3 ms·
If A is the event that x is causing y, and B is the event that x and y are correlated, what you are essentially saying is that P(both A and B) > P(A)P(B). There
by pavelrub 13y ago
If A is the event that x is causing y, and B is the event that x and y are correlated, what you are essentially saying is that P(both A and B) > P(A)P(B). There is nothing about the world that makes this statement correct without further assumptions.