ParceLiNGAM: A causal ordering method robust against latent confounders
This work addresses the issue of distorted causal order estimation due to latent confounders in causal inference, which is incremental as it builds on existing LiNGAM methods.
The authors tackled the problem of learning causal orders in linear non-Gaussian acyclic models (LiNGAM) by proposing a new algorithm, ParceLiNGAM, that is robust against latent confounders, demonstrating its effectiveness on artificial and simulated brain imaging data.
We consider learning a causal ordering of variables in a linear non-Gaussian acyclic model called LiNGAM. Several existing methods have been shown to consistently estimate a causal ordering assuming that all the model assumptions are correct. But, the estimation results could be distorted if some assumptions actually are violated. In this paper, we propose a new algorithm for learning causal orders that is robust against one typical violation of the model assumptions: latent confounders. The key idea is to detect latent confounders by testing independence between estimated external influences and find subsets (parcels) that include variables that are not affected by latent confounders. We demonstrate the effectiveness of our method using artificial data and simulated brain imaging data.