Optimal experimental designs when an independent variable is potentially censored

Abstract

This paper considers the problem of constructing optimal approximate designs when an independent variable might be censored. The problem is which design should be applied in practice to obtain the best approximate design when a censoring distribution is assumed known in advance. The approach for finite or continuous design spaces deserves different attention. In both cases, equivalent theorems and algorithms are provided in order to calculate optimal designs. Some examples illustrate this approach for D-optimality.

Keywords:

• approximate designs
• censoring distribution
• D-optimality
• information matrix

AMS 2000 Mathematics Subject Classification :

• 62K05
• 62N01

Acknowledgements

This paper has been written when the second author was doing a postdoc in the University of Salamanca and it was partially sponsored by the Ministerio de Educación y Ciencia and Fondos FEDER MTM2007 67211 C03-01, Junta de Comunidades de Castilla-La Mancha PAI07-0019-2036 and University of Castilla-La Mancha. The authors want to thank the referees for their helpful their comments.