Can graphical models represent independence besides conditional independence?

Question

In a graphical model, two random variables are conditional independent given their common ancestors.

score 2 · Answer 1 · answered Mar 29 '14 at 18:24

2

Yes, if two random variables are independent, you just have two nodes with no edges between them.

answered Mar 29 '14 at 18:24

queenbee

How would you construct a Bayesian network for the example given by Antonio Valerio Miceli-Barone in his answer to this question? It seems to me that you need an edge from X to B and an edge from A to B. However, A and B are mutually independent. – Franco Aug 15 '18 at 13:56

1 Answers1