缺失项二阶混料模型I最优设计  

I-optimal design of missing terms second-order mixture experiments

作　　者：朱小渊 郝红花 张崇岐 ZHU Xiao-yuan;HAO Hong-hua;ZHANG Chong-qi(School of Econ.and Sta.,Guangzhou Univ.,Guangzhou 510006,China)

机构地区：[1]广州大学经济与统计学院

出　　处：《高校应用数学学报（A辑）》2019年第4期379-388,共10页Applied Mathematics A Journal of Chinese Universities(Ser.A)

基　　金：国家自然科学基金(11671104; 11926305)

摘　　要：混料试验设计的应用热点已经从工农业转到医学领域,模型从通常的规范和中心多项式模型发展到适合应用的具有缺失项的模型,最优性准则中从参数估计准确出发的D最优研究较多,但是能准确预测响应面估计的I最优准则更适合研究混料设计,又因为需要复杂运算所以研究较少.现在利用MATLAB软件,在I最优准则下,对具有缺失项的二阶混料模型进行计算,区别于过去的线性优化,使用非线性优化,给出缺失项二阶混料模型的I最优设计重要的结论性定理.The application hotspot of mixture experiment design has shifted from industry and agriculture to medical field. The model has developed from normal form and centroid form to model with missing terms suitable for application. There are many D-optimum studies in optimality criteria for accurate parameter estimation, but the I-optimum criterion for accurate prediction of response surface estimation is more suitable for research. Mixture design, but because of the need for computer operations, so less research. By using MATLAB software and under the I-optimal criterion, the secondorder mixture model with missing terms is calculated, which is different from the linear optimization in the past. By using non-linear optimization, the conclusion theorem of the I-optimal design of the second-order mixture model with missing terms is given.

关 键 词：缺失项模型 最优设计 I最优 D最优 

分 类 号：O212.6[理学—概率论与数理统计]