9.2) OLS Matrix Notation

[@davidcriss9014]

Yes, this expression is very clear to me now. Thank you for drawing it out!

[@fayezsalame7134]

A well detailed explanation regarding OLS. A lot to digest but for the most part, it is relatively simple mathematics. The multiple figures and lines might be confusing at first but the step by step process you show makes it simple to understand. I am seeing the importance of randomized experiments and how it mathematically affects the regressions. This is very interesting to see in such detail

[@causaldeeplearning4738]

Matrix notation is very helpful to get the nuances of many estimators. Many properties and derivations are too hard to see using scalar notation. All advanced books in Econometrics are written using Linear Algebra.

[@pedrocolangelo5844]

Such a good explanation in just 5:40. You, sir, are amazing.

[@causaldeeplearning4738]

Thanks Pedro!!

[@kimberlykrafft9548]

Very clear explanation of the error term and how mathematically it's in the formula that it must not be correlated to x

[@causaldeeplearning4738]

You got it!

[@rheak.5470]

at 1:56, what are the steps to get to RSS(b) in matrix form?

[@mohssenify]

very well explained keep up the good work

[@Livcalona]

this was amazingly explained, thank you so much!

[@causaldeeplearning4738]

Glad you enjoyed it!

[@joebucket1471]

Good Explanation, but at 2:57 I don't understand why we are allowed to do this. Edit: It's because of this identity: C'B'A' = (ABC)' or (C'B'A) = (A'BC)' and the fact that y'Xb is a scalar and the transpose of a scalar is the scalar itself: so b'X'y = (y'Xb)' = y'Xb