Show that A and its transpose have the same eigenvalues

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
4 replies · 3K views
Mr Davis 97
Messages
1,461
Reaction score
44

Homework Statement


Show that ##A## and ##A^T## have the same eigenvalues.

Homework Equations

The Attempt at a Solution


If they have the same eigenvalues, then ##Ax = \lambda x## iff ##A^T x = \lambda x##. In other words, we have to show that ##|A - \lambda I| = 0## iff ##|A^T - \lambda I| = 0##. From the properties of transpose, we see that ##(A - \lambda I)^T = A^T - \lambda I##. It's a property of transposes that ##A^T## is invertible iff ##A## is also invertible. So we have shown that ##A - \lambda I## is invertible iff ##A^T - \lambda I## is also invertible. So this shows that they have the same eigenvalues.

I think that this is the correct solution, but I am a little confused about the beginning part of the proof. Does this imply that A and its transpose also have the same eigenvectors? Or do they have the same eigenvlues but different eigenvectors?
 
Physics news on Phys.org
Mr Davis 97 said:

Homework Statement


Show that ##A## and ##A^T## have the same eigenvalues.

Homework Equations

The Attempt at a Solution


If they have the same eigenvalues, then ##Ax = \lambda x## iff ##A^T x = \lambda x##. In other words, we have to show that ##|A - \lambda I| = 0## iff ##|A^T - \lambda I| = 0##. From the properties of transpose, we see that ##(A - \lambda I)^T = A^T - \lambda I##. It's a property of transposes that ##A^T## is invertible iff ##A## is also invertible. So we have shown that ##A - \lambda I## is invertible iff ##A^T - \lambda I## is also invertible. So this shows that they have the same eigenvalues.

I think that this is the correct solution, but I am a little confused about the beginning part of the proof. Does this imply that A and its transpose also have the same eigenvectors? Or do they have the same eigenvlues but different eigenvectors?
You don't need to consider invertibility here. And you have already said everything you need. Just clean it up a little.
What are the eigenvalues of ##A##?
What are the eigenvalues of ##A^T##?
And how are ##\det B## and ##\det B^T## related for any matrix ##B##? (Esp. for ##B=A-\lambda I##.)
 
fresh_42 said:
You don't need to consider invertibility here. And you have already said everything you need. Just clean it up a little.
What are the eigenvalues of ##A##?
What are the eigenvalues of ##A^T##?
And how are ##\det B## and ##\det B^T## related for any matrix ##B##? (Esp. for ##B=A-\lambda I##.)
##det(A) = det(A^T)## for all square ##A##. So we have that ##det(A - \lambda I) = det((A - \lambda I)^T) = det(A^T - \lambda I)##
 
  • Like
Likes   Reactions: fresh_42
fresh_42 said:
You don't need to consider invertibility here. And you have already said everything you need. Just clean it up a little.
What are the eigenvalues of ##A##?
What are the eigenvalues of ##A^T##?
And how are ##\det B## and ##\det B^T## related for any matrix ##B##? (Esp. for ##B=A-\lambda I##.)
So as a last note do ##A## and ##A^T## also have the same eigenvectors?