Super learner
This user is a Super Learner. To become a Super Learner, you need to reach Level 8.
Last answered:

31 Oct 2022

Posted on:

05 Apr 2022

0

Linear regression implies causality?

Linear regression is not supposed to imply causality, why in this video do they say yes?
I'm very confused

3 answers ( 0 marked as helpful)
Instructor
Posted on:

05 Apr 2022

1

Hey Luis,

Thank you for your question!

Indeed neither correlation, nor regression necessarily imply causation. When faced with a problem, very careful analysis needs to be performed to determine the true cause-and-effect relationship.

This lesson aims at telling the difference between correlation and regression, which could be a subtle one.

Kind regards,
365 Hristina

Super learner
This user is a Super Learner. To become a Super Learner, you need to reach Level 8.
Posted on:

05 Apr 2022

0

my confusion is because in the lessons and in the 365 booklet it says that linear regression is used for causal relationships, why? causal and causality is different?

Posted on:

31 Oct 2022

0

I think Luis,  one could make a regression with two totally different and unrelated variables Y and a X and still get a Beta coefficient. So regression does not mean causality, the causality remains in the theory or context behind the model.

Submit an answer