Causal DAGs are used in statistics for causal analysis. Also, widely misused. When real causality isn’t acyclic and real dependence is highly nonlinear, inference based on the assumption that there exists some (quasi-linear) causal DAG can go very very wrong. It may be closer to being true than just assuming that the most complicated structure is bivariate interaction (Z depends on the combination of X and Y), but it is also a lot more dangerous.
Causal DAGs are used in statistics for causal analysis. Also, widely misused. When real causality isn’t acyclic and real dependence is highly nonlinear, inference based on the assumption that there exists some (quasi-linear) causal DAG can go very very wrong. It may be closer to being true than just assuming that the most complicated structure is bivariate interaction (Z depends on the combination of X and Y), but it is also a lot more dangerous.