D-and A-optimal designs for mixture experiments with multiresponse models
D-and A-optimal designs for mixture experiments with multiresponse models
复制标题
DOI：
--
复制
发表时间：
2008
期刊：
影响因子：
--
通讯作者：
Yue Rong-xian
中科院分区：
文献类型：
--
作者：
Yue Rong-xian
考虑多响应模型E（Yi（X））=ηi（X）= fTi（X）βi（i= 1，2），其中η1和η2分别为Scheffe线性模型和二次模型.本文研究了D-和A-最优设计。更确切地说，我们将自己限制在一个特殊类型的设计中，其支撑点是X <$（1，0，...，0）和X <$（1/2，1/2，0，...，0），并根据D-和A-最优设计准则获得权重r*1和r*2。对这些最优设计效率的数值结果表明，在某些混料试验中，D-最优设计和A-最优设计的效率相差不大。
The multiresponse models E(Yi(X))=ηi(X)=f Ti(X)βi (i=1,2) are considered,where η1 and η2 are the Scheffe′s linear model and quadratic model, respectively. D- and A-optimal designs are investigated in this paper. More precisely, we restrict ourselves to a particular type of design ξ* whose support points are X(1, 0,…, 0) and X(1/2, 1/2, 0,…, 0), and obtain the weights r*1 and r*2 according to the D- and the A-optimal design criteria. The numerical results about the efficiencies of these optimal designs show that D- and A-optimal designs do not differ too much in some mixture experiments.