...
首页> 外文期刊>Quality Control and Applied Statistics >Testing multiple dispersion effects in unreplicated fractional factorial designs
【24h】

Testing multiple dispersion effects in unreplicated fractional factorial designs

机译:在未复制的分数阶乘设计中测试多重色散效应

获取原文
获取原文并翻译 | 示例
   

获取外文期刊封面封底 >>

       

摘要

The article shows that two active dispersion effects may create a serious dispersion effect in their interaction column,and the existing methods for dispersion-effect testing in unreplicated 2~k-p are subject to those spurious effects. The exact dispersion-effect test (D~BH test),developed by Bergman and Hynen (Ref. 1),is biased in the presence of multiple dispersion effects. The authors illustrate by conducting simulation studies that the F~ML test has many desirable properties lacking in other tests.If two active dispersion effects occur in columns that are not in the adapted location model,F~ML is biased for their interaction column only but no other.Also,unidentified location effects though create bias for both D~BH and F~ML tests,F~ML seems less susceptible.(23 refs.)
机译:该文章表明,两个主动色散效应可能会在它们的相互作用栏中产生严重的色散效应,而现有的未复制2〜k-p色散效应测试方法会受到这些伪效应的影响。 Bergman和Hynen(参考文献1)开发的精确色散效应测试(D〜BH测试)在存在多重色散效应的情况下存在偏差。作者通过仿真研究表明,F〜ML测试具有许多其他测试所欠缺的理想特性。如果在非自适应位置模型中的色谱柱中出现两个主动色散效应,则F〜ML仅因其相互作用色谱柱而偏此外,尽管不确定的位置效应虽然会为D〜BH和F〜ML测试造成偏差,但F〜ML似乎不那么敏感(23个参考文献)。

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