ECONOMICAL EXPERIMENTS - BAYESIAN EFFICIENT EXPERIMENTAL-DESIGN

We propose and implement a Bayesian optimal design procedure. Our proc edure takes as its primitives a class of parametric models of strategi c behavior, a class of games (experimental designs), and priors on the behavioral parameters. We select the experimental design that maximiz es the information from the experiment. We sequentially sample with th e given design and models until only one of the models has viable post erior odds. A model which has low posterior odds in a small collection of models will have an even lower posterior odds when compared to a l arger class, and hence we can dismiss it. The procedure can be used se quentially by introducing new models and comparing them to the models that survived earlier rounds of experiments. The emphasis is not on ru nning as many experiments as possible, but rather on choosing experime ntal designs to distinguish between models in the shortest possible ti me period. We illustrate this procedure with a simple experimental gam e with one-sided incomplete information.