...
首页> 外文期刊>Quality Control and Applied Statistics >Hypothesis testing about proportions in two finite populations
【24h】

Hypothesis testing about proportions in two finite populations

机译:关于两个有限总体中的比例的假设检验

获取原文
获取原文并翻译 | 示例
           

摘要

The paper explores two tests on equality of proportions in two finite populations: The Z-test (based on the usual normal approximation). The approximate E-test. (See the Results below for further explanations on both tests). Computing size and power of the tests numerically for various arbitrarily selected sample sizes and population proportions, the paper concludes that the approximate E-test is: Almost exact since the sizes of the test seldom exceed the nominal level by a significant amount. Satisfactory even for small sample sizes and can be recommended for practical use. The paper comments that the necessary programs for computing p-values and powers of the approximate E-test are provided at the Web page http://lib.stat.cmu.edu/TAS/, and the prospective users need not face any practical difficulties in using the E-test. (15 refs.)
机译:本文探讨了两个关于两个有限总体中的比例相等性的检验:Z检验(基于通常的法线近似)。近似的E检验。 (有关两种测试的进一步说明,请参见下面的结果)。通过对各种任意选择的样本数量和人口比例进行数字计算,计算出样本的大小和功效,得出的结论是,近似的E检验为:几乎是精确的,因为样本的数量很少超过标称水平。即使是小样本量也令人满意,可以建议实际使用。该论文评论说,在网页http://lib.stat.cmu.edu/TAS/上提供了用于计算p值和近似E检验的幂的必要程序,并且潜在的用户无需面对任何实际操作。使用电子考试的困难。 (15个参考)

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号