A Note on the Relationship between Two Approaches to Optimal Design under Correlation
The note demonstrates the relationship between two recently developed methods for characterizing optimal designs, when the errors/observations in the experiments are correlated according to a given correlation structure. The understanding of this relationship can help to improve the applicability of the methods by providing new frameworks for their tuning parameters.
KeywordsOptimal Design Information Matrix Design Measure Correlate Error Information Matrice
Unable to display preview. Download preview PDF.
We are grateful to two anonymous referees, whose comments helped to improve our paper, particularly our English.
- Fedorov, V. (1996). Design of spatial experiments: Model fitting and prediction. In S. Gosh and C. R. Rao (Eds.), Handbook of Statistics 13, 515–553. Amsterdam: Elsevier.Google Scholar
- Möller, W. G. (2007). Collecting Spatial Data: Optimum Design of Experiments for Random Fields, 3rd ed. Heidelberg: Springer-Verlag.Google Scholar
- Möller, W. G. and L. Pronzato (2009). Towards an optimal design equivalence theorem for random fields? (Submitted).Google Scholar
- Pázman, A. and W. G. Möller (1998). A new interpretation of design measures. In A. C. Atkinson, L. Pronzato, and H. P. Wynn (Eds.), MODA 5 - Advances in Model-Oriented Data Analysis and Experimental Design, 239–246. Heidelberg: Physica-Verlag.Google Scholar
- Rao, C. R. (2001). Linear Statistical Inference and its Applications, 2nd ed. New York: Wiley.Google Scholar