Causal Dantzig: fast inference in linear structural equation models with hidden variables under additive interventions

DOI10.1214/18-AOS1732zbMATH Open1420.62309arXiv1706.06159OpenAlexW2964243609WikidataQ128374405 ScholiaQ128374405MaRDI QIDQ2414102FDOQ2414102

Authors: Dominik Rothenhäusler, Nicolai Meinshausen, Peter Bühlmann 

Publication date: 10 May 2019

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

Abstract: Causal inference is known to be very challenging when only observational data are available. Randomized experiments are often costly and impractical and in instrumental variable regression the number of instruments has to exceed the number of causal predictors. It was recently shown in Peters et al. [2016] that causal inference for the full model is possible when data from distinct observational environments are available, exploiting that the conditional distribution of a response variable is invariant under the correct causal model. Two shortcomings of such an approach are the high computational effort for large-scale data and the assumed absence of hidden confounders. Here we show that these two shortcomings can be addressed if one is willing to make a more restrictive assumption on the type of interventions that generate different environments. Thereby, we look at a different notion of invariance, namely inner-product invariance. By avoiding a computationally cumbersome reverse-engineering approach such as in Peters et al. [2016], it allows for large-scale causal inference in linear structural equation models. We discuss identifiability conditions for the causal parameter and derive asymptotic confidence intervals in the low-dimensional setting. In the case of non-identifiability we show that the solution set of causal Dantzig has predictive guarantees under certain interventions. We derive finite-sample bounds in the high-dimensional setting and investigate its performance on simulated datasets.

Full work available at URL: https://arxiv.org/abs/1706.06159

Recommendations

• Robust Causal Structure Learning with Some Hidden Variables
• Causal inference by using invariant prediction: identification and confidence intervals. With discussion and authors' reply
• CAM: causal additive models, high-dimensional order search and penalized regression
• Causal statistical inference in high dimensions
• High-dimensional learning of linear causal networks via inverse covariance estimation

zbMATH Keywords

causal inferenceconfidence intervalsstructural equation modelshigh-dimensional consistencyDantzig estimator

Mathematics Subject Classification ID

Asymptotic properties of nonparametric inference (62G20) Nonparametric tolerance and confidence regions (62G15) Estimation in multivariate analysis (62H12) Ridge regression; shrinkage estimators (Lasso) (62J07)

Cited In (16)

• Causal inference by using invariant prediction: identification and confidence intervals. With discussion and authors' reply
• Estimating causal effects with hidden confounding using instrumental variables and environments
• Title not available (Why is that?)
• Invariant causal prediction for sequential data
• Estimating bounds on causal effects in high-dimensional and possibly confounded systems
• Invariance, causality and robustness
• Discussion of the Paper “Prediction, Estimation, and Attribution” by B. Efron
• Discussion of the Paper “Prediction, Estimation, and Attribution” by B. Efron
• Estimating exogenous variables in data with more variables than observations
• Switching regression models and causal inference in the presence of discrete latent variables
• Likelihood ratio tests for a large directed acyclic graph
• CAM: causal additive models, high-dimensional order search and penalized regression
• Marginal integration for nonparametric causal inference
• Environment invariant linear least squares
• Improved baselines for causal structure learning on interventional data
• Estimating high-dimensional intervention effects from observational data

This page was built for publication: Causal Dantzig: fast inference in linear structural equation models with hidden variables under additive interventions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2414102)