Replication code for: \"Causal inference with a continuous treatment: Addressing positivity constraints, nonlinearity, and effect heterogeneity\"

Causal inference approaches often emphasize binary treatments. But in many applications, the underlying constructs are continuous. In the potential outcomes framework, a continuous treatment can take on numerous values, each corresponding to a potential outcome that may be realized. In this setting, common estimands may be intractable due to a common issue in social research, particularly research on social inequality: the exposure is highly stratified by confounders. We show how to avoid drawing inferences about counterfactuals where data are unlikely to exist by carefully selecting the causal estimand. We adopt an additive shift estimand that adds a small, fixed amount to each unit's income. Our approach is preferable to population-average dose-response curves in settings where some treatment values rarely occur in some subgroups. We also show how to estimate and summarize patterns of nonlinearity and effect heterogeneity with continuous treatments. As a motivating example, we consider the causal effect of parental income on college attendance, a setting in which the exposure is highly stratified by confounders (e.g., parental education). Our approach applies to a wide range of possible treatment conditions in sociology.