运用Khuri(1985)组合响应的方法,将Wiens(1991)的结论推广至多响应情形.证明均匀设计在对多响应线性模型作显著性检验时,在模型失拟情况下使检验的功效函数最小值达到最大,在模型未失拟情况下使检验的功效函数最大值达到最小.
The paper extends the result of Wiens( 1991 ) to the multiresponse model. It is shown that the uniform design maximizes the minimum power of the standard test for lack of fit when the model is lack of fit and minimizes the maximum power when the model is true.