Consider the zero matrix $M=o$.

Is it correct to say that $M$ then has no eigenvalues and eigenvectors? A natural guess for a candidate would be $\lambda=0$. It solves the characteristic equation $\det (M-\lambda \mathbb I)=0$. But there is no associated eigenvector, that is a nonzero vector $v$ such that: \begin{equation} Mv=\lambda v=0 \end{equation} Hence no eigenvectors and no eigenvalues? Or would one say that 0 is an eigenvalue without a corresponding eigenvector?

  • 2
    $\begingroup$ Why do you say there is no nonzero $ v $ such that $ Mv = 0 $? $\endgroup$ – user571438 Jul 1 '18 at 17:17
  • 1
    $\begingroup$ quite right. I'm not seeing the forest for the trees. Eigenvectors everywhere :) I knew this question was embarrassing, but fortunately I'm not afraid of embarrassment. $\endgroup$ – Marlo Jul 1 '18 at 17:22

All nonzero vectors are eigenvectors, since all vectors $v$ satisfy $Mv=0v$ if $M$ is the zero matrix.

  • $\begingroup$ Thanks @Hans. I can't believe how I have not seen this :) $\endgroup$ – Marlo Jul 1 '18 at 17:24

The equation $Mv=\lambda v$ is satisfied by $\lambda =0$ and every non-zero vector.

Thus you have an eigenvalue of $0$ and all non-zero vectors as your eigenvectors.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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