So, we all know that correlation does not imply causation. Yet it sounds like this paper does some fancy dynamical systems theory stuff to prove causation based on correlation. Can anyone shed some light on how this works?
Well, it's explained in Methods, but over my head. I gather that instead of just looking at correlation between X and Y, they first map X to a ‘shadow attractor manifold’. Then they fit a relationship between this transformed X and Y.
They cite Sugihara, G. et al. Detecting causality in complex ecosystems. Science 338, 496–500 (2012) for this.
