Causal inference by using invariant prediction: identification and confidence intervals. With discussion and authors' reply
From MaRDI portal
DOI10.48550/ARXIV.1501.01332zbMATH Open1414.62297OpenAlexW2790376986WikidataQ61308239 ScholiaQ61308239MaRDI QIDQ153568FDOQ153568
Authors: Jonas Peters, Peter Bühlmann, Nicolai Meinshausen, Jonas Peters, Nicolai Meinshausen, Peter Bühlmann 
Publication date: 6 January 2015
Published in: Journal of the Royal Statistical Society Series B: Statistical Methodology (Search for Journal in Brave)
Abstract: What is the difference of a prediction that is made with a causal model and a non-causal model? Suppose we intervene on the predictor variables or change the whole environment. The predictions from a causal model will in general work as well under interventions as for observational data. In contrast, predictions from a non-causal model can potentially be very wrong if we actively intervene on variables. Here, we propose to exploit this invariance of a prediction under a causal model for causal inference: given different experimental settings (for example various interventions) we collect all models that do show invariance in their predictive accuracy across settings and interventions. The causal model will be a member of this set of models with high probability. This approach yields valid confidence intervals for the causal relationships in quite general scenarios. We examine the example of structural equation models in more detail and provide sufficient assumptions under which the set of causal predictors becomes identifiable. We further investigate robustness properties of our approach under model misspecification and discuss possible extensions. The empirical properties are studied for various data sets, including large-scale gene perturbation experiments.
Full work available at URL: https://arxiv.org/abs/1501.01332
Recommendations
Nonparametric hypothesis testing (62G10) General nonlinear regression (62J02) Linear regression; mixed models (62J05) Multivariate analysis (62Hxx) Research exposition (monographs, survey articles) pertaining to statistics (62-02)
Cited In (63)
- A Two-Sample Conditional Distribution Test Using Conformal Prediction and Weighted Rank Sum
- Higher-Order Least Squares: Assessing Partial Goodness of Fit of Linear Causal Models
- Individualized causal discovery with latent trajectory embedded Bayesian networks
- Data integration in causal inference
- Detecting responsible nodes in differential Bayesian networks
- Corrected score methods for estimating Bayesian networks with error-prone nodes
- Variable selection in linear regression models: choosing the best subset is not always the best choice
- Online feature selection for multi-source streaming features
- From statistical to causal learning
- Statistical Inference for Maximin Effects: Identifying Stable Associations across Multiple Studies
- Causal inference for oncology: past developments and current challenges
- A consistent estimator for confounding strength
- Environment invariant linear least squares
- tramicp
- Estimating causal effects with hidden confounding using instrumental variables and environments
- Sound and complete causal identification with latent variables given local background knowledge
- Title not available (Why is that?)
- Comment: Invariance and causal inference
- Causal network learning with non-invertible functional relationships
- Tests for differential Gaussian Bayesian networks based on quadratic inference functions
- Statistics for big data: a perspective
- Distributionally robust and generalizable inference
- Causal Dantzig: fast inference in linear structural equation models with hidden variables under additive interventions
- Invariant causal prediction for sequential data
- Federated causal inference in heterogeneous observational data
- Conformal Sensitivity Analysis for Individual Treatment Effects
- Causal modelling of heavy-tailed variables and confounders with application to river flow
- Conditional variance penalties and domain shift robustness
- Deconfounding and Causal Regularisation for Stability and External Validity
- Invariance, causality and robustness
- On semiparametric instrumental variable estimation of average treatment effects through data fusion
- Regularized high dimension low tubal-rank tensor regression
- Causal learning via manifold regularization
- Toward Causal Inference for Spatio-Temporal Data: Conflict and Forest Loss in Colombia
- Learning linear non-Gaussian causal models in the presence of latent variables
- The blessings of multiple causes
- Reconstructing regime-dependent causal relationships from observational time series
- Network modeling in biology: statistical methods for gene and brain networks
- Switching regression models and causal inference in the presence of discrete latent variables
- Models as approximations. II. A model-free theory of parametric regression
- Comments on ``Data science, big data and statistics
- Network Structure Learning Under Uncertain Interventions
- Title not available (Why is that?)
- Causal discovery from heterogeneous/nonstationary data
- nonlinearICP
- The Hardness of Conditional Independence Testing and the Generalised Covariance Measure
- Learning stable and predictive structures in kinetic systems
- Likelihood ratio tests for a large directed acyclic graph
- Marginal integration for nonparametric causal inference
- Improved baselines for causal structure learning on interventional data
- Title not available (Why is that?)
- Stabilizing variable selection and regression
- Estimation of Gaussian directed acyclic graphs using partial ordering information with applications to DREAM3 networks and dairy cattle data
- Towards out of distribution generalization for problems in mechanics
- Two-sample instrumental variable analyses using heterogeneous samples
- Discussion: Models as approximations
- Rejoinder: Models as approximations
- Veridical data science
- Beyond the mean: a flexible framework for studying causal effects using linear models
- Rejoinder: Invariance, causality and robustness
- Causal network reconstruction from time series: from theoretical assumptions to practical estimation
- Robustifying independent component analysis by adjusting for group-wise stationary noise
- Learning to increase the power of conditional randomization tests
This page was built for publication: Causal inference by using invariant prediction: identification and confidence intervals. With discussion and authors' reply
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q153568)
