MALTS: Matching After Learning to Stretch
By Harsh Parikh, Cynthia Rudin, Alexander Volfovsky. 2018
30 citations recorded by the source corpus.
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 covariates according to their contribution to outcome prediction. The framework is flexible in that the user can choose the form of the distance metric and the type of optimization algorithm. Our ability to learn flexible distance metrics leads to matches that are interpretable and useful for the estimation of conditional average treatment effects.
Subfields: Statistics and Probability
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Statistics and Probability
- Cynthia Rudin
- Alexander Volfovsky
- Sudeepa Roy
- Berk Ustun
- Harsh Parikh
- Kush R. Varshney
- Nadia Burkart
- Marco F. Huber
Citations in the corpus
- FLAME: A Fast Large-scale Almost Matching Exactly Approach to Causal Inference (2017, methodology): “This idea is the foundation for the MALTS algorithm (Parikh et al., 2018; Parikh et al., 2019).”
- dame-flame: A Python Library Providing Fast Interpretable Matching for Causal Inference (2021, unknown): “Instead of a predetermined distance metric like MatchIt , MALTS gives covariates that contribute more towards predicting the outcome higher weights (Parikh, Rudin, and Volfovsky 2018).”