And its not just causation. There are often unaccounted variables that are driving both the cause and effect and observing that decorrelates the cause and effec

And its not just causation. There are often unaccounted variables that are driving both the cause and effect and observing that decorrelates the cause and effect completely. There is some fascinating research going on on this in the field of economics. Matt Taddy, University of Chicago, Microsoft Research is doing some interesting research on this very kind of problem.