dame-flame: A Python Library Providing Fast Interpretable Matching for Causal Inference.

dc.contributor.author

Gupta, Neha R

dc.contributor.author

Orlandi, Vittorio

dc.contributor.author

Chang, Chia-Rui

dc.contributor.author

Wang, Tianyu

dc.contributor.author

Morucci, Marco

dc.contributor.author

Dey, Pritam

dc.contributor.author

Howell, Thomas J

dc.contributor.author

Sun, Xian

dc.contributor.author

Ghosal, Angikar

dc.contributor.author

Roy, Sudeepa

dc.contributor.author

Rudin, Cynthia

dc.contributor.author

Volfovsky, Alexander

dc.date.accessioned

2021-04-01T14:46:46Z

dc.date.available

2021-04-01T14:46:46Z

dc.date.issued

2021

dc.date.updated

2021-04-01T14:46:45Z

dc.description.abstract

dame-flame is a Python package for performing matching for observational causal inference on datasets containing discrete covariates. This package implements the Dynamic Almost Matching Exactly (DAME) and Fast Large-Scale Almost Matching Exactly (FLAME) algorithms, which match treatment and control units on subsets of the covariates. The resulting matched groups are interpretable, because the matches are made on covariates (rather than, for instance, propensity scores), and high-quality, because machine learning is used to determine which covariates are important to match on. DAME solves an optimization problem that matches units on as many covariates as possible, prioritizing matches on important covariates. FLAME approximates the solution found by DAME via a much faster backward feature selection procedure. The package provides several adjustable parameters to adapt the algorithms to specific applications, and can calculate treatment effects after matching. Descriptions of these parameters, details on estimating treatment effects, and further examples, can be found in the documentation at https://almost-matching-exactly.github.io/DAME-FLAME-Python-Package/

dc.identifier.uri

https://hdl.handle.net/10161/22495

dc.relation.ispartof

CoRR

dc.subject

cs.LG

dc.subject

cs.LG

dc.subject

cs.MS

dc.title

dame-flame: A Python Library Providing Fast Interpretable Matching for Causal Inference.

dc.type

Journal article

duke.contributor.orcid

Volfovsky, Alexander|0000-0003-4462-1020

pubs.organisational-group

Trinity College of Arts & Sciences

pubs.organisational-group

Computer Science

pubs.organisational-group

Statistical Science

pubs.organisational-group

Duke

pubs.volume

abs/2101.01867

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
2101.01867v2.pdf
Size:
709.56 KB
Format:
Adobe Portable Document Format