2011
DOI: 10.2139/ssrn.1943090
|View full text |Cite
|
Sign up to set email alerts
|

Ebalance: A Stata Package for Entropy Balancing

Abstract: The Stata package ebalance implements entropy balancing, a multivariate reweighting method described in Hainmueller (2012) that allows users to reweight a dataset such that the covariate distributions in the reweighted data satisfy a set of specified moment conditions. This can be useful to create balanced samples in observational studies with a binary treatment where the control group data can be reweighted to match the covariate moments in the treatment group. Entropy balancing can also be used to reweight a… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

2
236
0
2

Year Published

2014
2014
2023
2023

Publication Types

Select...
10

Relationship

0
10

Authors

Journals

citations
Cited by 252 publications
(247 citation statements)
references
References 21 publications
(12 reference statements)
2
236
0
2
Order By: Relevance
“…While we view this work as a step towards understanding the linkages between emigration and the different subjective well-being dimensions (evaluative well-being, positive hedonic 17 We rely on the user-written command ebalance in Stata (Hainmueller and Xu 2013). 18 More precisely, entropy balancing is more efficient and reduces covariate imbalance compared with techniques such as propensity score matching (PSM).…”
Section: Limitationsmentioning
confidence: 99%
“…While we view this work as a step towards understanding the linkages between emigration and the different subjective well-being dimensions (evaluative well-being, positive hedonic 17 We rely on the user-written command ebalance in Stata (Hainmueller and Xu 2013). 18 More precisely, entropy balancing is more efficient and reduces covariate imbalance compared with techniques such as propensity score matching (PSM).…”
Section: Limitationsmentioning
confidence: 99%
“…Entropy balancing achieves balance over specified moments of selected covariates by deriving sample weights. The retrieved weights are then used in subsequent weighted estimations (Hainmueller, 2011;Hainmueller and Xu, 2013). Intuitively, this can be understood as the creation of a synthetic control group, where the observations in the control group are reweighted so that their specified sample moments mimic those of the treatment group as closely as possible (cf.…”
Section: Estimation Approach and Identificationmentioning
confidence: 99%
“…However, irrespective of the significance of these differences, the size of some of the differences may trigger concerns about the comparability of treatment and control group students. In order to increase the similarity of treated and control group students we rerun our estimation using entropy balancing weights (Hainmueller, 2012;Hainmueller and Xu, 2013). Entropy balancing weights reweigh control group students such that a set of pre-specified moment conditions are equal across treatment status.…”
Section: Robustness Testsmentioning
confidence: 99%