The sequential generation of multiresponse d-optimal designs when the variance-covariance matrix is not known
- 1 January 1987
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 16 (1) , 239-259
- https://doi.org/10.1080/03610918708812589
Abstract
The construction of Doptimal designs for multiresponse experiments was consideredby Fedorov (1972, Ch. 5). His algorithm required that the variance-covariance matrix, £, of the responses be known. This is rarely the case in practice. The primary objective of this paper isto develop a sequential procedure for the construction of multiresponse designs when £ is not known. Several numerical examples are given to illustrate this procedureKeywords
This publication has 5 references indexed in Scilit:
- Optimal DesignPublished by Springer Nature ,1980
- A controlled random search procedure for global optimisationThe Computer Journal, 1977
- The Sequential Generation of $D$-Optimum Experimental DesignsThe Annals of Mathematical Statistics, 1970
- An Efficient Method of Estimating Seemingly Unrelated Regressions and Tests for Aggregation BiasJournal of the American Statistical Association, 1962
- The Equivalence of Two Extremum ProblemsCanadian Journal of Mathematics, 1960