A generalized back-door criterion
arXiv:1307.5636 · doi:10.1214/14-AOS1295
Abstract
We generalize Pearl's back-door criterion for directed acyclic graphs (DAGs) to more general types of graphs that describe Markov equivalence classes of DAGs and/or allow for arbitrarily many hidden variables. We also give easily checkable necessary and sufficient graphical criteria for the existence of a set of variables that satisfies our generalized back-door criterion, when considering a single intervention and a single outcome variable. Moreover, if such a set exists, we provide an explicit set that fulfills the criterion. We illustrate the results in several examples. R-code is available in the R-package pcalg.
Published at http://dx.doi.org/10.1214/14-AOS1295 in the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)
References in corpus (7)
- Causal Inference and Causal Explanation with Background Knowledge
- Estimating high-dimensional intervention effects from observational data
- Markov equivalence for ancestral graphs
- On the Validity of Covariate Adjustment for Estimating Causal Effects
- Identification of Conditional Interventional Distributions
- Adjustment Criteria in Causal Diagrams: An Algorithmic Perspective
- Towards Characterizing Markov Equivalence Classes for Directed Acyclic Graphs with Latent Variables
Cited by in corpus (14)
- Graphical Criteria for Efficient Total Effect Estimation via Adjustment in Causal Linear Models
- Separators and Adjustment Sets in Causal Graphs: Complete Criteria and an Algorithmic Framework
- Complete Graphical Characterization and Construction of Adjustment Sets in Markov Equivalence Classes of Ancestral Graphs
- Causal Effect Identification from Multiple Incomplete Data Sources: A General Search-based Approach
- A Local Method for Identifying Causal Relations under Markov Equivalence
- Structural Intervention Distance (SID) for Evaluating Causal Graphs
- Efficient adjustment sets in causal graphical models with hidden variables
- Efficient least squares for estimating total effects under linearity and causal sufficiency
- Causal query in observational data with hidden variables
- Bias Correction For Paid Search In Media Mix Modeling
- A unified survey of treatment effect heterogeneity modeling and uplift modeling
- Identifying causal effects in maximally oriented partially directed acyclic graphs
- Towards unique and unbiased causal effect estimation from data with hidden variables
- Local search for efficient causal effect estimation