Consistency of Causal Inference under the Additive Noise Model
This addresses a gap in causal inference research by focusing on consistency rather than soundness, which is incremental but important for theoretical reliability.
The paper tackles the problem of statistical consistency for causal inference methods under the Additive Noise Model, deriving general conditions under which these methods consistently infer causal direction in nonparametric settings.
We analyze a family of methods for statistical causal inference from sample under the so-called Additive Noise Model. While most work on the subject has concentrated on establishing the soundness of the Additive Noise Model, the statistical consistency of the resulting inference methods has received little attention. We derive general conditions under which the given family of inference methods consistently infers the causal direction in a nonparametric setting.