Optimal experimental control in econometrics: the simultaneous equation problem
Panagiotis A. Papakyriazis · International Journal of Systems Science · 1985
An effective controlled economic experiment is an extremely expensive undertaking. This suggests that efficient design is crucial, and there is need for economists to extend design theory to handle peculiarities of economic experimentation. In this paper, we investigate the class of experimental control strategies that are optimal for estimation in the context of a simultaneous equation model where attention is limited to / admissible sample points which have been chosen by the experimenter so that the relevant region of the design space is adequately covered. Since the nonlinear restrictions of reduced-form coefficients implied by the structural form cause the design criteria to depend on unknown parameters, the experimental control problem is formulated in an adaptive framework. This approach is shown to resolve the conceptual difficulties inherent in alternative formulations. The experimental control problem considered in this paper can also be treated as an initial phase of a stochastic control problem. This will avoid solution to a difficult dual-control problem. Finally, an example is presented illustrating the improvement in estimation accuracy that can be obtained.