MugaSofer comments on Confound it! Correlation is (usually) not causation! But why not? - LessWrong

44 Post author: gwern 09 July 2014 03:04AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (34)

Sort By: Best

You are viewing a single comment's thread.

Comment author: MugaSofer 10 July 2014 09:35:26PM 3 points [-]

So, um ... how do we assess the likelihood of causation, assuming we can't conduct an impromptu experiment on the spot?

Comment author: IlyaShpitser 10 July 2014 09:56:21PM 7 points [-]

The keywords are 'causal discovery,' 'structure learning.' There is a large literature.