a reference probability measure, with corresponding observations and information structures available for decisions, which are not affected by any of the decisions.