Living with the Curse of Dimensionality : Closed-Loop Optimization in a Large-Scale Fisheries Simulation Model

Facilitated by remarkable increases in computational speed, simulation models are becoming more and more complex and are being increasingly used in applied economic analysis. However, computational limitations remain a major barrier to the study of dynamically optimal policies. We study the problem of carrying out dynamic optimization in conjunction with large simulation models and propose a method for working around the computational difficulties that arise in such problems. Our methods are applied to a model of the Gulf of Mexico's red snapper fishery to study the dynamically optimal total allowable catch