FLAME: A Fast Large-scale Almost Matching Exactly Approach to Causal Inference

By Sudeepa Roy, Cynthia Rudin, Alexander Volfovsky. 2017

54 citations recorded by the source corpus.

A classical problem in causal inference is that of matching, where treatment units need to be matched to control units based on covariate information. In this work, we propose a method that computes high quality almost-exact matches for high-dimensional categorical datasets. This method, called FLAME (Fast Large-scale Almost Matching Exactly), learns a distance metric for matching using a hold-out training data set. In order to perform matching efficiently for large datasets, FLAME leverages techniques that are natural for query processing in the area of database management, and two implementations of FLAME are provided: the first uses SQL queries and the second uses bit-vector techniques. The algorithm starts by constructing matches of the highest quality (exact matches on all covariates), and successively eliminates variables in order to match exactly on as many variables as possible, while still maintaining interpretable high-quality matches and balance between treatment and control groups. We leverage these high quality matches to estimate conditional average treatment effects (CATEs). Our experiments show that FLAME scales to huge datasets with millions of observations where existing state-of-the-art methods fail, and that it achieves significantly better performance than other matching methods.

Subfields: Artificial Intelligence, Statistics and Probability, Causal Inference & Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

In Causal Inference & Discovery

Citations in the corpus

dame-flame: A Python Library Providing Fast Interpretable Matching for Causal Inference (2021, background): “As discussed in Dieng et al. (2019), and Wang et al. (2019), the two algorithms produce high-quality interpretable matched groups, by using machine learning on a holdout training set to learn distance metrics.”
Interpretable Almost-Exact Matching for Causal Inference (2019, methodology): “This reduction is discussed formally by Wang et al. (2017).”
Interpretable Almost Matching Exactly With Instrumental Variables (2019, methodology): “We create an Almost-Matching Exactly framework (Wang et al., 2019; Dieng et al., 2019) for the purpose of instrumental variable analysis.”
Almost-Matching-Exactly for Treatment Effect Estimation under Network Interference (2020, methodology): “The Fast Large Almost Matching Exactly (FLAME) algorithm (Wang et al., 2019) approximates the solution to the AME problem.”
Assessing Treatment Effect Variation in Observational Studies: Results from a Data Challenge (2019, background): “Chernozhukov et al. (2018) offer some promising directions in the randomized trial setting; see also the discussion in Athey & Wager and Zhao & Panigrahi.”
MALTS: Matching After Learning to Stretch (2018, methodology): “We consider both a conservative variance estimator (Wang et al. 2021) and estimators that sacrifice some interpretability for better coverage.”