Linear Models: A Useful "Microscope" for Causal Analysis
Linear Models: A Useful "Microscope" for Causal Analysis
复制标题
DOI：
10.1515/jci-2013-0003
复制
发表时间：
2013-05-01
影响因子：
1.4
通讯作者：
Pearl, Judea
中科院分区：
文献类型：
--
作者：
Pearl, Judea
本说明回顾了线性路径分析的基本技术，并通过简单示例展示了如何利用图表和一些代数步骤来理解、例证和分析具有非平凡特征的因果现象。这些技术能够快速评估模型的各种特征如何影响所研究的现象。这包括：辛普森悖论、病例 - 对照偏倚、选择偏倚、缺失数据、对撞偏倚、反向回归、偏倚放大、近似工具变量以及测量误差。
This note reviews basic techniques of linear path analysis and demonstrates, using simple examples, how causal phenomena of non-trivial character can be understood, exemplified and analyzed using diagrams and a few algebraic steps. The techniques allow for swift assessment of how various features of the model impact the phenomenon under investigation. This includes: Simpson's paradox, case-control bias, selection bias, missing data, collider bias, reverse regression, bias amplification, near instruments, and measurement errors.