Ancestor regression in linear structural equation models

05/18/2022
by   Christoph Schultheiss, et al.
0

We present a new method for causal discovery in linear structural equation models. We propose a simple "trick" based on statistical testing in linear models that shall distinguish between ancestors and non-ancestors of any given variable. Naturally, this can then be extended to estimating full graphs. Unlike many methods, we provide explicit error control for false causal discovery, at least asymptotically. This holds true even under Gaussianity where various methods fail due to non-identifiable structures.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset