Nonparametric causal structure learning in high dimensions
arXiv:2106.11415 · doi:10.3390/e24030351
Abstract
The PC and FCI algorithms are popular constraint-based methods for learning the structure of directed acyclic graphs (DAGs) in the absence and presence of latent and selection variables, respectively. These algorithms (and their order-independent variants, PC-stable and FCI-stable) have been shown to be consistent for learning sparse high-dimensional DAGs based on partial correlations. However, inferring conditional independences from partial correlations is valid if the data are jointly Gaussian or generated from a linear structural equation model -- an assumption that may be violated in many applications. To broaden the scope of high-dimensional causal structure learning, we propose nonparametric variants of the PC-stable and FCI-stable algorithms that employ the conditional distance covariance (CdCov) to test for conditional independence relationships. As the key theoretical contribution, we prove that the high-dimensional consistency of the PC-stable and FCI-stable algorithms carry over to general distributions over DAGs when we implement CdCov-based nonparametric tests for conditional independence. Numerical studies demonstrate that our proposed algorithms perform nearly as good as the PC-stable and FCI-stable for Gaussian distributions, and offer advantages in non-Gaussian graphical models.
References in corpus (10)
- Estimating high-dimensional directed acyclic graphs with the PC-algorithm
- Kernel-based Conditional Independence Test and Application in Causal Discovery
- Variable selection with error control: Another look at Stability Selection
- Learning high-dimensional directed acyclic graphs with latent and selection variables
- Geometry of the faithfulness assumption in causal inference
- High-dimensional learning of linear causal networks via inverse covariance estimation
- Markov equivalence for ancestral graphs
- High-Dimensional Gaussian Graphical Model Selection: Walk Summability and Local Separation Criterion
- The Reduced PC-Algorithm: Improved Causal Structure Learning in Large Random Networks
- On Distance and Kernel Measures of Conditional Independence