Is there an example such that $A$ and $B$, three by three, that have the same eigenvectors, but different eigenvalues?

What would be the eigenvectors and eigenvalues if it exists because I'm stuck on this practice problem.

I know that if matrices $A$ and $B$ can be written such that $AB=BA$, they share the same eigenvectors, but what about their eigenvalues? precisely if they're squared matrices ($3\times 3$ case)


How about $I$ and $-I$? Then all $x\neq0$ are eigenvectors. But the eigenvalues are $1$ and $-1$ respectively.

For a less trivial example, how about $\begin{pmatrix}1&0&0\\0&2&1\\0&0&2\end{pmatrix}$ and $\begin{pmatrix}4&0&0\\0&3&1\\0&0&3\end{pmatrix}$?

They share eigenvectors $(1,0,0)$ and $(0,1,0)$, but for different eigenvalues.

As to your question about commuting matrices, you could take any two diagonal matrices, with different entries on the respective diagonals.

  • 1
    $\begingroup$ It only works for the identity matrices? $\endgroup$ – xim Jan 29 at 3:04
  • $\begingroup$ Hmm. I'll have to think about it. It looks like multiples of identity matrices will work. How about a permutation matrix? $\endgroup$ – Chris Custer Jan 29 at 3:33
  • $\begingroup$ It would work if you put different values on the diagonal. Maybe using Jordan normal form we can make examples. $\endgroup$ – Chris Custer Jan 29 at 3:55

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.