Optimality criteria for regression models based on predicted variance
Optimality criteria for regression models based on predicted variance
复制标题
DOI：
10.1093/biomet/86.1.93
复制
发表时间：
1999-03-01
期刊：
影响因子：
2.7
通讯作者：
O'Brien, TE
中科院分区：
文献类型：
--
作者：
Dette, H;O'Brien, TE
在非线性回归模型的背景下，一类新的最优设计准则的开发和说明。这个新的类，称为I-L-最优性，类似于基弗的Phi（k）-准则，但基于预测方差，而基弗的类是基于信息矩阵的特征值; I-L-最优设计是不变的，相对于不同的参数化模型，并包含G-和D-最优性作为特殊情况。我们给出了一个一般的等价定理，用于获得和验证I-L-最优设计。该方法通过各种示例来说明。
In the context of nonlinear regression models, a new class of optimum design criteria is developed and illustrated. This new class, termed I-L-optimality, is analogous to Kiefer's Phi(k)-criterion but is based on predicted variance, whereas Kiefer's class is based on the eigenvalues of the information matrix; I-L-optimal designs are invariant with respect to different parameterisations of the model and contain G- and D-optimality as special cases. We provide a general equivalence theorem which is used to obtain and verify I-L-optimal designs. The method is illustrated by various examples.