A General Method for Detecting Interference Between Units in Randomized Experiments
By P. M. Aronow. 2012
113 citations recorded by the source corpus.
Subfields: Education, 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 Education
- Susan Athey
- Vittorio Orlandi
- Panos Toulis
- Dean Eckles
- M. Usaid Awan
- René F. Kizilcec
- Eytan Bakshy
- Sudeepa Roy
In Statistics and Probability
- Cynthia Rudin
- Alexander Volfovsky
- Sudeepa Roy
- Susan Athey
- Avi Feller
- Guido W. Imbens
- Stefan Wager
- Marco Morucci
Citations in the corpus
- The Econometrics of Randomized Experiments (2016, methodology): “First, the null hypothesis of no interactions whatsoever, that is the null hypothesis that changing the treatment status for friends does not change an individual’s outcome, also considered in Aronow (2012), and second, the null hypothesis that the treatment of a friend of a friend does not have a c”
- Almost-Matching-Exactly for Treatment Effect Estimation under Network Interference (2020, methodology): “Other approaches lay out methods based on randomization inference to test a variety of hypotheses under interference and treatment randomization (Rosenbaum, 2007; Aronow, 2012; Athey et al., 2018).”
- Estimating peer effects in networks with peer encouragement designs (2016, background): “That is, compared with the conditional randomization methods in Aronow (2) and Athey et al.”
- Randomization tests of causal effects under interference (2019, methodology): “Aronow (2012), who outlines some ideas that we discuss here, develops a test for the null hypothesis of no spillover effect. Although that paper does not exclude conditioning on Z , it gives limited guidance on how such conditioning would work. Athey et al. (2018) extend the method of Aronow (2012) ”
- Causal inference with misspecified exposure mappings: separating definitions and assumptions (2021, background): “Aronow and Samii (2017) recognize that the key methodological tool in Hudgens and Halloran (2008) was a terse description of the treatments assigned to all units. This insight suggests a generalization beyond proportions of treated units in some neighborhood to arbitrary summaries of the nominal tre”