Delta Approximation: Parameter Identification for Simulation Models of Human Behavior

摘要

Procedures are demonstrated for using data in the development of simulation models of long term human behavior with respect to environment and resources. The data is used to identify the unknown parameters in the hypothetical variable interactions. In this way, both the heuristic value to experts and the value to decision making of simulation modeling is enhanced. The technique is illustrated with an identification for a non-linear, discrete time simulation model (17th order) of the use of land and energy in California. This model has 80 parameters which were identified with 150 data points and a few minutes of computer time. Also included are: discussions of the uses and misuses of simulation models; a statement of the parameter identification problem; and several simple examples of the numerical solution of the problem.