Correlation and causality in economic data

- [Instructor] You know the question,…if a tree falls in the forest…and there's no one there to hear it,…does it make a sound?…Well, there's a similar kind of question…for data scientists,…and it goes like this,…if an algorithm finds a correlation…but there's no one there to see it,…does it still lead to a causal inference?…There's a few reasons to ask this.…Number one is because we all know…correlation doesn't imply causation.…Just because two things are correlated,…doesn't mean that one causes the other.…This is a dictum that's been hammered in…for a long time,…but another reason is this,…algorithms don't make inferences about causation,…people do.…

Machine learning algorithms simply find consistencies…or patterns in the data,…making the leap to causation is a psychological process.…Now, let me tell you why this matters,…there's a couple of reasons.…Number one, it actually may turn out…that correlation is sufficient for your purposes…and maybe all you need is an association between variables.…In fact, correlation under the proper circumstances…