CAM: causal additive models, high-dimensional order search and penalized regression
From MaRDI portal
(Redirected from Publication:482906)
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.
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
Cites work
- scientific article; zbMATH DE number 6378135 (Why is no real title available?)
- scientific article; zbMATH DE number 5957408 (Why is no real title available?)
- scientific article; zbMATH DE number 1134987 (Why is no real title available?)
- scientific article; zbMATH DE number 1493045 (Why is no real title available?)
- scientific article; zbMATH DE number 845714 (Why is no real title available?)
- 10.1162/153244303321897717
- A linear non-Gaussian acyclic model for causal discovery
- A simple smooth backfitting method for additive models
- Boosting With theL2Loss
- Boosting algorithms: regularization, prediction and model fitting
- CAM: causal additive models, high-dimensional order search and penalized regression
- Causation, prediction, and search
- Estimating Optimal Transformations for Multiple Regression and Correlation
- Generalized additive models. An introduction with R.
- Graph estimation with joint additive models
- High-dimensional additive modeling
- High-dimensional graphs and variable selection with the Lasso
- High-dimensional learning of linear causal networks via inverse covariance estimation
- Identifiability of Gaussian structural equation models with equal error variances
- Learning high-dimensional directed acyclic graphs with latent and selection variables
- Model Selection and Estimation in Regression with Grouped Variables
- Model-based boosting 2.0
- On measures of dependence
- On the uniform convergence of empirical norms and inner products, with application to causal inference
- Practical variable selection for generalized additive models
- Score-based causal learning in additive noise models
- Sharp Thresholds for High-Dimensional and Noisy Sparsity Recovery Using $\ell _{1}$-Constrained Quadratic Programming (Lasso)
- Statistics for high-dimensional data. Methods, theory and applications.
- The Adaptive Lasso and Its Oracle Properties
- The adaptive and the thresholded Lasso for potentially misspecified models (and a lower bound for the Lasso)
- The sparsity and bias of the LASSO selection in high-dimensional linear regression
- \(\ell_{0}\)-penalized maximum likelihood for sparse directed acyclic graphs
Cited in
(34)- Efficient learning of nonparametric directed acyclic graph with statistical guarantee
- A review of Gaussian Markov models for conditional independence
- Kernel-based tests for joint independence
- Learning high-dimensional Gaussian linear structural equation models with heterogeneous error variances
- High-dimensional causal discovery under non-Gaussianity
- Invariance, causality and robustness
- Marginal integration for nonparametric causal inference
- scientific article; zbMATH DE number 7370619 (Why is no real title available?)
- Causal discovery in heavy-tailed models
- Structural intervention distance for evaluating causal graphs
- Foundations of structural causal models with cycles and latent variables
- Why BDeu? Regular Bayesian network structure learning with discrete and continuous variables
- High-dimensional Poisson structural equation model learning via \(\ell_1\)-regularized regression
- Causal structure learning: a combinatorial perspective
- Distance metrics for measuring joint dependence with application to causal inference
- Learning causal networks via additive faithfulness
- Nonparametric and high-dimensional functional graphical models
- Efficient Learning of Quadratic Variance Function Directed Acyclic Graphs via Topological Layers
- Likelihood ratio tests for a large directed acyclic graph
- Identification of vector autoregressive models with nonlinear contemporaneous structure
- Causal Dantzig: fast inference in linear structural equation models with hidden variables under additive interventions
- A new causal discovery heuristic
- Causal effect identification in acyclic directed mixed graphs and gated models
- Invariant causal prediction for sequential data
- A survey on causal discovery: theory and practice
- Model selection and local geometry
- CAM: causal additive models, high-dimensional order search and penalized regression
- Causal inference in partially linear structural equation models
- Nonlinear Causal Discovery with Confounders
- Order-independent constraint-based causal structure learning
- Score-based causal learning in additive noise models
- Testing Directed Acyclic Graph via Structural, Supervised and Generative Adversarial Learning
- Identifiability of Gaussian linear structural equation models with homogeneous and heterogeneous error variances
- Max-linear models on directed acyclic graphs
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)
