herashi / mrt-mhealthmoderation

Assessing Time-Varying Causal Effect Moderation in the Presence of Cluster-Level Treatment Effect Heterogeneity

Assessing Time-Varying Causal Effect Moderation in the Presence of Cluster-Level Treatment Effect Heterogeneity

The directories above correspond to the simulation and case study in the paper Assessing Time-Varying Causal Effect Moderation in the Presence of Cluster-Level Treatment Effect Heterogeneity. Results for each scenario are generated by the following R files.

File	Description
Scenario 1	The first scenario concerns the estimation when an importantindividual-level moderator exists and proximal outcomes share a random cluster-level intercept term that does not interact with treatment.
Scenario 2	The second scenario extends the above generative model to includea random cluster-level intercept term that interacts with treatment.
Scenario 3	The third scenario assumes the treatment effect for an individualdepends on the average state of all individuals in the cluster.
Scenario 4	The fourth scenario considers the indirect effect.
Case Study	This folder contains all the code for generating both the direct and indirect effect estimates using the Intern Health Study.