-1
$\begingroup$

Let $T$ be a self-adjoint invertible operator on $\mathcal{H}$ with a continuous spectrum, means the spectral measure is nonatomic. For which class of invertible operators $V$( with continuous spectrum) other than unitary $VTV^{*}$ also have continuous spectrum?

$\endgroup$
4
  • 2
    $\begingroup$ Obviously it works for nonzero scalar multiples of unitaries. $\endgroup$ Commented Apr 29, 2019 at 12:31
  • $\begingroup$ I dont want that @Israel!! Something nontrivial I am expecting $\endgroup$
    – user136400
    Commented Apr 29, 2019 at 12:32
  • 4
    $\begingroup$ "Something nontrivial I am expecting" -- you should not expect other people to do the work of formulating a well-defined question for you. $\endgroup$
    – Yemon Choi
    Commented Apr 29, 2019 at 15:16
  • $\begingroup$ @Yemon I formulated the correct question, somebody answered it in the wrong way, I made him understand the mistake!! $\endgroup$
    – user136400
    Commented Apr 30, 2019 at 7:25

2 Answers 2

5
$\begingroup$

In any infinite-dimensional Hilbert space, the only such operators $V$ that work for all $T$ are scalar multiples of the identity.

Suppose $V$ is not a scalar multiple of a unitary. Then there are linearly independent vectors $v$, $w$ such that $V^* V v = w$. Let $T$ be a self-adjoint invertible operator such that $Tw = v$. Then $VTV^* (Vv) = V T w = V v$, so $V T V^*$ does not have continuous spectrum. All that remains is to show that such $T$ can be chosen with continuous spectrum, which is not difficult.

$\endgroup$
0
$\begingroup$

The spectrum of an operator $T$ is invariant under conjugation by invertible operators $V$, i.e. $$\sigma(T)=\sigma(VTV^{-1}).$$ So, if your operator has continuous spectrum also its conjugate $VTV^{-1}$ does, because the spectral measures are the same.

$\endgroup$
2
  • $\begingroup$ yes I know $\sigma(T)=\sigma(VTV^{-1})$, but my question about $VTV^{*}$ not $VTV^{-1}$ $\endgroup$
    – user136400
    Commented Apr 29, 2019 at 13:11
  • $\begingroup$ Ok, so perhaps you may want to slightly rephrase your question so that it is clearer. $\endgroup$
    – John
    Commented Apr 29, 2019 at 13:16

You must log in to answer this question.

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