...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Quality Control and Applied Statistics >Confidence intervals for causal effects with invalid instruments by using two-stage hard thresholding with voting
【24h】

Confidence intervals for causal effects with invalid instruments by using two-stage hard thresholding with voting

机译：通过使用投票的两阶段硬阈值与仪器的因果效应的置信区间

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

Instrumental variable (IV) analysis is used to find instruments that are valid, or have no direct effect on the outcome and are ignorable. Since one is unsure about the presence of only valid instruments, a general inference procedure called two-stage hard thresholding with voting, is proposed in the presence of invalid IVs. By generating candidate sets of valid IVs, this article determines valid confidence intervals for the causal effect that has the oracle optimal width. Two major cases are considered: the number of covariates and instruments is small and fixed relative to the sample size and the number of covariates and/or instruments is growing and may exceed the sample size. After presenting a theoretical justification for the proposed procedure, the performance is evaluated using a simulation study and results are compared with the existing methods. From the results, the proposed procedure outperforms the traditional and recent methods in the presence of invalid IVs. The method is also applied to reanalyze the causal effect of education on earnings.
机译：工具变量（IV）分析用于寻找有效的工具，或对结果没有直接影响且可忽略的工具。由于不确定是否只存在有效工具，因此在存在无效IVs的情况下，提出了一种称为两阶段投票硬阈值的通用推理程序。通过生成有效IVs的候选集，本文确定了具有oracle最佳宽度的因果效应的有效置信区间。考虑了两种主要情况：相对于样本量，协变量和工具的数量较小且固定，并且协变量和/或工具的数量正在增加，并且可能超过样本量。在给出了该方法的理论依据后，通过仿真研究对其性能进行了评估，并将结果与现有方法进行了比较。从结果来看，在存在无效IVs的情况下，所提出的方法优于传统和最近的方法。该方法也被用于重新分析教育对收入的因果关系。

著录项

相似文献

  • 外文文献
  • 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