※ [本文转录自 NCTU-STAT96G 看板] 作者: pael (就是喜欢) 看板: NCTU-STAT96G 标题: 0806演讲公告@清大统研所 时间: Wed Jul 23 09:33:49 2008 清华大学、交通大学 统 计 学 研 究 所 专 题 演 讲 题 目: Componentwise Variance Dispersion Graphs for Mixture Experiments 主讲人: Prof. John J. Borkowski Department of Mathematical Sciences Montana State University 时 间: 97年8月6日（星期三）10：40 - 11：30 （上午10：20-10：40茶会於统计所821室举行） 地 点: 清大综合三馆837室 Abstract Much research has been devoted to graphical methods for comparison of prediction properties of response surface (RS) designs for experiments in spherical and cuboidal design spaces (Borkowski 2006). In these experiments, the design factor settings can be varied independently. Consider, however, a q-component mixture experiment constraints defined by lower and upper bounds ( for component proportion (i = 1, 2, …, q). For a mixture experiment, there is also the additional constraint that the sum of component proportions must equal 1 (). Whether the design region for a mixture experiment is a simplex or some irregularly shaped subregion of the simplex defined by these constraints, the problem of interest addressed in this research is the extension of variance dispersion graphs (VDGs) to experiments involving mixtures. As an alternative to studing the prediction variance on shrunken polyhedral spaces (Piepel and Anderson 1993), a new graphical tool, called the Componentwise variance dispersion graph (CVDG), will be proposed. 敬请公布 欢迎参加 --

※ 发信站: 批踢踢实业坊(ptt.cc) ◆ From: 140.113.114.203 --

※ 发信站: 批踢踢实业坊(ptt.cc) ◆ From: 140.113.114.222