SimcAusal R Package: Conducting Transparent and Reproducible Simulation Studies of Causal Effect Estimation with Complex Longitudinal Data

• Activity Number: 539
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Oleg Sofrygin* and Mark Johannes van der Laan and Romain Neugebauer
• Companies: Kaiser Permanente Northern California/UC Berkeley and UC Berkeley and Kaiser Permanente Northern California
• Keywords: simulation study ; causal inference ; longitudinal data ; MSMs ; R package

Abstract

We introduce the simcausal R package as a tool for specification and simulation of complex longitudinal data structures based on structural equation models. The main motivation is to provide a flexible tool to facilitate the conduct of transparent and reproducible simulation studies with a particular emphasis on the types of data and interventions frequently encountered in real-world causal inference problems. We developed new R interface that allows for concise expression of complex functional dependencies for a large number of time-varying nodes. The package can simulate counterfactual data under various interventions (e.g., static, dynamic, deterministic, or stochastic), on a sequence of nodes that may represent exposures to treatment regimens, right-censoring events, or clinical monitoring events. It also enables the computation of a selected set of parameters from the counterfactual data that represent common causal quantities of interest, such as, the average treatment effect (ATE) and coefficients from the working marginal structural model (MSM). We demonstrate the use of simcausal by replicating results of two simulation studies from the causal inference literature.