Using dagR to identify minimal sufficient adjustment sets and to simulate data based on directed acyclic graphs

Abstract

Abstract Motivation Directed acyclic graphs (DAGs) are causal diagrams that can be used to identify confounding or selection bias in observational studies, and are increasingly used in many areas of medical research. In the add-on package dagR for the statistical software R, a set of simple graphical rules was implemented to identify minimal sufficient adjustment sets or harmful adjustment. General features The dagR package allows an automated approach for realistic causal structures featuring numerous variables and dependencies, where a manual approach may be too tedious and error-prone. The algorithmic adherence to the graphical step-by-step approach often used in DAG theory introductory courses, together with functionalities for plotting and simulating data conforming to the causal structure of an arbitrary DAG, renders the dagR package particularly useful for both teaching purposes and methodological research. Availability dagR is available under the GNU general public licence (GPL-2) from within R or by download at [https://CRAN-R-project.org/package=dagR].