Optimal Design for Compositional Data
In this paper optimal designs for experiments involving compositional data, specifically locally D-optimal designs for the additive logistic normal model and locally D S -optimal designs for Dirichlet regression, are investigated. The theory underpinning the construction of these designs is based on the appropriate information matrices and the development, while new, is relatively straightforward. The ideas are illustrated by means of a simple example, that of two consecutive reactions.
KeywordsOptimal Design Design Space Information Matrix Compositional Data Consecutive Reaction
Unable to display preview. Download preview PDF.
Linda Haines would like to thank the University of Cape Town, the National Research Foundation and SASOL, South Africa, for financial support.
- Aitchison, J. (2004). The Statistical Analysis of Compositional Data. New Jersey: Blackburn Press.Google Scholar
- Campbell, G. and J. E. Mosimann (1987). Multivariate analysis of size and shape: modelling with the dirichlet distribution. ASA Proceedings of Section on Statistical Graphics, 93–101.Google Scholar
- Fedorov, V. V. (1972). Theory of Optimum Experiments. New York: Academic Press.Google Scholar
- Hijazi, R. H. (2006). Residuals and diagnostics in dirichlet regression. ASA Proceedings of the General Methodology Section, 1190–1196.Google Scholar
- Hijazi, R. H. and R. W. Jernigan (2007). Modelling compositional data using dirichlet regression models. Technical report, Preprint.Google Scholar