Analysis of cause-effect inference by comparing regression errors

Abstract

We address the problem of inferring the causal direction between two variables by comparing the least-squares errors of the predictions in both possible directions. Under the assumption of an independence between the function relating cause and effect, the conditional noise distribution, and the distribution of the cause, we show that the errors are smaller in causal direction if both variables are equally scaled and the causal relation is close to deterministic. Based on this, we provide an easily applicable algorithm that only requires a regression in both possible causal directions and a comparison of the errors. The performance of the algorithm is compared with various related causal inference methods in different artificial and real-world data sets.

arXiv commentThis is an extended version of the AISTATS 2018 paper

References in corpus (7)

• Distinguishing cause from effect using observational data: methods and benchmarks
• On the Identifiability of the Post-Nonlinear Causal Model
• Causal Inference on Discrete Data using Additive Noise Models
• Inferring deterministic causal relations
• Telling Cause from Effect using MDL-based Local and Global Regression
• Distinguishing Cause and Effect via Second Order Exponential Models
• Error Asymmetry in Causal and Anticausal Regression

Cited by in corpus (2)

• Latent Instrumental Variables as Priors in Causal Inference based on Independence of Cause and Mechanism
• On the Role of Entropy-based Loss for Learning Causal Structures with Continuous Optimization