Learning high-dimensional directed acyclic graphs with latent and selection variables
From MaRDI portal
Publication:450035
DOI10.1214/11-AOS940zbMATH Open1246.62131arXiv1104.5617OpenAlexW3103539622WikidataQ57707416 ScholiaQ57707416MaRDI QIDQ450035FDOQ450035
Authors: Diego Colombo, Markus Kalisch, Marloes H. Maathuis, Thomas S. Richardson 
Publication date: 3 September 2012
Published in: The Annals of Statistics (Search for Journal in Brave)
Abstract: We consider the problem of learning causal information between random variables in directed acyclic graphs (DAGs) when allowing arbitrarily many latent and selection variables. The FCI (Fast Causal Inference) algorithm has been explicitly designed to infer conditional independence and causal information in such settings. However, FCI is computationally infeasible for large graphs. We therefore propose the new RFCI algorithm, which is much faster than FCI. In some situations the output of RFCI is slightly less informative, in particular with respect to conditional independence information. However, we prove that any causal information in the output of RFCI is correct in the asymptotic limit. We also define a class of graphs on which the outputs of FCI and RFCI are identical. We prove consistency of FCI and RFCI in sparse high-dimensional settings, and demonstrate in simulations that the estimation performances of the algorithms are very similar. All software is implemented in the R-package pcalg.
Full work available at URL: https://arxiv.org/abs/1104.5617
Recommendations
consistencysparsityhigh-dimensionalitycausal structure learningFCI algorithmmaximal ancestral graphs (MAGs)partial ancestral graphs (PAGs)RFCI algorithm
Applications of graph theory (05C90) Applications of statistics (62P99) Directed graphs (digraphs), tournaments (05C20) Nonparametric inference (62G99)
Cites Work
- Causation, prediction, and search
- High-dimensional graphs and variable selection with the Lasso
- Title not available (Why is that?)
- Title not available (Why is that?)
- A characterization of Markov equivalence classes for acyclic digraphs
- Estimating high-dimensional directed acyclic graphs with the PC-algorithm
- Title not available (Why is that?)
- 10.1162/153244302760200696
- Estimating high-dimensional intervention effects from observational data
- Ancestral graph Markov models.
- Conditional independence for statistical operations
- Learning high-dimensional directed acyclic graphs with latent and selection variables
- Causal inference in statistics: an overview
- Markov equivalence for ancestral graphs
- On the completeness of orientation rules for causal discovery in the presence of latent confounders and selection bias
- Causal reasoning with ancestral graphs
Cited In (39)
- A uniformly consistent estimator of causal effects under the \(k\)-triangle-faithfulness assumption
- On scoring maximal ancestral graphs with the max-min hill climbing algorithm
- Title not available (Why is that?)
- A survey on causal discovery: theory and practice
- Learning high-dimensional directed acyclic graphs with latent and selection variables
- Learning causal structure from mixed data with missing values using Gaussian copula models
- Nonlinear Causal Discovery with Confounders
- Computation of maximum likelihood estimates in cyclic structural equation models
- Title not available (Why is that?)
- Discovering causal graphs with cycles and latent confounders: an exact branch-and-bound approach
- Structural intervention distance for evaluating causal graphs
- On the causal interpretation of acyclic mixed graphs under multivariate normality
- High-dimensional learning of linear causal networks via inverse covariance estimation
- Estimating bounds on causal effects in high-dimensional and possibly confounded systems
- Uniform random generation of large acyclic digraphs
- Causal Structural Learning via Local Graphs
- Causal learning via manifold regularization
- Learning Bayesian network structures using weakest mutual-information-first strategy
- Inferring gene regulatory networks by an order independent algorithm using incomplete data sets
- Towards a fast and efficient algorithm for learning Bayesian network
- Robust Causal Structure Learning with Some Hidden Variables
- Learning Moral Graphs in Construction of High-Dimensional Bayesian Networks for Mixed Data
- Learning genetic and environmental graphical models from family data
- Learning minimal latent directed information polytrees
- Nonparametric estimation of a latent variable model
- Learning causal graphs with latent confounders in weak faithfulness violations
- Title not available (Why is that?)
- A generalized back-door criterion
- CAM: causal additive models, high-dimensional order search and penalized regression
- An Approach to Reduce the Number of Conditional Independence Tests in the PC Algorithm
- Marginal integration for nonparametric causal inference
- Discovery and Inference of a Causal Network with Hidden Confounding
- Causal structure learning: a combinatorial perspective
- Improved baselines for causal structure learning on interventional data
- Empirical likelihood for linear structural equation models with dependent errors
- Order-independent constraint-based causal structure learning
- Log-mean linear parameterization for discrete graphical models of marginal independence and the analysis of dichotomizations
- Complete graphical characterization and construction of adjustment sets in Markov equivalence classes of ancestral graphs
- A decomposition-based algorithm for learning the structure of multivariate regression chain graphs
Uses Software
This page was built for publication: Learning high-dimensional directed acyclic graphs with latent and selection variables
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q450035)
