Switching Regression Models and Causal Inference in the Presence of Discrete Latent Variables

Abstract

Given a response $Y$ and a vector $X=(X^1,\dots ,X^d)$ of $d$ predictors, we investigate the problem of inferring direct causes of $Y$ among the vector $X$. Models for $Y$ that use all of its causal covariates as predictors enjoy the property of being invariant across different environments or interventional settings. Given data from such environments, this property has been exploited for causal discovery. Here, we extend this inference principle to situations in which some (discrete-valued) direct causes of $Y$ are unobserved. Such cases naturally give rise to switching regression models. We provide sufficient conditions for the existence, consistency and asymptotic normality of the MLE in linear switching regression models with Gaussian noise, and construct a test for the equality of such models. These results allow us to prove that the proposed causal discovery method obtains asymptotic false discovery control under mild conditions. We provide an algorithm, make available code, and test our method on simulated data. It is robust against model violations and outperforms state-of-the-art approaches. We further apply our method to a real data set, where we show that it does not only output causal predictors, but also a process-based clustering of data points, which could be of additional interest to practitioners.

arXiv comment46 pages, 14 figures; real-world application added in Section 5.2; additional numerical experiments added in the Appendix E

References in corpus (14)

• Estimating high-dimensional intervention effects from observational data
• Distinguishing cause from effect using observational data: methods and benchmarks
• A simple approach for finding the globally optimal Bayesian network structure
• Characterization and Greedy Learning of Interventional Markov Equivalence Classes of Directed Acyclic Graphs
• Causal Inference in the Presence of Latent Variables and Selection Bias
• Bayesian network learning with cutting planes
• Joint Causal Inference from Multiple Contexts
• Nested Markov Properties for Acyclic Directed Mixed Graphs
• Anchor regression: heterogeneous data meets causality
• Learning Sparse Causal Models is not NP-hard
• Learning stable and predictive structures in kinetic systems: Benefits of a causal approach
• Causal Markov condition for submodular information measures
• Causal Inference through a Witness Protection Program
• Identifying Finite Mixtures of Nonparametric Product Distributions and Causal Inference of Confounders

Cited by in corpus (4)

• Causality in cognitive neuroscience: concepts, challenges, and distributional robustness
• Reconstructing regime-dependent causal relationships from observational time series
• Model-based causal feature selection for general response types
• Orthogonal Structure Search for Efficient Causal Discovery from Observational Data