Software:
SpatialEffect
This package allows researchers to estimate the causal quantity called the "average marginalized effect" (AME) proposed by Wang et al., (2025), which is equal to the average effect of activating a treatment at an intervention node that is a given distance away, averaging ambient effects emanating from other intervention nodes. This approach provides credible causal effect estimates even in the absence of any knowledge on the interference structure.
​
New version available soon.
CTB
This package implements the covariate-tightened trimming bounds (CTB) method proposed by Samii, Wang, and Zhou (2025). CTB generalizes the classic trimming bounds (TB) approach of Lee (2009) by leveraging the generalized random forest algorithm (Athey, Tibshirani, and Wager, 2019) to incorporate information from a potentially high-dimensional set of covariates. The method enables users to construct trimming bounds for the average treatment effect among always-responders conditional on covariate values, and to aggregate these conditional bounds into a tighter identified set. It is built upon the idea of double machine learning (Chernozhukov et al., 2018) and provides users with honest confidence intervals.​
​
fect (winner of SPM's best statistical software award in 2025)
This package implements counterfactual estimators in TSCS data analysis based on Liu, Wang, and Xu (2024). These estimators first impute counterfactuals for each treated observation in a TSCS dataset by fitting an outcome model (fixed effects model, interactive fixed effects model, or matrix completion) using the untreated observations. They then estimate the individualistic treatment effect for each treated observatio n by subtracting the predicted counterfactual outcome from its observed outcome. Finally, the average treatment effect on the treated (ATT) or period-specific ATTs are calculated. A placebo test and an equivalence test are included to evaluate the validity of identification assumptions behind these estimators.
​