首页 | 本学科首页   官方微博 | 高级检索  
     检索      

���ڿ�ζ�Ƚϵļ���
引用本文:刘莉,茆诗松.���ڿ�ζ�Ƚϵļ���[J].应用概率统计,2006,22(1):21-26.
作者姓名:刘莉  茆诗松
作者单位:??????????????, ???, 200062?????????????, ?人, 430062
摘    要:本文用Bayes方法来检验两种药品之间的口味是否相同, 得到了与人们的直观感觉相吻合的结论\bd 设$(x_1,x_2,x_3)$服从参数为$(p_1,p_2,p_3)$的多项分布, 则检验口味的问题转为检验$H_i:(p_1,p_2,p_3)$ $\in\Theta_ii=1,2,3$三个假设的问题. 在使用Bayes方法时, 其关键是选择合理的先验分布, 这里选择相应的先验分布为无信息先验, 据此求出每个假设的后验概率, 从而得到结论\bd

关 键 词:Bayes????  ????  ??????  ????  
收稿时间:2003-11-26
修稿时间:2003年11月26

Testing on Flavor Analysis
LIU LI,MAO SHISONG.Testing on Flavor Analysis[J].Chinese Journal of Applied Probability and Statisties,2006,22(1):21-26.
Authors:LIU LI  MAO SHISONG
Institution:Department of Statistics, East China Normal University, Shanghai, 200062Institute of Mathematics and Computer Science, Hubei University, Wuhan, 430062
Abstract:In testing whether the flavors of two samples are the same by applying Bayesian methods, the paper draws the conclusion that it is identical with human intuition. Suppose that x_1,x_2,x_3) is distributed as polynomial distribution with parameters $(p_1,p_2,p_3)$, then the question of flavor testing is changed to test three hypotheses, $H_i:(p_1, p_2, p_3)in\Theta_i$ $i=1,2,3. In applying the Bayesian methods, the key lies in the choice of proper prior distribution, and herein noninformation prior is selected. Accordingly, the posterior probabilities of the above hypotheses are derived, and the conclusion is reached.
Keywords:
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《应用概率统计》浏览原始摘要信息
点击此处可从《应用概率统计》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号