We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

econ.EM

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Economics > Econometrics

Title: Evidence-Based Policy Learning

Abstract: The past years have seen seen the development and deployment of machine-learning algorithms to estimate personalized treatment-assignment policies from randomized controlled trials. Yet such algorithms for the assignment of treatment typically optimize expected outcomes without taking into account that treatment assignments are frequently subject to hypothesis testing. In this article, we explicitly take significance testing of the effect of treatment-assignment policies into account, and consider assignments that optimize the probability of finding a subset of individuals with a statistically significant positive treatment effect. We provide an efficient implementation using decision trees, and demonstrate its gain over selecting subsets based on positive (estimated) treatment effects. Compared to standard tree-based regression and classification tools, this approach tends to yield substantially higher power in detecting subgroups with positive treatment effects.
Subjects: Econometrics (econ.EM); Machine Learning (cs.LG); Methodology (stat.ME); Machine Learning (stat.ML)
Cite as: arXiv:2103.07066 [econ.EM]
  (or arXiv:2103.07066v1 [econ.EM] for this version)

Submission history

From: Jann Spiess [view email]
[v1] Fri, 12 Mar 2021 03:36:03 GMT (296kb,D)

Link back to: arXiv, form interface, contact.