A Heterogeneous Agent, Indivisible Labor Model Solved By Means Of Genetic Algorithms

In this paper, I develop an algorithm for solving heterogeneous agent dynamic models in which individual decision rules influence each other. This approach is more general than the other heterogeneous agent solution methods in that it only requires a well formed objective function. The key innovation of the algorithm is to parameterize the policy rules of the individuals and then to find the Nash equilibrium in policy rules by using genetic algorithms to evolve optimal policy responses. To illustrate the algorithm, I solve an indivisible labor model without full unemployment insurance that endogenously induces heterogeneous agents.