2
$\begingroup$

I feel like self-adjoint / Hermitian operators are the "best" operators, since an operator that is self-adjoint can be orthogonally diagonalized, according to the Spectral Theorem (over the complex number field, an operator only needs to be normal to be orthogonally diagonalizable).

I know that self-adjoint implies normal, but does self-adjoint also imply orthogonal / unitary: $AA^T = A^TA = I$?

Thanks,

$\endgroup$
2
  • 2
    $\begingroup$ No. Consider $\begin{bmatrix} 2 & 0 \\ 0 & 1\end{bmatrix}$. This matrix is self-adjoint but not orthogonal. It is also untrue that Hermitian matrices or necessarily unitary. $\endgroup$
    – user137731
    Dec 2, 2014 at 3:31
  • $\begingroup$ Ok, thanks so much, @Bye_World :) $\endgroup$
    – User001
    Dec 2, 2014 at 4:06

1 Answer 1

3
$\begingroup$

Definitely not. A self-adjoint operator just need $A^T=A$. Note the matrix of $T^*$ is the transposed of the matrix of $T$.

$\endgroup$
1
  • $\begingroup$ Got it - thanks@user193702. $\endgroup$
    – User001
    Dec 2, 2014 at 4:06

You must log in to answer this question.

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