CAM: causal additive models, high-dimensional order search and penalized regression

DOI10.1214/14-AOS1260zbMATH Open1309.62063arXiv1310.1533OpenAlexW1977446441MaRDI QIDQ482906FDOQ482906

Authors: Jan Ernest, Jonas Peters, Peter Bühlmann 

Publication date: 6 January 2015

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: We develop estimation for potentially high-dimensional additive structural equation models. A key component of our approach is to decouple order search among the variables from feature or edge selection in a directed acyclic graph encoding the causal structure. We show that the former can be done with nonregularized (restricted) maximum likelihood estimation while the latter can be efficiently addressed using sparse regression techniques. Thus, we substantially simplify the problem of structure search and estimation for an important class of causal models. We establish consistency of the (restricted) maximum likelihood estimator for low- and high-dimensional scenarios, and we also allow for misspecification of the error distribution. Furthermore, we develop an efficient computational algorithm which can deal with many variables, and the new method's accuracy and performance is illustrated on simulated and real data.

Full work available at URL: https://arxiv.org/abs/1310.1533

Recommendations

• Causal Dantzig: fast inference in linear structural equation models with hidden variables under additive interventions
• ParceLiNGAM: a causal ordering method robust against latent confounders
• scientific article
• Causal statistical inference in high dimensions
• Causal inference in high dimensions: A marriage between Bayesian modeling and good frequentist properties
• scientific article; zbMATH DE number 2232736
• High-dimensional causal mediation analysis based on partial linear structural equation models
• Causal inference in partially linear structural equation models

zbMATH Keywords

nonparametric regressionstructural equation modelsparsityregularized estimationgraphical modelingintervention calculus

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Measures of association (correlation, canonical correlation, etc.) (62H20) Graphical methods in statistics (62A09) Artificial intelligence (68T99)

Cites Work

• Model-based boosting 2.0
• High-dimensional additive modeling
• Causation, prediction, and search
• Generalized additive models. An introduction with R.
• The Adaptive Lasso and Its Oracle Properties
• Title not available (Why is that?)
• Statistics for high-dimensional data. Methods, theory and applications.
• High-dimensional graphs and variable selection with the Lasso
• Title not available (Why is that?)
• Sharp Thresholds for High-Dimensional and Noisy Sparsity Recovery Using $\ell _{1}$-Constrained Quadratic Programming (Lasso)
• Model Selection and Estimation in Regression with Grouped Variables
• On measures of dependence
• \(\ell_{0}\)-penalized maximum likelihood for sparse directed acyclic graphs
• Estimating Optimal Transformations for Multiple Regression and Correlation
• Title not available (Why is that?)
• Title not available (Why is that?)
• Boosting algorithms: regularization, prediction and model fitting
• Practical variable selection for generalized additive models
• The sparsity and bias of the LASSO selection in high-dimensional linear regression
• Boosting With theL₂Loss
• The adaptive and the thresholded Lasso for potentially misspecified models (and a lower bound for the Lasso)
• Graph estimation with joint additive models
• Title not available (Why is that?)
• Learning high-dimensional directed acyclic graphs with latent and selection variables
• Identifiability of Gaussian structural equation models with equal error variances
• On the uniform convergence of empirical norms and inner products, with application to causal inference
• A linear non-Gaussian acyclic model for causal discovery
• 10.1162/153244303321897717
• A simple smooth backfitting method for additive models
• High-dimensional learning of linear causal networks via inverse covariance estimation
• CAM: causal additive models, high-dimensional order search and penalized regression
• Score-based causal learning in additive noise models

Cited In (30)

• Causal discovery in heavy-tailed models
• A survey on causal discovery: theory and practice
• Learning high-dimensional Gaussian linear structural equation models with heterogeneous error variances
• Nonlinear Causal Discovery with Confounders
• Model selection and local geometry
• Identification of vector autoregressive models with nonlinear contemporaneous structure
• Max-linear models on directed acyclic graphs
• Invariance, causality and robustness
• Likelihood Ratio Tests for a Large Directed Acyclic Graph
• A new causal discovery heuristic
• Why BDeu? Regular Bayesian network structure learning with discrete and continuous variables
• Title not available (Why is that?)
• Testing Directed Acyclic Graph via Structural, Supervised and Generative Adversarial Learning
• Foundations of structural causal models with cycles and latent variables
• Causal effect identification in acyclic directed mixed graphs and gated models
• CAM: causal additive models, high-dimensional order search and penalized regression
• Marginal integration for nonparametric causal inference
• Score-based causal learning in additive noise models
• Causal structure learning: a combinatorial perspective
• Identifiability of Gaussian linear structural equation models with homogeneous and heterogeneous error variances
• Distance Metrics for Measuring Joint Dependence with Application to Causal Inference
• Efficient Learning of Quadratic Variance Function Directed Acyclic Graphs via Topological Layers
• Nonparametric and high-dimensional functional graphical models
• Invariant Causal Prediction for Sequential Data
• Kernel-Based Tests for Joint Independence
• A review of Gaussian Markov models for conditional independence
• Efficient learning of nonparametric directed acyclic graph with statistical guarantee
• Structural Intervention Distance for Evaluating Causal Graphs
• Title not available (Why is that?)
• Causal inference in partially linear structural equation models

Uses Software

• ElemStatLearn
• gamair
• hgam
• TETRAD

This page was built for publication: CAM: causal additive models, high-dimensional order search and penalized regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q482906)