当前位置: X-MOL 学术arXiv.stat.OT › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Controlling for Latent Confounding with Triple Proxies
arXiv - STAT - Other Statistics Pub Date : 2022-04-28 , DOI: arxiv-2204.13815
Ben Deaner

We apply results in Hu and Schennach (2008) to achieve nonparametric identification of causal effects using noisy proxies for unobserved confounders. We call this the `triple proxy' approach because it requires three proxies that are jointly independent conditional on unobservables. We consider three different choices for the third proxy: it may be an outcome, a vector of treatments, or a collection of auxiliary variables. We compare to an alternative identification strategy introduced by Miao et. al. (2018) in which causal effects are identified using two conditionally independent proxies. We refer to this as the `double proxy' approach. We show that the conditional independence assumptions in the double and triple proxy approaches are non-nested, which suggests that either of the two identification strategies may be appropriate depending on the particular setting.

中文翻译:

使用三重代理控制潜在混杂

我们应用 Hu 和 Schennach (2008) 的结果,使用未观察到的混杂因素的噪声代理来实现因果效应的非参数识别。我们称其为“三重代理”方法,因为它需要三个共同独立的代理,以不可观测为条件。我们考虑了第三个代理的三种不同选择:它可能是一个结果、一个治疗向量或一组辅助变量。我们与 Miao 等人介绍的另一种识别策略进行了比较。人。(2018),其中使用两个条件独立的代理来确定因果关系。我们将此称为“双重代理”方法。我们表明,双重和三重代理方法中的条件独立假设是非嵌套的,
更新日期:2022-05-02
down
wechat
bug