当前位置: 首页 > 期刊 > 《第四军医大学学报》 > 2001年第7期
编号:10857376
医学重复观测数据裂区方差分析的假定条件的检验
http://www.100md.com 《第四军医大学学报》 2001年第7期
重复观测,,重复观测;假设检验;方差分析,0引言,1原理和方法,2实例分析,3讨论,参考文献
     关键词: 重复观测;假设检验;方差分析

    摘 要:目的 重复观测数据存在自相关性常导致一元方差分析的误用,探讨一元方差分析两个假定条件的检验. 方法 通过标准正交对比变换克服数据间的自相关性,应用似然比统计量进行假定条件的检验. 结果 给出了重复观测数据一元方差分析的假定条件检验方法,并用软件REP得以实现. 结论 只有满足假定条件,才可以保证重复观测资料单变量方差分析的有效性.

    Tests for presuppositions concerning validity of split┐plot analysis of variance with repeated mea┐sures

    CHEN Chang-Sheng,XU Yong-Yong,ZHAO Dong-Tao

    1 Department of Health Statistics,

    2 Department of Mathematics,Faculty of Biomedical Engineering,Faculty of Preventive Medicine,Fourth Military Medical University,Xi'an710033,China

    Keywords:repeated measures;hypothesis test;analysis of variance;

    Abstract:AIM To explore the tests for two presuppositions concerning the validity of the univariate analysis of variance with repeated measures.METHODS The authors intro-duced orthonormal contrast transformation to avoid auto-cor-relation,and proposed the tests of presuppositions by likeli-hood ratio statistics.RESULTS Hypothesis tests for the presuppositions of the univariate ANOVA with repeated mea-sures were given ......

您现在查看是摘要页,全文长 7845 字符