arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2103.08390econ.EMstat.MEstat.ML
收藏

评估新疗法的长期效果

Estimating the Long-Term Effects of Novel Treatments

Keith Battocchi, Eleanor Dillon, Maggie Hei, Greg Lewis, Miruna Oprescu, Vasilis Syrgkanis

摘要:

决策者通常面临的问题是,想要估计新疗法的长期效果,而只有旧疗法的历史数据。我们假设访问一个长期数据集,其中只管理过去的治疗和一个短期数据集,其中管理新的治疗。我们提出了一个基于代理的方法,我们假设长期的影响是通过大量的可用的短期代理。我们的工作结合了因果机器学习文献中的三个主要的最新技术:代理指数,动态处理效应估计和双重机器学习,在一个统一的管道。我们表明,我们的方法是一致的,并提供根N渐近正态估计下的马尔可夫假设的数据和观测政策。我们使用来自一家大公司的数据集,其中包括三年内的客户投资,以创建一个半合成的数据分布,其中保留了真实数据集的主要定性属性。我们评估我们的方法的性能,并讨论部署我们的正式方法的实际挑战,以及如何解决这些问题。

英文摘要:

Policy makers typically face the problem of wanting to estimate the long-term effects of novel treatments, while only having historical data of older treatment options. We assume access to a long-term dataset where only past treatments were administered and a short-term dataset where novel treatments have been administered. We propose a surrogate based approach where we assume that the long-term effect is channeled through a multitude of available short-term proxies. Our work combines three major recent techniques in the causal machine learning literature: surrogate indices, dynamic treatment effect estimation and double machine learning, in a unified pipeline. We show that our method is consistent and provides root-n asymptotically normal estimates under a Markovian assumption on the data and the observational policy. We use a data-set from a major corporation that includes customer investments over a three year period to create a semi-synthetic data distribution where the major qualitative properties of the real dataset are preserved. We evaluate the performance of our method and discuss practical challenges of deploying our formal methodology and how to address them.

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。

↑