Wang Linbo, Robins James M, Richardson Thomas S
Department of Biostatistics, Harvard School of Public Health, 677 Huntington Avenue, Boston, Massachusetts 02115, U.S.A.
Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston, Massachusetts 02115,
Biometrika. 2017 Mar;104(1):229-236. doi: 10.1093/biomet/asw064. Epub 2017 Jan 23.
Instrumental variables are widely used for estimating causal effects in the presence of unmeasured confounding. The discrete instrumental variable model has testable implications for the law of the observed data. However, current assessments of instrumental validity are typically based solely on subject-matter arguments rather than these testable implications, partly due to a lack of formal statistical tests with known properties. In this paper, we develop simple procedures for testing the binary instrumental variable model. Our methods are based on existing techniques for comparing two treatments, such as the [Formula: see text]-test and the Gail-Simon test. We illustrate the importance of testing the instrumental variable model by evaluating the exogeneity of college proximity using the National Longitudinal Survey of Young Men.
在存在未测量混杂因素的情况下，工具变量被广泛用于估计因果效应。离散工具变量模型对观测数据的规律具有可检验的含义。然而，目前对工具有效性的评估通常仅基于专业知识论证，而非这些可检验的含义，部分原因是缺乏具有已知性质的形式化统计检验。在本文中，我们开发了用于检验二元工具变量模型的简单程序。我们的方法基于现有的比较两种处理方法的技术，如t检验和盖尔 - 西蒙检验。我们通过使用全国青年男性纵向调查评估大学ximity的外生性来说明检验工具变量模型的重要性。
原文中“college proximity”表述似乎有误，可能是“college proximity to residence”之类的准确表述，但按照要求未做修改直接翻译。
