• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Sequential analysis >A Two-Stage Procedure for Selection and Assessment of Agreement of the Best Instrument with a Gold Standard
【24h】

A Two-Stage Procedure for Selection and Assessment of Agreement of the Best Instrument with a Gold Standard

机译：选择和评估最佳工具与黄金标准协议的两阶段程序

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

We consider a method-comparison study of assessing agreement between the gold standard and the unknown best among the k instruments under consideration. The goal is to identify the best when this agreement is close enough in terms of certain metric. We propose a two-stage procedure that employs the logarithm of the mean squared deviation as the metric for comparison and uses its large-sample properties. Our procedure has a prespecified significance level under the null hypothesis of insufficient agreement. Further, it correctly rejects the null hypothesis and selects the true best with probability no less than a prespecified level (in the asymptotic sense), when the best has sufficiently close agreement with the gold standard and is not practically equivalent to other competing instruments. For k = 2, we assume bivariate normality for the differences of paired measurements and study the small-sample properties of the procedure using simulation. We observe that a first-stage sample of size 15 is adequate for its application and illustrate the procedure using a real data set.
机译：我们考虑了一种方法比较研究，用于评估所考虑的k种工具中的黄金标准与未知最佳之间的一致性。目的是在某种度量标准上接近该协议时，确定最佳方案。我们提出了一个两阶段程序，该程序采用均方差的对数作为比较的指标，并使用其大样本属性。在协议不足的无效假设下，我们的程序具有预先指定的显着性水平。此外，当最佳假设与黄金标准足够接近并且实际上不等同于其他竞争工具时，它会正确地拒绝零假设并选择不小于预定水平的真实最佳（渐近意义上）。对于k = 2，我们假设配对测量值的差异为双变量正态性，并使用模拟研究程序的小样本属性。我们观察到大小为15的第一阶段样本足以满足其应用，并使用实际数据集说明了该过程。

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