arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.13544stat.MEstat.AP
收藏

利用支持集不重叠的观察性数据与实验数据估计处理效应

Estimating Treatment Effects Using Observational Data and Experimental Data with Non-overlapping Support

Kevin Han

AI总结:

针对随机实验受限且观察性数据无混杂假设难验证的问题,本文在两类数据支持集不重叠、实验仅观测替代结果的设定下,提出结合二者估计目标平均处理效应的简单估计量。

摘要:

在估计处理效应时,黄金标准是开展随机实验,然后比较处理组与对照组对应的结果。然而,在许多情况下,随机实验的规模相比目标人群要小得多,或伴随某些伦理问题,因而难以实施。因此,研究人员通常依赖观察性数据来研究因果联系。其缺点在于,作为验证观察性数据使用之关键的无混杂假设很难检验,而且几乎总是被违背。因此,从观察性数据得出的任何结论都应进一步谨慎分析。鉴于观察性数据的丰富性和实验数据的有用性,研究人员希望开发可靠的方法,以结合二者的优势。本文考虑一种设定:观察性数据包含感兴趣的结果以及一个替代结果,而实验数据仅包含替代结果。我们提出一个简单的估计量,利用观察性数据和实验数据来估计感兴趣的平均处理效应。

英文摘要:

When estimating treatment effects, the golden standard is to conduct a randomized experiment and then contrast outcomes associated with the treatment group and the control group. However, in many cases, randomized experiments are either conducted with a much smaller scale compared to the size of the target population or accompanied with certain ethical issues and thus hard to implement. Therefore, researchers usually rely on observational data to study causal connections. The downside is that the unconfoundedness assumption, the key to validate the use of observational data is hard to verify and almost always violated. Hence, any conclusion drawn from observational data should be further analyzed with great care. Given the richness of observational data and usefulness of experimental data, researchers hope to develop credible method to combine the strength of the two. In this paper, we consider a setting where the observational data contain the outcome of interest as well as a surrogate outcome while the experimental data contain only the surrogate outcome. We propose a simple estimator to estimate the average treatment effect of interest using both the observational data and the experimental data.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑