Estimating high-dimensional intervention effects from observational data
From MaRDI portal
Publication:1043733
DOI10.1214/09-AOS685zbMATH Open1191.62118arXiv0810.4214OpenAlexW2050502901WikidataQ57707432 ScholiaQ57707432MaRDI QIDQ1043733FDOQ1043733
Authors: Marloes H. Maathuis, Markus Kalisch, Peter Bühlmann 
Publication date: 9 December 2009
Published in: The Annals of Statistics (Search for Journal in Brave)
Abstract: We assume that we have observational data generated from an unknown underlying directed acyclic graph (DAG) model. A DAG is typically not identifiable from observational data, but it is possible to consistently estimate the equivalence class of a DAG. Moreover, for any given DAG, causal effects can be estimated using intervention calculus. In this paper, we combine these two parts. For each DAG in the estimated equivalence class, we use intervention calculus to estimate the causal effects of the covariates on the response. This yields a collection of estimated causal effects for each covariate. We show that the distinct values in this set can be consistently estimated by an algorithm that uses only local information of the graph. This local approach is computationally fast and feasible in high-dimensional problems. We propose to use summary measures of the set of possible causal effects to determine variable importance. In particular, we use the minimum absolute value of this set, since that is a lower bound on the size of the causal effect. We demonstrate the merits of our methods in a simulation study and on a data set about riboflavin production.
Full work available at URL: https://arxiv.org/abs/0810.4214
Recommendations
- Estimating the effect of joint interventions from observational data in sparse high-dimensional settings
- Bayesian inference of causal effects from observational data in Gaussian graphical models
- Causal Dantzig: fast inference in linear structural equation models with hidden variables under additive interventions
- Estimating bounds on causal effects in high-dimensional and possibly confounded systems
- Jointly interventional and observational data: estimation of interventional Markov equivalence classes of directed acyclic graphs
sparsitygraphical modelingPC-algorithmintervention calculuscausal analysisdirected acyclic graph (DAG)riboflavin data
Multivariate analysis (62H99) Applications of graph theory (05C90) Directed graphs (digraphs), tournaments (05C20)
Cites Work
- Causation, prediction, and search
- Weak convergence and empirical processes. With applications to statistics
- High-dimensional graphs and variable selection with the Lasso
- Title not available (Why is that?)
- Bayesian analysis in expert systems. With comments and a rejoinder by the authors
- Learning Bayesian networks: The combination of knowledge and statistical data
- Causal inference from graphical models
- Uniform consistency in causal inference
- Estimating high-dimensional directed acyclic graphs with the PC-algorithm
- Statistics and Causal Inference
- Title not available (Why is that?)
- Title not available (Why is that?)
- 10.1162/153244302760200696
- Approximating discrete probability distributions with dependence trees
- A note on the Lasso for Gaussian graphical model selection
- Large-Scale Simultaneous Hypothesis Testing
- Ancestral graph Markov models.
- Statistics and causal inference: A review. (With discussion)
- Confounding and collapsibility in causal inference
- Causal diagrams for empirical research
- Incidence matrices and interval graphs
- On rigid circuit graphs
- A linear non-Gaussian acyclic model for causal discovery
- 10.1162/153244303321897717
- On the Desirability of Acyclic Database Schemes
- Causal Inference Without Counterfactuals
- Title not available (Why is that?)
- Causal reasoning with ancestral graphs
Cited In (53)
- Discussion of big Bayes stories and BayesBag
- Geometry of the faithfulness assumption in causal inference
- The dual PC algorithm and the role of Gaussianity for structure learning of Bayesian networks
- Reconstruction of a directed acyclic graph with intervention
- Learning high-dimensional directed acyclic graphs with latent and selection variables
- High-dimensional consistency in score-based and hybrid structure learning
- Objective Bayes model selection of Gaussian interventional essential graphs for the identification of signaling pathways
- High dimensional sparse covariance estimation via directed acyclic graphs
- Causal statistical inference in high dimensions
- Variable selection for high-dimensional incomplete data using horseshoe estimation with data augmentation
- Bayesian sample size determination for causal discovery
- Structural intervention distance for evaluating causal graphs
- Estimating the effect of joint interventions from observational data in sparse high-dimensional settings
- Estimating bounds on causal effects in high-dimensional and possibly confounded systems
- A local method for identifying causal relations under Markov equivalence
- Invariance, causality and robustness
- Functional Bayesian networks for discovering causality from multivariate functional data
- Individualized causal discovery with latent trajectory embedded Bayesian networks
- \(\mathsf{PenPC}\): a two-step approach to estimate the skeletons of high-dimensional directed acyclic graphs
- Causal learning via manifold regularization
- Discussion to: Bayesian graphical models for modern biological applications by Y. Ni, V. Baladandayuthapani, M. Vannucci and F.C. Stingo
- Inferring gene regulatory networks by an order independent algorithm using incomplete data sets
- Sufficient dimension reduction for average causal effect estimation
- Network modeling in biology: statistical methods for gene and brain networks
- Approximation of generalized ridge functions in high dimensions
- Estimating exogenous variables in data with more variables than observations
- Robust Causal Structure Learning with Some Hidden Variables
- Switching regression models and causal inference in the presence of discrete latent variables
- Causal inference in genetic trio studies
- Bayesian causal inference in probit graphical models
- Multiple imputation and test-wise deletion for causal discovery with incomplete cohort data
- Model free estimation of graphical model using gene expression data
- An efficient algorithm for counting Markov equivalent DAGs
- Foundations of structural causal models with cycles and latent variables
- A generalized back-door criterion
- Fast causal orientation learning in directed acyclic graphs
- Marginal integration for nonparametric causal inference
- Learning instrumental variables with structural and non-Gaussianity assumptions
- Capturing ridge functions in high dimensions from point queries
- \(\ell_{0}\)-penalized maximum likelihood for sparse directed acyclic graphs
- Improved baselines for causal structure learning on interventional data
- Title not available (Why is that?)
- Structural learning and estimation of joint causal effects among network-dependent variables
- Testing conditional independence in supervised learning algorithms
- Log-mean linear parameterization for discrete graphical models of marginal independence and the analysis of dichotomizations
- Beyond the mean: a flexible framework for studying causal effects using linear models
- Efficient Sampling and Structure Learning of Bayesian Networks
- Inferring network structure from interventional time-course experiments
- Complete graphical characterization and construction of adjustment sets in Markov equivalence classes of ancestral graphs
- Bayesian inference of causal effects from observational data in Gaussian graphical models
- Reversible MCMC on Markov equivalence classes of sparse directed acyclic graphs
- Causal discoveries for high dimensional mixed data
- Estimating causal effects from nonparanormal observational data
Uses Software
This page was built for publication: Estimating high-dimensional intervention effects from observational data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1043733)
