Optimal Timing of Decisions: A General Theory Based on Continuation Values
We develop a comprehensive theory of optimal timing of decisions based around continuation values as functions of the state and operators that act on them. Rewards can be bounded or unbounded. One advantage of this approach over standard Bellman equation methods is that continuation value functions are smoother than value functions. Another is that, for a range of problems, the continuation value function exists in a lower dimensional space than the value function. We exploit these advantages to obtain new results and more efficient computation.