摘要
本文指出，在混合位级的正交试验设计中，以往的极差分析方法存在着不足。本文给出了极差的修正方法，并用大量实例做了数值验证。
In this paper,we indicate that there has been existed the defect on method of range analysis in orthogonal experimental design of mixed levels,and we give a simple method to correct it.A large number of examples are used to test this work and the result shows this correction is very effective indeed.
出处
《数理统计与管理》
CSSCI
北大核心
1998年第6期31-37,共7页
Journal of Applied Statistics and Management
关键词
试验设计
正交设计
位级
极差分析
平均极差
Experimental Design
Orthogonal Design
Level
Range Analysis
Resolution of Sum of Squares.
