期刊文献+

对假设检验使用中一个常见错误的澄清 被引量:2

Clarifying a Common Mistake in Using Hypothesis-test
下载PDF
导出
摘要 是否接受某个假设应该取决于这个假设成立的概率.当从实验获得样本观察值X之后,假设H成立的概率就是P(H/X).所以,应该把P(H/X)≥e(e是一个足够大的概率常数)规定为接受假设H的条件并进而规定为确定H接受域的方法.这个方法对备择假设和零假设是同样适用的,据此可以证明:零假设的接受域不是备择假设接受域的余集. Whether we will accept a hypothesis should be determined by the probability that the hypothesis is right. When we obtain the observed value X of a sample from an experiment, the probability that hypothesis H is right is P(HIX). Therefore, the qualification for receiving H and thereby the method for prescribing the receiving area for H should be P(HIX)≥e (e is a large enough probability constant). The method is true not only for alternative hypothesis but also for null hypothesis, so it can be proved that the receiving area for null hypothesis is not the surplus set of the receiving area for alternative hypothesis.
作者 仲晓波
出处 《温州大学学报(自然科学版)》 2008年第1期37-42,共6页 Journal of Wenzhou University(Natural Science Edition)
关键词 假设检验 备择假设 零假设 Hypothesis-test Alternative hypothesis Null hypothesis
  • 相关文献

参考文献3

  • 1[1]Raymond S N.Null Hypothesis Significance Testing:A Review of an Old and Continuing Controversy[J].Psychological Methods,2000,5(2):241-301.
  • 2韩兆洲,魏章进.假设检验的一个常见误区[J].统计与信息论坛,2005,20(1):9-11. 被引量:10
  • 3[5]Falk R.In criticism of the null hypothesis statistical test[J].American Psychologist,1998,53:787-799.

二级参考文献1

  • 1茆诗松,周继芗.概率论与数理统计[M]. 北京:中国统计出版社,1999.

共引文献9

同被引文献4

引证文献2

二级引证文献6

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部