常用正交表的构造原理及SAS实现 |
| |
引用本文: | 陈远方,林曦晨,徐利华,汤洪秀,汪宏晶,尹平. 常用正交表的构造原理及SAS实现[J]. 中国卫生统计, 2012, 29(4): 470-474 |
| |
作者姓名: | 陈远方 林曦晨 徐利华 汤洪秀 汪宏晶 尹平 |
| |
作者单位: | 华中科技大学同济医学院公共卫生学院流行病与卫生统计学系 430030 |
| |
摘 要: | 目的介绍常用正交表的构造原理,并应用SAS宏语句编制通用程序模块生成各种常用的正交表。方法应用哈达玛矩阵法、正交拉丁方法和并列法构造不同类型的常用正交表。结果采用哈达玛矩阵直积法、特征函数法、正交拉丁方法和并列法可以分别构造试验次数N=2s型、N≠2s且N=2(p+1)型(p为素数且p≡1(mod4))的2水平正交表、Lt2(tm)型正交表和混合型正交表。结论不同类型的正交表构造原理方法不同,应用SAS软件可生成各种常用的正交表,方便实用。
|
关 键 词: | 正交表 原理 SAS实现 |
Structure Principle of Orthogonal Array and Implementation of SAS Software |
| |
Affiliation: | Chen Yuanfang,Lin Xichen,Xu Lihua,et al.Department of Epidemiology and Health Statistics,School of Public Health,Tongji Medical College of Huazhong University of Science and Technology(430030),Wuhan |
| |
Abstract: | Objective The structuresprincipleof commonorthogonal arrays were analyzed,and their common computer models were implemented by macro statements of SAS softwale.Methods Construction of Harlamard Matrix,Design of Orthogonal Latin Squares and Parallel principle were used to construct different kinds of orthogonal arrays.Results The methods of Kronecker product Harlamard Matrix,Eigenvalue Function,orthogonal Latin squares and Parallel principle constructed the model of N = 2s of second level orthogonal arrays,N≠2s and N = 2(p + 1)(p is prime number and p≡1(mod4)) of second level orthogonal arrays,Lt2(tm) of orthogonal arrays,and Mixed orthogonal arrays respectively.Conclusion Different kinds of orthogonal arrays have different structure principle.Common program modules of SAS software were used convenient,simple and practical to construct different kinds of orthogonal arrays. |
| |
Keywords: | Orthogonal array Principle Implementation of SAS software |
本文献已被 CNKI 万方数据 等数据库收录! |
|