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

Download:

Current browse context:

stat.ME

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

Statistics > Methodology

Title: MALTS: Matching After Learning to Stretch

Abstract: We introduce a flexible framework that produces high-quality almost-exact matches for causal inference. Most prior work in matching uses ad-hoc distance metrics, often leading to poor quality matches, particularly when there are irrelevant covariates. In this work, we learn an interpretable distance metric for matching, which leads to substantially higher quality matches. The learned distance metric stretches the covariate space according to each covariate's contribution to outcome prediction: this stretching means that mismatches on important covariates carry a larger penalty than mismatches on irrelevant covariates. Our ability to learn flexible distance metrics leads to matches that are interpretable and useful for the estimation of conditional average treatment effects.
Comments: 40 pages, 5 Tables, 12 Figures
Subjects: Methodology (stat.ME); Machine Learning (cs.LG); Econometrics (econ.EM)
Cite as: arXiv:1811.07415 [stat.ME]
  (or arXiv:1811.07415v8 [stat.ME] for this version)

Submission history

From: Harsh Parikh [view email]
[v1] Sun, 18 Nov 2018 22:29:59 GMT (5393kb,D)
[v2] Sun, 2 Dec 2018 08:26:20 GMT (5557kb,D)
[v3] Fri, 16 Aug 2019 01:24:24 GMT (3359kb,D)
[v4] Wed, 16 Oct 2019 23:14:28 GMT (4285kb,D)
[v5] Sat, 15 Aug 2020 10:57:46 GMT (1423kb,D)
[v6] Wed, 22 Sep 2021 19:00:48 GMT (4484kb,D)
[v7] Wed, 20 Jul 2022 01:50:12 GMT (4521kb,D)
[v8] Thu, 22 Sep 2022 17:54:58 GMT (4451kb,D)

Link back to: arXiv, form interface, contact.