Inferring Causal Direction from Observational Data: A Complexity Approach
This addresses a fundamental challenge in causal inference for researchers and practitioners, though it appears incremental as it builds on existing simplicity-based approaches.
The paper tackles the problem of inferring causal direction from observational data by proposing fast and simple criteria based on the idea that predicting the effect from the cause is simpler than the reverse, and demonstrates their accuracy on synthetic data across various causal mechanisms and noise types.
At the heart of causal structure learning from observational data lies a deceivingly simple question: given two statistically dependent random variables, which one has a causal effect on the other? This is impossible to answer using statistical dependence testing alone and requires that we make additional assumptions. We propose several fast and simple criteria for distinguishing cause and effect in pairs of discrete or continuous random variables. The intuition behind them is that predicting the effect variable using the cause variable should be `simpler' than the reverse -- different notions of `simplicity' giving rise to different criteria. We demonstrate the accuracy of the criteria on synthetic data generated under a broad family of causal mechanisms and types of noise.