المستخلص: |
In this paper the logarithms of the treatment parameters in a mixture paired comparison experiments, are approximated by a second-order response surface model in the transformed mathematically independent variables. The null hypothesis that the regression coefficient of W21, say ξ11 equals zero is considered. A class of partially balanced designs D2, corresponding to the null hypothesis is constructed. The simplified form of the non-centrality parameter λ2 of the asymptotic chi-square distribution of the likelihood ratio statistic is derived for testing the corresponding null hypothesis under defined local alternatives. the optimal design d2, maximizing, the no-centrality parameter λ2 in the class of designs D2 is determined
|