We can see conditional independence for D and Y given X means E [ D &#x2223;<!-

cooloicons62

cooloicons62

Answered question

2022-07-08

We can see conditional independence for D and Y given X means E [ D Y , X ] = E [ D X ] or E [ Y D , X ] = E [ Y X ] in the causal analysis. However, by Wiki the conditional independence is defined as E [ Y D X ] = E [ Y X ] E [ D X ]   a . e .
I cannot bridge these definitions. Is one of them stronger than the other one?
Thanks a lot.

Answer & Explanation

Alexzander Bowman

Alexzander Bowman

Beginner2022-07-09Added 19 answers

MathJax(?): Can't find handler for document MathJax(?): Can't find handler for document Use the property for sigma algebras G 1 G 2 F ,
E [ E [ X | G 2 | G 1 ] ] = E [ X | G 1 ] .
The σ-algebra generated by X alone is a sub- σ-algebra of the one generated by X , Y, so the above property applies.

Assume that E [ D | X Y ] = E [ D | X ]. Then,
E [ Y D | X ] = E [ E [ Y D | X Y ] | X ] = E [ Y E [ D | X Y ] | X ] = E [ Y E [ D | X ] | X ] = E [ D | X ] E [ Y | X ] .
The first equality follows by the property, the second equality by measurability of Y with respect to σ ( X , Y ), the third by the conditional independence hypothesis, and the fourth by measurability of E [ D | X ] with respect to σ ( X ) .
The reverse direction is similar.

Do you have a similar question?

Recalculate according to your conditions!

Ask your question.
Get an expert answer.

Let our experts help you. Answer in as fast as 15 minutes.

Didn't find what you were looking for?