The usefulness and popularity of nonlinear models have spurred a large literature on data analysis, but research on design selection has not kept pace. One complication in studying optimal designs for nonlinear models is that information matrices and optimal designs depend on unknown parameters. Besides the popular locally optimal designs strategy, another common approach is to use Bayesian optimal design approach, which typically means an optimality problem has to be solved through numerical approaches. However, very few algorithm approaches are available for Bayesian optimal design. In this talk, I will first give two simple examples demonstrating the set up and the impact of efficient designs. Then I will review the latest advances. The main purpose is to introduce a general and efficient algorithm for deriving Bayesian designs. We prove convergence of the algorithm, and demonstrate in numerous examples that the new approach is very efficient.
More information on Min Yang may be found at http://www.math.uic.edu/people/profile?netid=myang2