Skip to main content
MathOverflow

Return to Answer

added 3 characters in body
Source Link
Glorfindel
Glorfindel
  • 2.8k
  • 6
  • 28
  • 38

Suppose that $\sigma_{\mathrm{ess}}(T)=\{0\}$. So, zero is the only one possible accumulation point.

Thus if $\sigma_d(T)$ is finite then $E(\{\lambda\in\mathbb{R}:|\lambda|>\varepsilon\})$ has finite range because it is the sum of projections on finite dimensional subespacessubspaces of $H$.

In the same way, if $\sigma_d(T)$ is infinite then $E(\{\lambda\in\mathbb{R}:|\lambda|>\varepsilon\})$ has finite range because $\{\lambda:|\lambda|>\varepsilon\}$ contains only finite eigenvalues ​​of finite multiplicity because $T$ is bounded.

Suppose that $\sigma_{\mathrm{ess}}(T)=\{0\}$. So, zero is the only one possible accumulation point.

Thus if $\sigma_d(T)$ is finite then $E(\{\lambda\in\mathbb{R}:|\lambda|>\varepsilon\})$ has finite range because it is sum of projections on finite dimensional subespaces of $H$.

In the same way, if $\sigma_d(T)$ is infinite then $E(\{\lambda\in\mathbb{R}:|\lambda|>\varepsilon\})$ has finite range because $\{\lambda:|\lambda|>\varepsilon\}$ contains only finite eigenvalues ​​of finite multiplicity because $T$ is bounded.

Suppose that $\sigma_{\mathrm{ess}}(T)=\{0\}$. So, zero is the only one possible accumulation point.

Thus if $\sigma_d(T)$ is finite then $E(\{\lambda\in\mathbb{R}:|\lambda|>\varepsilon\})$ has finite range because it is the sum of projections on finite dimensional subspaces of $H$.

In the same way, if $\sigma_d(T)$ is infinite then $E(\{\lambda\in\mathbb{R}:|\lambda|>\varepsilon\})$ has finite range because $\{\lambda:|\lambda|>\varepsilon\}$ contains only finite eigenvalues ​​of finite multiplicity because $T$ is bounded.

added 4 characters in body
Source Link
Andrés Felipe
Andrés Felipe
  • 173
  • 4

Suppose that $\sigma_{\mathrm{ess}}(T)=\{0\}$. So, zero is the only one possible accumulation point.

Thus if $\sigma_d(T)$ is finite then $E({\lambda\in\mathbb{R}:|\lambda|>\varepsilon})$$E(\{\lambda\in\mathbb{R}:|\lambda|>\varepsilon\})$ has finite range because it is sum of projections on finite dimensional subespaces of $H$.

In the same way, if $\sigma_d(T)$ is infinite then $E({\lambda\in\mathbb{R}:|\lambda|>\varepsilon})$$E(\{\lambda\in\mathbb{R}:|\lambda|>\varepsilon\})$ has finite range because $\{\lambda:|\lambda|>\varepsilon\}$ contains only finite eigenvalues ​​of finite multiplicity because $T$ is bounded.

Suppose that $\sigma_{\mathrm{ess}}(T)=\{0\}$. So, zero is the only one possible accumulation point.

Thus if $\sigma_d(T)$ is finite then $E({\lambda\in\mathbb{R}:|\lambda|>\varepsilon})$ has finite range because it is sum of projections on finite dimensional subespaces of $H$.

In the same way, if $\sigma_d(T)$ is infinite then $E({\lambda\in\mathbb{R}:|\lambda|>\varepsilon})$ has finite range because $\{\lambda:|\lambda|>\varepsilon\}$ contains only finite eigenvalues ​​of finite multiplicity because $T$ is bounded.

Suppose that $\sigma_{\mathrm{ess}}(T)=\{0\}$. So, zero is the only one possible accumulation point.

Thus if $\sigma_d(T)$ is finite then $E(\{\lambda\in\mathbb{R}:|\lambda|>\varepsilon\})$ has finite range because it is sum of projections on finite dimensional subespaces of $H$.

In the same way, if $\sigma_d(T)$ is infinite then $E(\{\lambda\in\mathbb{R}:|\lambda|>\varepsilon\})$ has finite range because $\{\lambda:|\lambda|>\varepsilon\}$ contains only finite eigenvalues ​​of finite multiplicity because $T$ is bounded.

Source Link
Andrés Felipe
Andrés Felipe
  • 173
  • 4

Suppose that $\sigma_{\mathrm{ess}}(T)=\{0\}$. So, zero is the only one possible accumulation point.

Thus if $\sigma_d(T)$ is finite then $E({\lambda\in\mathbb{R}:|\lambda|>\varepsilon})$ has finite range because it is sum of projections on finite dimensional subespaces of $H$.

In the same way, if $\sigma_d(T)$ is infinite then $E({\lambda\in\mathbb{R}:|\lambda|>\varepsilon})$ has finite range because $\{\lambda:|\lambda|>\varepsilon\}$ contains only finite eigenvalues ​​of finite multiplicity because $T$ is bounded.