For OLS in matrix form, we are taught that Hat matrix is $X(X^TX)^-X^T$, and is idempotent etc, i.e. when it multiplies with itself, it will self cancel and thus lead back to the same Hat matrix.

I was wondering why can't the inverse in the middle just unravel itself, thus causing two self cancellations (due to X multiplying with its inverse, likewise for $X^T$), but this self-implosion can only occur when we mutliply the Hat onto something else (like when we multiply it with itself, for e.g.)

  • 2
    $\begingroup$ Could you tell us what you mean by the "inverse" of a non-square matrix $X$? $\endgroup$
    – whuber
    Commented Dec 10, 2020 at 16:11
  • $\begingroup$ Ah... I see now. Thank you so much, clearly need to brush up my linear algebra... $\endgroup$
    – jojorabbit
    Commented Dec 11, 2020 at 7:27

1 Answer 1


As you observed, the inverse is a distributive operation, thus in principle it could be applied separately to terms in a product.

However, as whuber mentioned in comments, the actual inverse operation is not defined for non-square matrices.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.