LocallyD-optimal designs for heteroscedastic polynomial measurement error models
LocallyD-optimal designs for heteroscedastic polynomial measurement error models
复制标题
异方差多项式测量误差模型的局部D最优设计
DOI：
10.1007/s00184-019-00745-2
复制
发表时间：
2020
期刊：
影响因子：
0.7
通讯作者：
Yue Rong-Xian
中科院分区：
文献类型：
--
作者：
Zhang Min-Jue;Yue Rong-Xian
本文考虑异方差多项式测量误差模型的最优设计的构造。利用修正得分函数法建立了相应的近似设计理论，从而导出了非凹优化问题。对于具有常用异方差结构的加权多项式测量误差模型，局部D-最优设计的支撑点个数的上界可以显式确定.最后给出了一个数值例子来说明异方差结构对最优设计的影响。
This paper considers constructions of optimal designs for heteroscedastic polynomial measurement error models. Corresponding approximate design theory is developed by using corrected score function approach, which leads to non-concave optimisation problems. For the weighted polynomial measurement error model of degreepwith some commonly used heteroscedastic structures, the upper bounds for the number of support points of locallyD-optimal designs can be determined explicitly. A numerical example is given to show how heteroscedastic structures affect the optimal designs.