Causal inference with invalid instruments: post-selection problems and a solution using searching and sampling

Causal inference with invalid instruments: post-selection problems and a solution using searching and sampling
复制标题

使用无效仪器进行因果推断:选择后问题以​​及使用搜索和采样的解决方案

DOI:
10.1093/jrsssb/qkad049
复制
发表时间:
2023
期刊:
Journal of the Royal Statistical Society Series B: Statistical Methodology
影响因子:
--
通讯作者:
Guo, Zijian
Guo, Zijian
中科院分区:
--
文献类型:
--
作者:
Guo, Zijian

文献摘要

参考文献

被引文献

相似文献

在观察性研究中,工具变量方法是处理未测量混杂因素的最常用的因果推断方法之一。无效工具的存在是实际应用中的主要问题,而一个快速增长的研究领域是对可能无效工具的因果关系的推断。本文说明了当有效工具和无效工具很难以数据依赖的方式分离时,现有的可信区间可能会掩盖这一点。为了解决这个问题,我们构造了对区分有效和无效工具的错误具有健壮性的一致有效的置信度区间。我们建议寻找一系列的治疗效应值,以产生足够多的有效器械。我们进一步设计了一种新的采样方法,该方法与搜索一起导致了更精确的置信度区间。我们提出的搜索和抽样可信区间是一致有效的,并且在有限样本多数和复数规则下达到了参数长度。我们应用我们的建议来研究教育对收入的影响。所提出的方法在CRAN提供的R包RobustIV中实现。
Instrumental variable methods are among the most commonly used causal inference approaches to deal with unmeasured confounders in observational studies. The presence of invalid instruments is the primary concern for practical applications, and a fast-growing area of research is inference for the causal effect with possibly invalid instruments. This paper illustrates that the existing confidence intervals may undercover when the valid and invalid instruments are hard to separate in a data-dependent way. To address this, we construct uniformly valid confidence intervals that are robust to the mistakes in separating valid and invalid instruments. We propose to search for a range of treatment effect values that lead to sufficiently many valid instruments. We further devise a novel sampling method, which, together with searching, leads to a more precise confidence interval. Our proposed searching and sampling confidence intervals are uniformly valid and achieve the parametric length under the finite-sample majority and plurality rules. We apply our proposal to examine the effect of education on earnings. The proposed method is implemented in the R package RobustIV available from CRAN.
DOI: 10.1016/j.jeconom.2018.07.002
发表时间: 2016
影响因子: 6.3
作者:
Zijian Guo;Hyunseung Kang;T. T. Cai;Dylan S. Small
通讯作者: Dylan S. Small
使用一些无效工具变量进行因果推断:准贝叶斯方法*
DOI: --
发表时间: 2022
影响因子: 2.5
作者:
Gyuhyeong Goh;Jisang Yu
通讯作者: Jisang Yu
DOI: 10.1111/biom.13415
发表时间: 2020-12-17
期刊: BIOMETRICS
影响因子: 1.9
作者:
Kang, Hyunseung;Lee, Youjin;Small, Dylan S.
通讯作者: Small, Dylan S.
DOI: 10.1111/rssb.12275
发表时间: 2018-09-01
影响因子: 5.8
作者:
Guo, Zijian;Kang, Hyunseung;Small, Dylan S.
通讯作者: Small, Dylan S.
使用机器学习进行两阶段曲率识别:使用可能无效的工具变量进行因果推理
DOI: --
发表时间: 2022
期刊:
影响因子: --
作者:
Zijian Guo;P. Bühlmann
通讯作者: P. Bühlmann