Identifying causal effects in maximally oriented partially directed acyclic graphs
arXiv:1910.02997
Abstract
We develop a necessary and sufficient causal identification criterion for maximally oriented partially directed acyclic graphs (MPDAGs). MPDAGs as a class of graphs include directed acyclic graphs (DAGs), completed partially directed acyclic graphs (CPDAGs), and CPDAGs with added background knowledge. As such, they represent the type of graph that can be learned from observational data and background knowledge under the assumption of no latent variables. Our identification criterion can be seen as a generalization of the g-formula of Robins (1986). We further obtain a generalization of the truncated factorization formula (Pearl, 2009) and compare our criterion to the generalized adjustment criterion of Perković et al. (2017) which is sufficient, but not necessary for causal identification.
17 pages, 5 figures, 2 columns, Updated: Proof of Lemma A.3, thanks to Sara LaPlante for pointing out an issue
References in corpus (14)
- Causal Inference and Causal Explanation with Background Knowledge
- Characterization and Greedy Learning of Interventional Markov Equivalence Classes of Directed Acyclic Graphs
- Probabilistic Evaluation of Sequential Plans from Causal Models with Hidden Variables
- Pearl's Calculus of Intervention Is Complete
- On the Validity of Covariate Adjustment for Estimating Causal Effects
- A generalized back-door criterion
- Nested Markov Properties for Acyclic Directed Mixed Graphs
- Graphical Criteria for Efficient Total Effect Estimation via Adjustment in Causal Linear Models
- Permutation-based Causal Inference Algorithms with Interventions
- Causal discovery of linear acyclic models with arbitrary distributions
- Testing Identifiability of Causal Effects
- A Complete Generalized Adjustment Criterion
- Interpreting and using CPDAGs with background knowledge
- Causality on Cross-Sectional Data: Stable Specification Search in Constrained Structural Equation Modeling