Ex Matrix Eigen Sol

Exercises: Eigenvalues and Eigenvectors
 
0 0 1
Problem 1. Find all the eigenvalues and eigenvectors of A =  0 1 0 .
1 0 0
Solution. Let λ be an eigenvalue of A. To obtain all possible λ, we solve the characteristic

equation of A (let I be the 3 × 3 identity matrix):
det(A − λI) = 0 ⇒

−λ 0 1

0
1−λ 0 = 0 ⇒
1 0 −λ
(λ − 1)2 (λ + 1) = 0
Hence, A has eigenvalues λ1 = 1 and λ2 = −1.
 
x1
To find all the eigenvectors of λ1 = 1, we need to solve x =  x2  from:
x3
(A − λ1 I)x = 0 ⇒
    
−1 0 1 x1 0
 0 0 0   x2  =  0 
1 0 −1 x3 0
 
x1
The set of solutions to the above equation—EigenSpace(λ1 )—includes all  x2  satisfying
x3
x1 = u
x2 = v
x3 = u
for any u, v ∈ R. Any non-zero vector in EigenSpace(λ1 ) is an eigenvector of A corresponding to
λ1 .
 
x1
Similarly, to find all the eigenvectors of λ2 = −1, we need to solve x =  x2  from:
x3
(A − λ2 I)x = 0 ⇒
    
1 0 1 x1 0
 0 2 0   x2  =  0 
1 0 1 x3 0
 
x1
The set of solutions to the above equation—EigenSpace(λ2 )—includes all  x2  satisfying
x3
x1 = u
x2 = 0
x3 = −u
1
for any u ∈ R. Any non-zero vector in EigenSpace(λ2 ) is an eigenvector of A corresponding to λ2 .
Problem 2. Let A be an n×n square matrix. Prove: A and AT have exactly the same eigenvalues.
Proof. Recall that an eigenvalue of a matrix is a root of the matrix’s characteristic equation,
which equates the matrix’s characteristic polynomial to 0. It suffices to show that the characteristic
polynomial of A is the same as that of AT . In other words, we want to show that det(A − λI) =
det(AT − λI). This is true because A − λI = (AT − λI)T .
Problem 3 (Hard). Let A be an n × n square matrix. Prove: A−1 exists if and only if 0 is not
an eigenvalue of A.
Proof. If-Direction. The objective is to show that if 0 is not an eigenvalue of A, then A−1 exists,
namely, the rank of A is n. Suppose, on the contrary, that the rank of A is less than n. Consider
the linear system Ax = 0 where x is an n × 1 matrix. The hypothesis that rank A < n indicates
that the system has infinitely many solutions. In other words, there exists a non-zero x satisfying
Ax = 0x = 0. This, however, indicates that 0 is an eigenvalue of A, which is a contradiction.
Only-If Direction. The objective is to show that if A−1 exists, then 0 is not an eigenvalue of A.
The existence of A−1 means that the rank of A is n, which in turn indicates that Ax = 0 has a
unique solution x = 0. In other words, there is no non-zero x0 satisfying Ax0 = 0x0 , namely, 0 is
not an eigenvalue of A.
Problem 4. Let A be an n × n square matrix such that A−1 exists. Prove: if λ is an eigenvalue
of A, then 1/λ is an eigenvalue of A−1 .
Proof. Since λ is an eigenvalue of A, there is a non-zero n × 1 matrix x satisfying
Ax = λx ⇒
−1
A Ax = λA−1 x ⇒
x = λA−1 x ⇒
A−1 x = (1/λ)x
which completes the proof.
Problem 5. Prove: if A2 = I, then the eigenvalues of A must be 1 or −1.
Proof. Consider any eigenvalue λ of A, and let x be an arbitrary eigenvector of A corresponding

to λ. Hence, we have:
Ax = λx ⇒
A2 x = λAx ⇒
Ix = λAx ⇒
x = λAx
Note that λ(Ax) = λ(λx) = λ2 x. Hence, we have
x = λ2 x.
As x is not 0, it follows that λ2 = 1, which completes the proof.
2
Problem 6. Suppose that λ1 and λ2 are two distinct eigenvalues of matrix A. Furthermore,
suppose that x1 is an eigenvector of A under λ1 , and that x2 is an eigenvector of A under λ2 .
Prove: there does not exist any real number c such that cx1 = x2 .
Proof. Assume, on the contrary, that such a c exists. Since Ax1 = λ1 x1 , we have A(cx1 ) =
λ1 (cx1 ), which leads to Ax2 = λ1 x2 .
On the other hand, Ax2 = λ2 x2 . Therefore, λ1 = λ2 (remember x2 cannot be 0), giving a
contradiction.
Problem 7. Suppose that λ1 and λ2 are two distinct eigenvalues of matrix A. Furthermore,
suppose that x1 is an eigenvector of A under λ1 , and that x2 is an eigenvector of A under λ2 .
Prove: x1 + x2 is not an eigenvector of A.
Proof. Assume, on the contrary, that x1 + x2 is an eigenvector under some eigenvalue λ3 . This
means that
A(x1 + x2 ) = λ3 (x1 + x2 ) ⇒
Ax1 + Ax2 = λ3 (x1 + x2 ) ⇒
λ1 x1 + λ2 x2 = λ3 (x1 + x2 ) ⇒
(λ1 − λ3 )x1 = (λ3 − λ2 )x2 .
As λ1 6= λ2 , at least one of λ1 − λ3 and λ3 − λ2 is non-zero. Without loss of generality, suppose

λ3 − λ2 6= 0, which gives:
λ1 − λ3
x1 = x2 .
λ3 − λ2
In Problem 6, we already showed that the above is impossible, thus giving a contradiction.

Ex Matrix Eigen Sol

Uploaded by

Copyright:

Available Formats

Ex Matrix Eigen Sol

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ex Matrix Eigen Sol

Uploaded by

Copyright:

Available Formats

Exercises: Eigenvalues and Eigenvectors

Solution. Let λ be an eigenvalue of A. To obtain all possible λ, we solve the characteristic

Proof. Since λ is an eigenvalue of A, there is a non-zero n × 1 matrix x satisfying

which completes the proof.

Problem 5. Prove: if A2 = I, then the eigenvalues of A must be 1 or −1.

Proof. Consider any eigenvalue λ of A, and let x be an arbitrary eigenvector of A corresponding

Note that λ(Ax) = λ(λx) = λ2 x. Hence, we have

As x is not 0, it follows that λ2 = 1, which completes the proof.

As λ1 6= λ2 , at least one of λ1 − λ3 and λ3 − λ2 is non-zero. Without loss of generality, suppose

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.