0
$\begingroup$

I am missing something about the last step of the variance of the OLS estimator for $b$. I understand up to the point when $\sigma^2$ is derived due to errors being mutually independent. However, when we go from this:

$V(b)=\sigma²(X'X)^{-1}X'X(X'X)^{-1}$

to

$V(b)=\sigma²(X'X)^{-1}$

Could you help me figuring out which properties of $X$ or any matrix whatsoever, make that substitution possible?

$\endgroup$

1 Answer 1

0
$\begingroup$

In this step you're simply canceling $(X'X)$ and $(X'X)^{-1}$ to get an identity matrix.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .