ABSTRACT
This chapter formalizes the policy problem as a stochastic dynamic program with learning. This formulation aids understanding of the problem, but it is mathematically intractable. Subsequently, I present a simplified decision framework that is tractable and provides useful guidance.
