The goal of stabilizedinterference is to estimate causal effects in the presence of interference, with “stabilized” IPTW estimators as introduced in Liu, Hudgens, and Becker-Dreps (2016) Biometrika.
These estimators estimate causal effects in the presence of partial or clustered interference, with estimates of the asymptotic variance from standard M-estimation theory.
IPW estimators are implemented in the function
This package is in working order, but currently under development. Suggestions are welcome; thanks for your patience!