二水平正交饱和设计的统计分析：零效应搜索法  

Statistical analysis of 2-level orthogonal saturated designs: the procedure of searching zero effects

作　　者：张晓琴[1] 张应山[1] 茆诗松[1] 

机构地区：[1]华东师范大学统计系

出　　处：《华东师范大学学报（自然科学版）》2008年第1期51-59,共9页Journal of East China Normal University(Natural Science)

摘　　要：分析正交饱和设计的困难在于,由于没有剩余的自由度用于误差方差的估计,所以一般的方差分析方法不再适用.基于次序统计量期望的一个相关性质,本文提出了一种确定零效应个数的方法一零效应搜索法.这样就能够给出误差方差的估计,进而确定模型的显著因子或显著效应.The difficulty of analyzing orthogonal saturated designs is that there are no degrees of freedom to estimate the error variance, so that the ordinary ANOVA is no more available. The paper proposes a new method to identify the number of the effect parameters which is equal to zero, namely the Procedure of Searching Zero Effects. This proposed method is based on the mean of the order statistics. Consequently, the error variance is estimated and the active effects are identified

关 键 词：正交饱和设计 次序统计量 零效应 显著效应 

分 类 号：O212.1[理学—概率论与数理统计]