于泽洋,刘媛媛,李长平,胡良平.如何正确运用t检验——两几何均值比较一般差异性t检验及SAS实现[J].四川精神卫生杂志,2020,33(3):222-225.
如何正确运用t检验——两几何均值比较一般差异性t检验及SAS实现
How to use t test correctly——comparison of two geometric means with general difference t test and SAS implementation
投稿时间:2020-05-26  
DOI:10.11886/scjsws20200526005
中文关键词:  几何均数  对数正态分布  变量变换  t检验  SAS软件
英文关键词:Geometric mean  Logarithmic normal distribution  Variable transformation  t test  SAS software
基金项目:国家自然科学基金项目(项目名称:贝叶斯生存分析方法在肝细胞癌肝移植患者预后预测中的应用研究,项目编号:81803333)
作者单位邮编
于泽洋 天津医科大学公共卫生学院天津 300070 300070
刘媛媛 天津医科大学公共卫生学院天津 300070 300070
李长平 天津医科大学公共卫生学院天津 300070
世界中医药学会联合会临床科研统计学专业委员会北京 100029 
100029
胡良平 世界中医药学会联合会临床科研统计学专业委员会北京 100029
军事科学院研究生院北京 100850 
100850
摘要点击次数:
全文下载次数:
中文摘要:
      本文目的是介绍几何均数以及两组近似对数正态分布数据几何均数的一般差异性t检验及SAS实现。首先对几何均数的概念、适用条件和计算公式进行介绍;并且将几何均数与统计学中常用的算术均数的特点进行比较;其次,呈现两组药物浓度数据,该数据近似服从对数正态分布,符合几何均数的适用条件,使用SAS的TTEST过程对两组几何均数的一般差异性进行t检验,并对相关结果进行解释和比较;最后,讨论几何均数使用时的注意事项。
英文摘要:
      The purpose of this article was to introduce the geometric mean, as well as the SAS implementation of general difference t test for logarithmic normal distribution data. Firstly, the concept of geometric mean, applicable conditions and calculation formulas were introduced. And the characteristics of geometric means were compared with those of arithmetic means commonly used in statistics. Secondly, two groups of concentration data were presented, which were logarithmic normal distribution approximately and met the applicable conditions of the geometric mean. The TTEST procedure of SAS was used to perform a t test on the general difference of the geometric mean. The relevant results were explained and compared. Finally, the precautions when using geometric mean were discussed.
      附件
查看全文  HTML  查看/发表评论  下载PDF阅读器
关闭
手机扫一扫