MATB24 Linear Algebra II Assignment 3 — Solutions

Prof. T. Kielstra, Dept. of CMS, University of Toronto, Scarborough

Instructions:
ˆ This assignment is for self work and is worth 0 percent of you final grade.
ˆ These questions are to give you an opportunity to engage with the material of the course.
ˆ By looking at the course outline you can see when your TA will review this material in lab.
ˆ The TA will not review the entire assignment. They will specifically focus on questions that students seem to be
struggling with.
ˆ You should attempt the questions on this assignment before you go to the tutorial.

1. For the following matrix. What are the characteristic values over the field
(a) C
(b) R
(c) Z5
 
2 0 0
 2 3 3 
4 2 2

Solution:
The characteristic polynomial is (x − 2)(x − 5)(x). Therefore the characteristics of the complex numbers and
the reals are x3 − 7x21 0x. Over Z5 the characteristics are 2 and 0.

2. For the following matrix. Compute the minimal polynomial over the field
(a) C
(b) R
(c) Z5
 
2 0 0
 2 3 3 
4 2 2

Solution:
From the previous question, we know that the characterisitic polynomial is (x − 2)(x − 5)(x), which using a
theory from class is the minimal polynomial. For Z5 , we must check (x + 3)x
 
0 0 0
(A + 3)(A) =  2 4 4  ̸= 0.
2 1 1

Therefore, the minimal polynomial is x3 + 3x2 .

3. For the following matrix. Determine if c = 1 is a characteristic over the field. If so, calculate its characteristic
space.

(a) C
(b) R
(c) Z2

1
  
0 0 1
 1 0 1 
0 1 1
Solution:
The characteristic polynomial is x3 − x2 − x − 1. f (1) = −2 = 0(mod 2). Therefore it is only a characteristic
over Z2 , thus c = 1. So we have to solve for a basis for the nullspace of A + 1.

   
1 0 1 0 1 0 1 0
A+1= 1 1 1 0 = 0 1 0 0 
0 1 1 0 0 1 1 0
   
1 0 1 0 1 0 0 0
= 0 1 0 0 = 0 1 0 0 
0 0 1 0 0 0 1 0
Therefore, the standard basis is a basis for the characteristic space corresponding to c = 1.
4. (6.2.3) Let A be an n × n triangular matrix over the field F. Prove that the characteristic values of A are the
P n
Q
diagonal entries of A. (Hint: Leibniz expansion should help det(A) = sign(σi ) Ai,σj (i) ).
σj ∈Sn i=1

Solution:

Note that A − λI is triangular. Use that for a triangular matrix T

n
Y
det T = Tii
i=1

To prove the 1st statement use co-factors or the Leibniz expansion

X n
Y
det A = (−1)σ Ai,σ(i)
σ∈Sn i=1

If A is triangular, then if σi (j) < j, then aj,sigmai (j) = 0. For all permutations σj accept the identity
permutation, there exists an index j such that sigmaj < j. As all the products for permutations except the
identity permutation,

X n
Y n
Y n
Y
sign(σi ) Ai,σj (i) = (−1)0 Ai,i + 0 = Ai,i ,
σj ∈Sn i=1 i=1 i=1

As A is triangular, xI − A is also triangular. Therefore

n
Y
det(xI − A) = x − aii
i=1

and so the eigenvalues are ai,i , 1 ≤ i ≤ n.

5. (6.2.7) Let T be a linear operator on an n- dimensional vector space A, and let T has n distinct characteristic
values. Prove that T is diagonalizable.
As T is an n-dimensional vector space, the characteristic polynomial will be degree n. The n distinct charac-
teristic values will all be factors of the characteristic polynomial by definition. As an n dimensional polynomial
has n roots, the characteristic polynomial will by
n
Y
f (x) = (x − c1 ) · · · (x − cn ) = (x − ci )di ,
i=1

where di = 1 for all i.

As the values ci are characteristics and di = 1 for all i, the dim(nullspace(T − ci )) = di for all i. By a theorem
from class, we are differentiable.

2
 6. (6.2.10) Suppose that A is a symmetric 2 × 2 real matrix. Prove that A is diagonalizable.
Solution:  
a b
Let A = , wherea, b, c ∈ R. Then,
b c

det(xI − A) = (x − a)(x − c) − b2
= x2 − (a + c) + ac − b2 .

The roots of the expression are:

p
a+c± (a + c)2 − 4ac + b2
c=
p 2
a + c ± (a2 + 2ac + c2 − 4ac + b2
=
p 2
a + c ± (a − 2ac + c2 + b2
2
=
p 2
a + c ± (a − c)2 + b2
= .
2

Therefore, A has real characteristic values. Unless b = 0 and c = a, there are two different characteristic values
and by the previous question is diagonalizable. If b = 0 the matrix is already a diagonal matrix and thus is
diagonalizable.

7. (6.2.11) Let N be a 2 × 2 complex matrix such that N 2 = 0. Prove that either N = 0 or N is similar over C to
 
0 0
1 0

Solution:
Long calculus argument. You can find the solution if you look.

8. (6.2.12) Use the result of the previous question to prove the following: If A is a 2 × 2 matrix with complex
entries, then A is similar over C to one of the following two matrices.
   
a 0 a 0
0 b 1 a

Solution:
Long calculus argument. You can find the solution if you look.
9. (6.3.1) Let V be a finite-dimensional vector space. What is the minimal polynomial for:

(a) the identity operator

(b) the zero operator
on V ?
Solution:

(a) Let β be the standard basis. [I]β = I. I − 1I = 0. Therefore, x − 1 is the minimal polynomial.
(b) Let β be the standard basis. [O]β = 0. Therefore, x is the minimal polynomial.
10. (6.3.2) Let a, b and c be elements of a field F and let A be the following 3 × 3 matrix over F:
 
0 0 c
A= 1 0 b 
0 1 a

3
 Prove that the characteristic polynomial for A is x3 − ax2 − bx − c and the this is also the minimal polynomial.
(Bonus: extend this concept to polynomials of general degree.)
Solution:
Check that det λI − A = λ3 − aλ2 − bλ − c.
Let c1 , c2 , c3 be the characteristic values of A.
 
−ci 0 c
A − ci =  1 −ci b = ̸ 0.
0 1 c − ci

Therefore, no linear function will annihilate A.

  
−c1 0 c −c2 0 c
(A − ci )(A − cj ) =  1 −c1 b  1 −c2 b 
0 1 c − c1 0 1 c − c2
 
c1 c2 c c(c − c1 − c2 )
=  −c2 − c1 b + c1 c2 c − bc2 + b(c − c2 )  ̸= 0
1 c − c1 − c2 b − (c − c1 )(c − c2 )

Therefore, no quadratic function will annihilate A. Thus, the minimal polynomial must be a cubic function.

11. (6.3.7) Let n be a positive integer, and let V be the space of polynomials over R which have degree at most n
(include the zero polynomial). Let D be the differentiation operator on V . What is the minimal polynomial
for D?
Solution:
n
ai x0 . Dn+1 p(x) = 0, therefore the minimal polynomial divides f (x) = xn+1 . Let p(x) = xn ,
P
Let p(x) =
i=0
then Dn (p) = n! ̸= 0, Therefore, the minimal polynomial is f (x) = xn+1 .
12. (6.3.8) Let P be the operator on R2 which projects each vector onto the x-axis, parallel to the y-axis. Show
that P is linear. What is the minimal polynomial for P ?
Solution:
The proof that shows the function is linear transformation is something we have done a lot up to this point. It
should be straight forward.
Now we can compute P in the standard matrix.
 
1 0
P =
0 0

The characteristic polynomial is (x − 1)x. Note that x ̸= 0 and x − 1 ̸= 0. Therefore the minimal polynomial
is x(x − 1).
13. (6.4.1) Let T be the linear operator on R2 , the matrix of which in the standard ordered basis is
 
1 −1
A= (1)
2 2

(a) Prove that the only subspaces of R2 invariant under T are R2 and the zero subspace.
(b) If U is the linear operator on C2 , the matrix of which in the standard ordered basis is A, show that U has
1-dimensional invariant subspaces.

Solution:
To say that there is not other is the same to say that we do not have real characteristic values , with a
corresponding characteristic vector ( otherwise they will be invariant subspaces). The characteristic polynomial
is (1 − x)(2 − x) + 2 = x2 + 3x + 4 with no real roots. As it has two complex root, each of these characteristic
values have a corresponding characteristic vector that forms an invariant subspace.

4
14. Consider the vector space of polynomial over the field F with degree at most n. List all subspaces that
are invariant under the differentiation operator Wi . Are you able to find a basis such that W0 = {α0 },
W1 = {α0 , α1 }, W2 = {α0 , α1 , α2 }, ..., Wn = {α0 , α1 , ..., αn }.
Solution:
The invariant subspaces are the polynomials of degree k, where k ̸= n are the invariant subspaces of under the
differentiation operator.
Let α0 = x0 , α1 = x1 , · · · , αn = xn . This satisfies the definition.
15. (6.4.2) Let W be an invariant subspace for T . prove that the minimal polynomial for the restriction operator
TW divides the minimal polynomial for T , without referring to matrices.
Solution:
The minimal polynomial for Tw divides any polynomial f (t), where f (TW ) = 0. Let f (t) be the minimal
polynomial for T . Then by definition, for all α ∈ W , as α ∈ V , f (TW )α = 0. Therefore f (TW ) = 0. Therefore,
the minimal polynomial of TW divides the minimal of T .
16. (6.4.3) Let c be a characteristic value of T and let W be the space of characteristic vectors associated with the
characteristic value c. What is the restriction operator TW ?
Solution:
The characteristic vectors are those such that T α = cα. Thus, for any basis, β of W , [TW ]β = cI. Thus,
cI − cI = 0 and x − c is the minimal polynomial.
17. (6.4.7) Let T be a linear operator on a finite-dimentsional vector space over the field of complex numbers.
Prove that T is diagonalizable if and only if T is annihilated by some polynomial over C which has distinct
roots.
Solution:
If T is diagonalizable, then the minimal polynomial is a product of distinct factors (Theorem from text). This
proves the forward direction. To prove in the reverse direction, if T is annihilated by a polynomial with distinct
roots, then the minimal polynomial, which must divide this polynomial must have distinct roots, therefore T
is diagonalizable.
18. (6.4.9) Let T be the indefinite integral operator
Zx
(T f )(x) = f (t)dt
0

on the space of continuous functions on the interval [0, 1].

(a) Is the space of polynomial functions invariant under T ?
(b) Is the space of differentiable functions invariant under T ?
(c) Is the space of functions which vanish at a point a ∈ R invariant under T ?
Solution:

(a) True, by definition.

(b) True, by definition.
(c) Let f (x) = x − a. This is false.
19. (6.4.10) Let A be a 3 × 3 matrix with real entries. Prove that, if A is not similar over R to a triangular matrix,
then A is similar over C to a diagonal matrix.
Solution:
If A is not triangularizable, then the characteristic polynomial must equal (x − c)(x2 + bx + c), where x2 + bx + c
has complex root. Thus, over the complex numbers, it has three characteristics. Therefore, it is diagonalizable.