Score-based causal learning in additive noise models

DOI10.1080/02331888.2015.1060237zbMATH Open1360.62135arXiv1311.6359OpenAlexW1531392120MaRDI QIDQ5739677FDOQ5739677

Authors: Christopher Nowzohour, Peter Bühlmann 

Publication date: 19 July 2016

Published in: Statistics (Search for Journal in Brave)

Abstract: Given data sampled from a number of variables, one is often interested in the underlying causal relationships in the form of a directed acyclic graph. In the general case, without interventions on some of the variables it is only possible to identify the graph up to its Markov equivalence class. However, in some situations one can find the true causal graph just from observational data, for example in structural equation models with additive noise and nonlinear edge functions. Most current methods for achieving this rely on nonparametric independence tests. One of the problems there is that the null hypothesis is independence, which is what one would like to get evidence for. We take a different approach in our work by using a penalized likelihood as a score for model selection. This is practically feasible in many settings and has the advantage of yielding a natural ranking of the candidate models. When making smoothness assumptions on the probability density space, we prove consistency of the penalized maximum likelihood estimator. We also present empirical results for simulated scenarios and real two-dimensional data sets (cause-effect pairs) where we obtain similar results as other state-of-the-art methods.

Full work available at URL: https://arxiv.org/abs/1311.6359

Recommendations

• scientific article
• Learning causal networks via additive faithfulness
• Distinguishing cause from effect using observational data: methods and benchmarks
• CAM: causal additive models, high-dimensional order search and penalized regression
• High-dimensional causal discovery under non-Gaussianity

zbMATH Keywords

causal inferencelatent variablesstructure learningmaximum likelihood estimatorstructural equation modelshidden variablespath diagramsnonparametric independence tests

Mathematics Subject Classification ID

Point estimation (62F10) Nonparametric estimation (62G05) Nonparametric regression and quantile regression (62G08) Learning and adaptive systems in artificial intelligence (68T05)

Cites Work

• Statistics for high-dimensional data. Methods, theory and applications.
• A study of logspline density estimation
• Identifiability of Gaussian structural equation models with equal error variances
• Theory of function spaces
• 10.1162/153244303321897717
• Bracketing metric entropy rates and empirical central limit theorems for function classes of Besov- and Sobolev-type
• Information-geometric approach to inferring causal directions
• CAM: causal additive models, high-dimensional order search and penalized regression

Cited In (5)

• Title not available (Why is that?)
• CAM: causal additive models, high-dimensional order search and penalized regression
• Marginal integration for nonparametric causal inference
• Title not available (Why is that?)
• Causal inference in partially linear structural equation models

This page was built for publication: Score-based causal learning in additive noise models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5739677)