Asymptotic Properties of Adaptive Penalized Optimal Designs over a Finite Space
Adaptive optimal design with a cost constraint is considered, both for LS estimation in nonlinear regression and ML estimation in Bernoulli-type experiments, with possible applications in clinical trials. We obtain the strong consistency of the estimators for designs over a finite space, both when the cost level is fixed (and the adaptive design converges to an optimum constrained design) and when the objective is to minimize the cost. Moreover, the asymptotic normality of the estimators is obtained in the first situation, with an asymptotic covariance matrix given by the inverse of the usual information matrix, calculated as if the design were not constructed sequentially.
KeywordsOptimal Design Asymptotic Property Design Point Asymptotic Normality Strong Consistency
Unable to display preview. Download preview PDF.
- Müller, W. and B. Pötscher (1992). Batch sequential design for a nonlinear estimation problem. In V. Fedorov, W. Müller, and I. Vuchkov (Eds.), Model-Oriented Data Analysis II, Proceedings 2nd IIASA Workshop, St Kyrik (Bulgaria), May 1990, pp. 77–87. Heidelberg: Physica Verlag.Google Scholar
- Pronzato, L. (2009b). One-step ahead adaptiveD-optimal design on a finite design space is asymptotically optimal. Metrika. (to appear, DOI: 10.1007/s00184-008-0227-y).Google Scholar