Abstract
We consider linear models where d potential causes X1 , . . . , Xd are correlated with one targ quantity Y and propose a method to infer whether the association is causal or whether it is an artif caused by overfitting or hidden common causes. We employ the idea that in the former case the vector of regression coefficients has ‘generic’ ori entation relative to the covariance matrix of X. Using an ICA based model for confounding, we show that both confounding and overfitting yield regression vectors that concentrate mainly in the space of low eigenvalues of .