A-Optimality Standardized Through the Coefficient of Variation

Many of the usual criteria for optimal experimental design do not take into account the different scale of the variance of the parameters. Dette (1997 Dette , H. ( 1997 ). Designing experiments with respect to “standardized” optimality criteria . J. J. Roy. Statist. Soc. Ser. B Statist. Methodol. 59 ( 1 ): 97 – 110 .Crossref] , Google Scholar]) provided a standardization which leads to designs with similar efficiencies for all of the parameters.

In this article, a new way of standardization through the coefficient of variation is given. This leads to designs useful when one of the parameters is expected to be very small. Thus, this criterion may be used if there is special interest in maximizing the power of some parameter tests. Locally standardized through the coefficient of variation A-optimal designs are computed for simple linear and quadratic regression.