Linear Algebra Qual Prep

Michael Scully
Mar 2022

1 Background
1.1 Introduction
Below, is a collection of typical Linear Algebra qualifying exam problems. They
will have be grouped into categories, as highlighted by Kent State’s algebra
study questions. As this study guide forms a basis for questions posed in qual-
ifying exams, when questions occurred in sampled quals it will be noted. The
sample list includes exams from:
1. Oregon State University - 2008 to 2021

2. University of Colorado (Denver) - 2010 to 2021

3. Penn State University - 2017 to 2020
4. UMass (Amherst) - 2009 to 2021
5. Duke - 2009 to 2014

6. UCF - 2018 to 2022

Furthermore, questions will be grouped into appetizer’s, entree and main course
groupings for level of (perceived) difficulty. This is in the opinion of the editor
and meant to help serve as a guideline for comfort with the material. As always,
individuals may find some questions easier or more difficult than suggested based
on comfort with material. It is not the goal to have this be a fully solutioned
manual, but serve as a baseline for discussion in reviewing material.

2 Generalized Matrix Theory

Generalized Matrix Theory will cover some concepts specifically for matrices.
Because of this, some questions may be calculation based in nature. They are
meant to be surface level interactions of deeper tested concepts.
2.1 Appetizers
1. (KS 1)Let m > n be positive integers. Show that there does not exist matrices
A ∈ Rm×n , B ∈ Rn×m such that AB = Im , where Im is the m × m identity
matrix.
2. (KS 5) Let T : R3 → R4 be given by T (v) = Av, where,
 
1 0 1
2 0 3
A= 0 1 0


3 4 2
.
a Find dim Null T
b Find dim Range T

3. (KS 6)Let  
0 1 0 ... 0
0 0 1 ... 0
 
J =  ... ... ... .. .. 

 . .
0 0 0 ... 1
0 0 0 ... 0
Show that J is similar to J t by a symmetric transforming matrix. [Recall:
Matrices X, Y are similar if there is a matrix P such that P −1 XP = Y , and P
is called the transforming matrix.]
4. (UMass F21)LetP3 (R) denote the real vector space of polynomials with real
coefficients of degree at most 3. Let T : P3 (R) → C the linear map of the
R-vetor spaces defined by p(x) = 3p(0) − ip”(1) + p′ (i). Compute a basis for
ker T .
5. (Denver W21)Suppose v1 , . . . , vm is linearly independent in V , w ∈ V and
v1 + w, . . . , vm + w is linearly dependent. Prove that w ∈ span{v1 , . . . , vm }.
6. (UMass
 W21)Consider
 the
two matrices:

1 3 0 2
2 6 and Y = 5 4 . Find all vectors x, y ∈ R2 and v ∈ R4
 
X = 3 9 1 6 
−2 −6 0 −4
such that Xx = v = Y y.
7. (UMass F20)Consider the vectors v1 = (1, 2, 2), v2 = (−1, 0, 2).
a Give an orthonormal basis for the subspace V ⊂ R3 that they span.

b Find a unit vector orthogonal to V .

8. (Denver F20)Complete the following problems:
 a Find bases for the null and column spaces of the matrix A.
 
2 −1 3 −1 1 −5
A = −1 4 2 11 −3 10 
1 1 3 4 2 −7
.
b Let S, T be subspaces of Rn .

(a) Show there exists matrices A, B such that S = Null(A), T = Null(B).

Given bases for S, T describe how such matrices A, B could be found.
(b) Find a matrix C such that S ∩ T = Null(C).
(c) Describe how to find a basis for S ∩ T .
9. (UMass W20)Let a1 = (1, 1, −1), a2 = (0, 2, 3), a3 = (z, −3, −7).

a Find all values z for which there will be a unique solution to the equation
a1 x1 + a2 x2 + a3 x3 = b ∈ R3 for every b. Explain your answer.
b Consider a4 = (1, 4, −5). Fine all values of z for which there will be a
unique solution to a1 y1 + a2 y2 + a3 y3 + a4 y4 = c ∈ R3 for every c. Explain
your answer.
c Use Gauss-Jordan elimination to find the general solution to the system
of equations:

x1 − 2x2 + x3 = 1
x1 − 2x2 − x3 = 1
2x1 − 5x2 + 2x3 = 1

d Use (c) to find the linear equation for the plane going through points
(1, −2, 1), (1, −2, −1), (2, −5, 2).
10. (UMass W20)Complete the following:
a Find an orthogonal basis for the subspace S spanned by the vectors v1 =
(−2, −1, 1, 2), v2 = (1, 1, 1, 1), v3 = (4, 1, 1, 4) that contains v1 .
b Project the vector (4, 1, 1, 3) onto S. Find the linear combination of the
projected vector in terms of v1 , v2 , v3 .
c The answer to (b) say coefficients a1 , a2 , a3 yields the least squares so-
lution for the parabola y = a3 x2 + a2 x + a1 going through the points
(−2, 4), (−1, 1), (1, 1), (2, 3). Explain why.
   
1 2 −1 −1
11. (UMass F19)Consider the matrix A = 2 2 c  and b =  2 .
3 5 −2 −1
 a Characterize the values of c such that the matrix has a nontrivial nullspace.
b For c = 1, find all solutions x to the system Ax = b.
c For c = 0, find all solutions x to the system Ax = b.
12. (Denver S19)Prove or the reject the following:

a There exists a matrix A ∈ M4 (R) for which the column space and null
space are identical.
b 0 ̸= A be an m × n matrix with m ≤ n let b ∈ Rm such that Ax = b has no
solution and let 0 ̸= d ∈ Rm for which there exists a solution to Ax = d.
What is the minimal / maximal dimension of the of solutions for AX = d?
Prove the best bounds available on given information, prove that your
bounds are correct, and prove that they can be tight for all well-defined
m, n.
c Suppose that S is a fixed, invertible n × n matrix. Let W be the set of all
matrices A for which S −1 AS is diagonal. Prove or reject: W is a vector
space.
13. (Denver S19)Complete the following:
a Let T : P3 → P3 be an operator that maps p(t) = a0 + a1 t + a2 t2 + a3 t3
onto q(t) = a3 t + a2 t2 + a3 t3 . Prove or reject: T is a linear transformation.
If so, provide a matrix representation.
b Let a, b ̸= 0 ∈ R be fixed. Find a basis for the subspace in R4 created
from intersecting:

S = span{(a, 0, 0, a), (a, 0, a, 0), (b, b, 0, 00)}

, and
T = span{(b, 0, 0, a), (0, a, b, 0)}
.

2.2 Entrees
14. (KS 2)Let A, B be nonsingular n × n matrices over C.

a Show that if A−1 B −1 AB = cI, c ∈ C, then cn = 1

b Show that if AB − BA = cI, c ∈ C, then c = 0
15. (KS 3)Let A be strictly upper triangular n × n matrix with real entries, and
let I be the n × n identity matrix. Show that I − A is invertible and express
the inverse as a function of A.
16. (KS 4)Solve the following:
 a Give an example of a complex 2 × 2 matrix that does not have a square
root.
b Show that every complex non-singular n × n matrix has a square root.
[Hint: Show that a Jordan block with non-zero eigenvalue has a square
root]
17. (UMass F21)Let A be the 3 × 2 matrix given by
 
1 2
A = −2 1
−1 0
Find a vector of length 5 in the orthogonal complement of the column space of
A. Find an orthonormal basis of the orthogonal complement of the null space
of A.
18. (UMass F21)For k ≥ 0, let the vector xk is defined by, x0 = (5, 1, 7) and
xk+1 = Axk . Where A is the matrix
 1
− 2 0 12


A =  0 1 0
−1 0 1
a Compute three linearly independent eigenvectors v1 , v2 , v3 of A.
b Give an explicit formula for xk in terms of k, v1 , v2 , v3 .
c Compute limk→∞ xk .
19. (UMass W21)Let Mn (C) be the set of n × n complex matrices and A be a
matrix in Mn (C).
a Prove that there exists B ∈ Mn (C) such that AB = 0 and rank(A) +
rank(B) = n − 1.
b Prove or provide counterexample: There exists B ∈ Mn (C) such that
AB = 0 = BA and rank(A) + rank(B) = n
20. (UMass S18)
a Let A, B ∈ Mn (R). If AB = 0, then rank(A) + rank(B) ≤ n.
b Let A ∈ Mn (R), then ∃B ∈ Mn (R) such that AB = 0 and rank(A) +
rank(B) = n
21. (PSU W21)Let n ≥ 1, define M to be a matrix given in the block form by:
 
A B
M=
C D
, where each is an n × n matrix. Show that if det A ̸= 0 and A, C commute,
then det M = det(AD − CB).
22. (OSU F20)Let Mn (R) be the set of all n × n real matrices.
 a Prove that the identity (a + b)−1 = a−1 + b−1 can never hold for a, b ∈ R,
but can hold for a, b ∈ C.
b Prove that there exists A, B ∈ Mn (R) such that (A + B)−1 = A−1 + B −1
if and only if there exists J ∈ Mn (R) such that J 2 = In .

c As a consequence of (b), show that such matrices A, B exist if and only if

n is even.
23. (UMass F20)View C as a two-dimensional vector space over $ with the basis
{1, i}.
a Given the complex number α = a1 + bi, express multiplication by α as a
real valued matrix Mα .
b Does Mα preserve angles between vectors ?
24. (Denver W20)Let A, B be two real 10 × 10 matrices. Suppose the rank of
A is 6, and the rank of B is 4. Justify your answers for the following:

a What is the minimum possible rank of the matrix A2 ?

b What is the maximum possible rank of the matrix AB t ?
c If the columns of A are orthogonal to the columns of B, must the rank of
A + B be equal to 10?

25. (Denver W19, UCF ???)Let v1 , v2 , . . . , vn be a list of independent vectors

in the vector space V . Show that the list of vectors v1 + v2 , v2 + v3 , . . . , vn−1 +
vn , vn + v1 is linearly independent if and only if n is odd.
26. (UMass F18)Let A, B be n × n matrices.
a If AB = 0, show that rank(A) + rank(B) ≤ n.

b For any A, there exists B such that AB = 0 and rank(A) + rank(B) = n.

2.3 Main Course

27. (OSU F21, Denver W19)Use linear algebra to find f (x) ∈ P2 (R) = {f (x) ∈
R1
R[x]| deg(f ) ≤ 2} such that f ′ (0) = 0 and f (x) minimizes 0 (2x − f (x))2 dx.
28. (Denver W19, OSU F21Find real numbers x, y, z such that
Z 1
(ln t − x − yt − zt2 )2 dt
0
is minimal.
R1 R1 k
1 1
Hints: 0 tk ln tdt = − (k+1) 2 ; 0 t dt = k+1 for k ≥ 0.

29.
R ∞ (OSU S21)Consider the continuous function f : R → (0, ∞) such that
j
−∞
|x| f (x)dx < ∞ for all j ≥ 0. Denote Pm the vector space of all real-
valued polynomials of degree at most m, equipped with the inner product.
 Z ∞
⟨g, h⟩ = g(x)h(x)f (x)dx
−∞
R∞
Let mj = −∞
xj f (x)dx, and consider polynomials p0 (x) = m0 and
 
m0 m1 m2 ... mn

 m1 m2 m3 ... mn+1 
 m2 m3 m4 ... mn+2 
pn (x) = det 
 
.. .. .. .. .. 

 . . . . . 
mn−1 mn mn+1 ... m2n−1 
1 x x2 ... x2n

. Prove that po , . . . pm are orthogonal in Pm .

30. (Denver W20) Let P n denote the real vector space of polynomials of degree
strictly less than n. For two functions f, g ∈ P n define the inner product by
Z 1
⟨f, g⟩ = f (t)g(t)dt
0
.
a Verify the above is an inner product.

b Apply the Gram-Schmidty procedure to the basis {1, t, t2 } to find an or-

thogonal basis for P 3 .
31. (OSU F19)Let A be an m × n real matrix, B an n × p real matrix. Prove
the following:

a rank(AB) = rank(B) − dim(N (A) ∩ R(B)), where N (M ) is the null space

of the matrix M , R(M ) is its column space.
b rank(At A) = rank(A) = rank(AAt ).
32. (OSU F19)Let M be the real n × n matrix such that each column sum of
M is zero.
a Show the determinant of M equals 0.
b Show that for each column, the n cofactors of elements in that column are
all equal.

c Show that the vector whose ith entry is the cofactor of any element of the
ith column of M gives an element of the nullspace of M .
33. (Oregon F18)Consider the n × n matrix Fn = (fi,j ) of binomial coefficients
fi,j = i−1+j−1 . Prove that det(Fn ) = 1 for all n. Recall nk = n−1 n−1





i−1 k−1 + k .
2.4 Uncategorized
34. (Denver W21)Let V be a vector space of dimension n over a field F. For
any nilpotent operator T on V definite the smallest integer p such that T p = 0
as the index of nilpotency of T .

a Suppse that N is nilpotent of index p. If v ∈ V such that N p−1 (v) ̸= 0,

prove that {v, N (v), . . . , N p−1 (v)} is linearly independent.
b Show that N is nilpotent of index n if and only if there is an ordered basis
v1 , v2 , . . . , vn ∈ V such that the matrix N with respect to the basis is of
the form  
0 0 0 ... 0 0
1 0 0 . . . 0 0
 
0 1 0 . . . 0 0
A = 0 0 1 . . . 0 0
 
 
 .. .. .. . . . .
. . . . .. .. 
0 0 0 ... 1 0

c Show that an n × n matrix M over F such that M n = 0 and M n−1 ̸= 0 if

and only if M is similar to a matrix of the above form.

3 Jordan Canonical Forms

This section covers Jordan Canonical Forms, along with diagonalization checks.
Also, Characteristic and Minimal Polynomial properties will be used to check
the Jordan Form.

3.1 Appetizers
35. (KS 9)Let A be an n×n Jordan block. Show that any matrix that commutes
with A is a polynomial in A.
36. (KS 21)The characteristic polynomial of a certain 4 × 4 matrix A has the
two distinct roots 2, 3, with the multiplicity of 3 less than the multiplicity of 2.
List all possible Jordan canonical forms of A, up to similarity.
37. (KS 22)Let A be an n × n matrix, I the n × n identity matrix. Show that
if A2 = I and A ̸= I, then λ = −1 is an eigenvalue of A.
38. (KS 24)Solve the following:

a Find all possible Jordan canonical forms of the 5 × 5 complex matrix with
minimal polynomial (x − 2)2 (x − 1).
b Find all possible Jordan canonical forms of the complex matrix with char-
acteristic polynomial (x − 3)3 (x − 5)2 .
39. (KS 25)Find all possible Jordan canonical forms for the following with
explanation.
 a A linear operator T with characteristic polynomial p(x) = (x − 2)4 (x − 3)2
and minimal polynomial m(x) = (x − 2)2 (x − 3)2 .
b A linear operator T with characteristic polynomial p(x) = (x − 4)5 , and
dim ker(T − 4I) = 3.

40. (KS 26)A matrix A has characteristic polynomial p(x) = (x − 3)5 and
minimal polynomial m(x) = (x − 3)3 .
a List all possible Jordan canonical forms for A.
b Determine the Jordan canonical form of the matrix
 
3 −1 2 0 0
2 3
 0 −2 0 
A = 1 0
 3 −1 0 
0 −1 2 3 0
0 2 −3 0 3

which has the given characteristic and minimal polynomial.

41. (KS 29) Let  
2 0 0 0
0 1 0 0
A
2

0 1 1
0 0 0 1
,  
1 1 0 0
−1 3 0 0
B
−1

1 1 1
−1 1 −1 3
,  
2 −1 1 −1
0 1 1 −1
C
 
0 −1 3 −1
0 0 0 2
.
a Find the characteristic polynomial of A, B, C
b Find the minimal polynomial of A, B, C

c Find the eigenvalues of A, B, C

d Find the dimensions of all eigenspaces of A, B, C
e Find the Jordan canonical form of A, B, C
42. (KS 30)Let A be the following matrix:
 
2 0 0 0 0 2
0 2 1 1 2 0
 
0 0 2 2 0 0
A= 0

 0 0 2 0 0
0 0 0 1 2 0
0 0 0 0 1 2

Find the characteristic polynomial, minimal polynomial, eigenvalues, eigenvec-

tors, dimensions of eigenspaces and the Jordan canonical form of this matrix.
43. (KS 31) Let  
2 0 a b
0 1 0 0 
A= 0 c 3 −2


0 d 2 −1
a Determine conditions on a, b, c, d such that there is only one Jordan block
for each eigenvalue of A in the Jordan canonical form of A.
b Suppose a = c = d = 2 and b = −2. Find the Jordan canonical form of A.
44. (KS 32) Let  
2 0 a b
0 1 0 0
A= 
0 c 3 −2
0 d a −1
a Determine conditions on a, b, c, d such that there is only one Jordan block
for each eigenvalue of A in the Jordan canonical form of A.
b Suppose a = b = c = d = 2. Find the Jordan canonical form of A.
45. (KS 36) Let  
−1 4 −2
A = −2 5 −2
−1 2 0
, with characteristic polynomial p(x) = (x − 1)2 (x − 2).
a For each eigenvalue λ of A, find a basis for the eigenspace Eλ
b Determine if A is diagonalizable. If so, give matrices P, B such that
P −1 AP = B and B is diagonal. If not, explain (carefully) why A is
not diagonalizable.
46. (KS 37) Let  
2 −1 −1
A = 1 0 −1
1 −1 0
.
 a Verify that the characteristic polynomial is p(x) = x(x − 1)2 .
b For each eigenvalue λ of A, find a basis for the eigenspace Eλ
c Determine if A is diagonalizable. If so, give matrices P, B such that
P −1 AP = B and B is diagonal. If not, explain (carefully) why A is
not diagonalizable.

47. (KS 38) Let  

5 0 6
A= 2 2 4
−2 0 −2
.

a Verify that the characteristic polynomial is p(x) = (x − 1)(x − 2)2 .

b For each eigenvalue λ of A, find a basis for the eigenspace Eλ

c Determine if A is diagonalizable. If so, give matrices P, B such that
P −1 AP = B and B is diagonal. If not, explain (carefully) why A is
not diagonalizable.

48. (KS 34)Let T : V → V be a linear transformation satisfying T 2 = 0. Prove

that the Jordan canonical form of T consists of dim(ker T ) Jordan blocks, and
dim(Im T ) of which 2 × 2 blocks.
49. (KS 35)Let A be an n × n nilpotent matrix such that An−1 ̸= 0. Show that
A has exactly one Jordan block.
50. (OSU F21)Find the Jordan Canonical Form and the associate basis for the
following complex matrix:
 
1 0 0 1
 0 3 −1 0
A= −1 0 2 1


0 1 −1 2

51. (Denver F21)Suppose T ∈ L(V ) and (T − 2I)(T − 3I)(T − 4I) = 0. Suppose

λ is an eigenvalue of T . Prove that λ ∈ {2, 3, 4}.
52. (OSU W21)Consider the following space of functions:

V = {(a2 x2 + a1 x + a0 ) exp 2x|a0 , a1 , a2 ∈ R}

. Then A : V → V defined by Af = f ” is a linear operator. Find a Jordan

basis of V for A as well as the
 corresponding
 Jordan canonical form.
2 0
53. (Denver W21)Let A = . Define: T : M2 (R) → M2 (R) by T (B) =
2 1
AB − BA.

a Fix an ordered basis B of M2 (R) and compute [T ]B that represents T

with respect to this basis.
 b Give eigenvalues of T .
c Compute a basis for each of the eigenspaces of T .
d Give the minimal and characteristic polynomials of T and the Jordan form
of T . Say whether T is diagonalizable or not.
54. (UMass W21)Consider the matrix:
 
2 −4 2 2
−2 0 1 3
M = −2

−2 3 3
−2 −6 3 7

. Find a change-of-basis matrix A such that A−1 M A is in Jordan form.

55. (OSU F20)Let A be an n × n matrix over the real numbers with eigenval-
Pn λ1 , λ2 ,k. . . λn counted with multiplicity. Show that any polynomial q(x) =
ues
k=0 ak x the eigenvalues of q(A) counted with multiplicity are q(λ1 ), q(λ2 ), . . . , q(λn ).
56. (UMass F20)Consider the following matrix:
 
2 0 1 −3
0 2 10 4 
A= 0 0 2

0
0 0 0 3
.
a Find the roots of the characteristic polynomial.
b Find a maximal set of independent eigenvectors of A. Is the matrix diag-
onalizable?
57. (UMass F19)Consider the linear transformation T : R2 → R2 the reflects
about the line y = x.
a What is T (e1 ), T (e2 )?
b Find a 2 × 2 matrix A representation of T with respect to the basis
{(1, 0), (0, 1)}.
c Find the eigenvectors and the corresponding eigenvalues of T .
d Diagonalize A.
e Give an example of a 2 × 2 nonzero matrix B that has no eigenvectors and
justify your answer.
58. (Oregon F18)Let A, B be n × n complex matrices. Prove or disprove the
following statements:
a If A, B are diagonalizable, so is AB.
b If A2 = A, then A is diagonalizable.
3.2 Entrees
59. (KS 8)Show that if A is an n × n matrix, then An can be written as a linear
Pn−1
combination of the matrices {I, A, A2 , . . . , An−1 }. [Namely, An = k=0 αk Ak ,
where A0 = I.]
60. (OSU S21)Suppose A ∈ Mn,n (C) is such that A has an eigenspace of dimen-
sion bigger than one. Prove that for any v ∈ Cn the vectors v, Av, . . . , An−1 v
are linearly dependent.
61. (KS 10) Let A be a square matrix whose minimal polynomial is equal to its
characteristic polynomial. Show that if B is any matrix that commutes with A,
then B is a polynomial in A.
62. (KS 12)Prove that any n × n complex matrix A is diagonalizable if and only
if the minimal polynomial of A has distinct roots.
63. (KS 14)Let V be a vector space and let T : V → V be a linear transforma-
tion.
a Show T is invertible if and only if the minimal polynomial of T has a
non-zero constant term.
b Show that if T is invertible, then T −1 can be expressed as a polynomial
in T .
64. (KS 15)Let A, B be complex 3 × 3 matrices having the same eigenvectors.
Suppose that the minimal polynomial of A is (x − 1)2 , and the characteristic
polynomial of B is x3 . Show the minimal polynomial of B is x2 .
65. (KS 16)Let A, B be 5 × 5 complex matrices having the same eigenvectors.
Show that if the minimal polynomial of A is (x + 1)2 and the characteristic
polynomial of B is x5 then B 3 = 0.
66. (KS 20)Let A be a 5 × 5 complex matrix such that A3 = 0. Find all possible
Jordan canonical forms for A.
67. (KS 27)Let T : V → V be a linear transformation defined on the finite
dimensional vector space V . Let λ be an eigenvalue of T , and the set Wi =
{v ∈ V |(T − λI)i (v) = 0}. If m is the multiplicity of the λ as a root of the
characteristic polynomial of T , prove Wm = Wm+1 .
68. (KS 33)Let A be a square complex matrix with a single eigenvalue λ. Show
that the number of blocks in the Jordan form of A is the dimension of the
λ-eigenspace.
69. (KS 40) Let A be a matrix of the form
 
0 0 0 ... −c0
1 0 0 . . .
 −c1  
0 1 0 . . . −c2 
. . . .. 
 
A=  .. .. .. . . .
 . 
0 0 0 . . . −cn−2 
 

0 0 0 ... cn−1
Find the characteristic polynomial.
70. (OSU F21) Complete the following:
a Let T be a linear operator on a vector space over field F, and suppose
Wi , . . . Wk are T -invariant subspaces of V such that V = W1 + · · · + Wk .
Let mi (x) be the minimal polynomial of T |Wi for each 1 ≤ i ≤ k. Prove
that the minimal polynomial of T is m = lcm(m1 , . . . , mk ).
b For α ∈ C set T : C3 → C3 be defined by T (x, y, z) = (αx + y, −x + y, 2z).
Let W1 = {(x, y, 0)|x, y ∈ C} and W2 = {(0, 0, z)|z ∈ C}. Use (a) to
determine which α ∈ C make T diagonalizable.
71. (PSU
  n ≥ 1, and let A be the n × n matrix given by the block form
F20)Let
0 1
A= . Determine the Jordan canonical form of A over C.
In−1 0
72. (Denver F20)Let A be an n × n matrix over C. Show that:
a there exists an invertible n × n matrix Q such that
 
λ1 b12 . . . b1n
 0 b22 . . . b2n 
Q−1 AQ =  .
 
.. .. .. 
 .. . . . 
0 bn2 . . . bnn

b Use induction to show that any n × n matrix over C is similar to an

uppertriangular matrix
 
λ1 ∗ . . . ∗
 0 λ2 . . . ∗ 
 
 .. .. .. .. 
. . . . 
0 0 . . . λn

73. (UMass W20)Complete the following:

 
1 −1 0
a is −1 1 0 diagonalizable? If so, find its diagonalization. If not,
3 4 1
explain why.
 
−2 3 1 5
 0 1 0 2
b is 
 0 0 3 4 diagonalizable? If so, find its diagonalization. If not,


0 0 0 1
explain why.
c One of the last two matrices was diagonalizable, call it A. Find A7 .
74. (Denver W19)Let n be an integer. Find all n × n matrices A with complex
entries, such that A is self-adjoint and A3 = 2A + 4I.
3.3 Main Course
75. (KS 7)State and prove the Cayley-Hamilton Theorem
76. (KS 11)Let A be an n×n matrix, v a column vector. Suppose {v, Av, . . . , An−1 v}
is a linearly independent set of vectors. Prove that if B is any matrix that com-
mutes with A, then B is a polynomial in A.
77. (KS 13)Let G = GLn (C) be the multiplicative group of invertible n ×
n matrices with complex entries. Let g ∈ G with finite order. Show g is
diagonalizable.
78. (KS 19)Let A, B be complex n × n matrices. Prove that if AB = BA, then
A, B share a common eigenvector.
79. (KS 23, OSU W19, Denver W19)Solve the following:
a Show that two 3 × 3 complex matrices are similar if and only if they have
the same characteristic and minimal polynomials.
b Is the conclusion of part (a) true for larger matrices? Prove or provide a
counter-example.
80. (OSU W19, Denver W19)Let A, B be nonsimilar n × n complex matrices
with both the same minimal polynomial and characteristic polynomial. Show
that n ≥ 4 and that the common minimal polynomial is not equal to the common
characteristic polynomial.
81. (Denver W19, OSU W19/Denver S19)Complete the following:
a Prove or disprove: Two n × n real matrices with the same characteristic
and minimal polynomials must be similar.
b Let A be an idempotent real matrix A2 = A, then A is diagonalizable.
82. (KS 28)Let  
1 2 3
A = 0 −1 4
0 0 3
be a matrix over the field F, where F is either the field of rational numbers or
the field of p elements for some prime p.
a Find a basis of the eigenvectors for A over those fields for which such a
basis exists.
b What is the Jordan canonical form of A over the fields not included in
part (a).
83. (KS 39) Let A be a matrix of the form
 
0 1 0 ... 0
0 0 1 ... 0
 
A =  ... .. .. .. .. 

 . . . .
0 0 0 ... 1
c1 c2 c3 ... cn
Show that the minimal polynomial and characteristic polynomial of A are equal.
84. (UMass F21)Let A be an invertible diagonalizable 4 × 4 matrix whose char-
acteristic polynomial is pA (λ) = (λ−5)(λ2 (λ−1)−λ+1), and P is an invertible
4 × 4 matrix. Compute pB (λ) if B = (P (AT )P −1 )−1
85. (Denver F20)Consider the map T : Mn (R) → Mn (R) by T (A) = At .
a Find the characteristic polynomial and minimal polynomial of T .
b Find the Jordan form of T .
86. (PSU W20) Let f (x) = (x2 − 1)(x2 + 1). Consider the set of real 3 × 3
matrices such that det A = 1 and f (A) = 0. Determine the number of (real)
similarity classes of such matrices and give representatives of each class.
87. (OSU F19)Let M be a complex n × n matrix whose eigenvalues λ1 , . . . , λn
satsify the following relations:
X
λi = (−1)1
i
X
λi1 λi2 = (−1)2
i1 <i2
X
λi1 λi2 . . . λin−1 = (−1)n−1
i1 <i2 <···<in−1

λ1 . . . λn = (−1)n

a Determine the eigenvalues of M .

b Is M diagonalizable ?
88. (PSU S19)Prove that there exists two non-similar invertible 3×3 real matri-
ces A which satisfy A−1 = A2 + A and give a representative for each similarity
class.

3.4 Uncategorized

4 Linear Transformations
This section covers Linear Transformations, dealing with properties of the ker-
nel, image and quotient spaces. The Rank-Nullity theorem will come into play
for many arguments in this section.

4.1 Appetizers
89. (KS 49)Let T : V → W be a surjective linear transformation of finite di-
mensional vector space over a field F. Show that there is a linear transformation
W → V such that T ◦ S is the identity map on W .
90. (KS 50)A linear transformation T : V → W is said to be independence
preserving if T (I) ⊆ W is linearly independent whenever I ⊆ V is a linearly
independent set. Show that T is independence preserving if and only if T is
injective.
91. (KS 54)Let A be a complex n × n matrix. Let L : Cn×n → Cn×n be the
linear transformation given by L(M ) = AX + XA for X ∈ Cn×n . Prove that if
A is nilpotent, then L is nilpotent.
92. (UMass F21)For any real number k, let Tk : R3 → R3 be the linear map
defined by the conditions T (e1 ) = e2 , T (e2 ) = (k 2 − 1, 0, k), T (e3 ) = (k 2 −
2k, 0, 0). Determine the values of k with the following properties:
a Tk is not invertible. For those values, say what is the rank and nullity of
Tk .
b Tk is an isometry.
c Tk has the same rank as Tk2 .
93. (OSU S21)Let A be an n × n real matrix. Suppose tr(AX) = 0 for any
n × n real matrix X with tr(X) = 0. Prove that A = λIn×n for some λ ∈ R.
94. (OSU W21)Suppose U, V, W be finite dimensional inner product spaces over
C. Let L : U → V, T : V → W be linear transformations and L∗ , T ∗ be the
corresponding adjoint transformations.
a Prove Range(L) = ker(L∗ )⊥
L T
b Assume the sequence U − → V − → W is exact. [Namely Range(L) =
ker(T ).] Define S = LL∗ + T ∗ T . Prove S is invertible.
95. (PSU W21)Let U, V, W be finite-dimensional vectors spaces over F. Let
A : U → V, B : V → W be linear maps. Show that rank(BA) ≥ rank(A) +
rank(B) − dim(V ).
96. (OSU F20)Let W be a subspace of an inner product space V , P : V → V
be a linear transformation. Prove that P is an orthogonal projection onto W
if and only if P is both idempotent and self-adjoint. [Recall: an orthogonal
projection onto W if for any x ∈ W, P (x) = x and for any y ⊥ W, P (y) = 0.]
97. (UMass F19)Complete the following:
a Find an orthogonal basis for the span of the following three vectors S2 =
{(1, 0, 1), (0, 1, 1), (1, 3, 3)} and write (1, 1, 2) in terms of such a basis.
b Let W = span((1, 1, 1)). Find a basis of the orthogonal complement W ⊥ ∈
C3 .
98. (UMass F18, Check a for multi)Complete the following:
 
2 2 1 −2
a Let T1 : R → R be the linear transformation that T1 (v) = v
−2 4
2 2 2
for every v ∈ R . Prove that for any linear transformation T2 : R → R ,
the linear transformation T2 ◦ T1 is never surjective.
b Let T : Rn → Rn be a linear transformation. Show that Rn ⊇ Im(T ) ⊇
Im(T 2 ) ⊇ . . . ,and show that there exists some m ∈ N+ such that Im(T k ) =
Im(T k+1 ) for all k ≥ m.
4.2 Entrees
99. (KS 41)Let φ : V → V be a linear transformation of a finite dimensional
vector space to itself. Prove there exits a decomposition of V as V = U ⊕ W ,
where each summand is φ-invariant, φ|U is nilpotent and φ|W is nonsingular.
100. (KS 42)Let V be a vector space and T : V → V a linear transformation
such that T 2 = T . Show that V = ker T ⊕ Im T .
101. (KS 45)Let V, W be finite dimensional vector spaces and let T : V → W
be a linear transformation. Show that dim(ker T ) + dim(Im T ) = dim(V ).
102. (KS 46)Let V be a finite dimensional vector space and T : V → V a non-
zero linear operator. Show that if ker T = Im T , then dim V is an even integer
and the minimal polynomial of T is m(x) = x2 .
103. (KS 47)Let V be a finite dimensional vector space over a field F and let
T : V → V be a nilpotent linear transformation. Show that the trace of T is 0.
104. (KS 48)Let V be the vector space of 2 × 2 matrices over a field F . Let
 
a b
A= ∈V
c d
. Let T : V → V be the linear transformation defined by T (X) = AX. Compute
det(T ).
105. (KS 51)Let V, W be vector spaces and let T : V → W be a surjective linear
transformation. Assume for all subsets S ⊆ V that if T (S) spans W , then S
spans V . Prove that T is injective.
106. (KS 52)Let T : V → W be a linear transformation of vector spaces over a
field F .
a Show that T is injective if and only if {T (v1 ), T (v2 ), · · · , T (vn )} is linearly
independent in W whenever {v1 , v2 , · · · , vn } is linearly independent in V .
b Show that T is surjective if and only if {T (x)|x ∈ X} is a spanning set for
W whenever X is a spanning set for V .
107. (KS 53)Let A be a complex n × n matrix. Let L : Cn×n → Cn×n be the
linear transformation given by L(M ) = AM for M ∈ Cn×n . Express det L in
terms of det A with proof.
108. (OSU S21)Suppose L : V → V , where dim V = n. If L is nilpotent, prove
Ln = 0.
109. (Denver F20)Let V be a finite-dimensional real inner product space. Let
T ∈ L(V ) and let U be a T -invariant subspace of V . Prove or give a counterex-
ample for the following:
a U ⊥ is T ∗ -invariant.
b U ⊥ is T -invariant.
110. (UMass W20)Complete the following proofs:
a Let T1 : Rm → Rn such that T1 (v) = Av, T2 : Rn → Rm . Prove that if T1
is not injective then neither is T2 ◦ T1 . Prove that if T2 is not surjective,
then neither is T2 ◦ T1 .
 b Let T1 : Rm → R, T2 : Rn → Rm , such that T1 (v) = Av, T2 (w) = At w
for every v ∈ Rm , w ∈ Rn . Prove that T1 is surjective if and only if T2 is
injective.
c Let A be an n × n matrix. Show that if rank(AB) = rank(B) for all n × n
matrices B, then A is invertible.
111. (Denver W20)We consider the inner product space Rn with its standard
inner product. Let T (z1 , z2 , . . . , zn ) = (z2 − z1 , z3 − z2 , . . . , z1 − zn ).
a Give an explicit expression for the adjoint T ∗ .
b is T invertible? Explain.
c Find the eigenvalues of T .
112. (Denver W20)Complete the following:
a Let n ≥ 2, and V be an n-dimensional vector space over C with basis
vectors e1 , . . . , en . Let T be the linear map of V satisfying T (ei ) = ei+1
for 1 ≤ i ≤ n − 1 and T (en ) = e1 . Is T diagonalizable?
b Let V be a finite-dimensional vector space and T : V → V a diagonalizable
linear transformation. Let W ⊆ V be a subspace which is mapped into
itself by T . Show that T |W is diagonalizable.
113. (Denver S19)Let V be a finite-dimensional vector space.
a Suppose T ∈ L(V ) is such that every vector in V is an eigenvector of T .
Prove or disprove that T is a scalar multiple of the identity operator.
b Suppose T ∈ L(V ) is such that every subspace of V with dimension
dim V − 1 is invariant under T . Prove that T is a scalar multiple of
the identify operator.
114. (Oregon F18)Suppose P is a linear operator on a vector space V such that
P2 = P.
a Prove that V = RP ⊕ NP .
b Further suppose that V is a real inner product space and that P has finite
dimensional range. Let T be the orthogonal projection from V to the
range of P . Determine T P and P T .
c With notation and hypotheses as in (b), find the adjoint operator of T .
115. (Oregon F18, UCF F21)Complete the following problems:
a Suppose that λ1 , . . . , λr are distinct non-zero complex numbers. Show
that the following matrix is invertible.
 
λ1 λ2 . . . λr
2 2
λ1 λ2 . . . λr  2
B= .
 
.. .. .. 
 .. . . .
λr1 λr2 ... λrr
 b Suppose that A is a complex square matrix such that the trace of Ak is
zero for each 1 ≤ k ∈ N. Show that all the eigenvalues of A are zero.

4.3 Main Course

116. (KS 43)Let V be a vector space over a field F. A linear transformation
T : V → V is said to be idempotent if T 2 = T . Prove that if T is idempotent,
then V = V0 ⊕ V1 , where T (v0 ) = 0 for all v0 ∈ V0 and T (v1 ) = v1 for all
v 1 ∈ V1 .
117. (KS 44)Let U, V, W be finite dimensional vector spaces with U ≤ V . Show
that if T : V → W is a linear transformation having the same rank as T |U :
U → W , then U is complemented in V by a subspace K satisfying T (x) = 0 for
all x ∈ K.
118. (KS 55)Let T : V → V be a linear transformation. Let X = ker T n−2 , Y =
ker T n−1 and Z = ker T n . Observe that X ⊆ Y ⊆ Z (no need to prove). Sup-
pose, {u1 , . . . , ur }, {u1 , . . . , ur , v1 , . . . , vs }, {u1 , . . . , ur , v1 , . . . , vs , w1 , . . . , wt } are
bases for X, Y, Z respectively. Show that S = {u1 , . . . , ur , T (w1 ), . . . , T (wt )} is
contained in Y and linearly independent.
119. (Denver F20)For x, y ∈ Cn let ⟨x, y⟩ be the Hermitian inner product
n
P
j xj y j . Let T be an operator on C such that ⟨T x, T y⟩ = 0 if ⟨x, y⟩ = 0.
Prove that T = kS for some scalar k and some unitary operator S.
120. (Denver W20)Suppose V is a finite-dimensional vector space over F.

a Prove or disprove: if S, T are nilpotent operators on V , then S + T is

nilpotent.
b Prove or disprove: if S, T are nilpotent operators on V and ST = T S,
then S + T is nilpotent.
c Prove that S is a nilpotent operator on V , then I ± S are invertible.

d Let N be an operator on an n−dimensional vector space, n ≥ 2, such that

N n = 0, N n−1 ̸= 0. Prove this is no operator T with T 2 = N .
121. (OSU F19)Let F be a field, V the vector space of n × n matrices over F.

a Fix B ∈ V . Define fB (A) = trace(B t A). Show that fB is a linear function

on V .
b Show that every linear functional on V is of the above form.
c Show that B 7→ fB is an isomorphism of V onto V ∗ , the dual space of V .

122. (UMass F19)Let V be a vector space and T : V → V be linear transfor-

mation.
1. Show that {x : T (x) = x} is subspace of V .
2. Show that if T 2 = T , then for any x ∈ V , we have that x − T (x) ∈ ker T .
 3. Show that if T 2 = T then V = {x : T (x) = X} ⊕ ker(T ).
4. Characterize all linear T such that T 2 = T .
123. (Denver S19)Let ∥∥˙ denote an arbitrary vector norm on Rp . The matrix
norm induced by ∥∥˙ is defined by:

∥P x∥
∥P ∥ = max
x̸=0 ∥x∥

, for each p × p real matrix P .

˙ a norm on the vector space of real p × p matrices.
a Prove that ∥∥is
b Let P be a p × p real matrix. Suppose ∥P ∥ < 1. Prove that I + P is
nonsingular and that
1 1
≤ ∥(I + P )−1 ∥ ≤
1 + ∥P ∥ 1 − ∥P ∥
.

124. (PSU S19)Let V be a finite-dimensional vector space, and let T : V → V

be a linear transformation. Suppose there is a vector v ∈ V such that the list
{v, T v, T 2 v, . . . } spans V . Show that if S : V → V is a linear transformation
that commutes with T , then there is a polynomial f such that S = f (T ).
125. (OSU W19)Let A(t) be a real-valued continuous n × n matrix function,
defined for all t ∈ R, and Φ(t) be a real-valued, continuously differentiable n × n
matrix function such that
Φ′ (t) = A(t)Φ(t)
for all t ∈ R. Here Φ′ (t) is the matrix function obtained from Φ(t) by taking
the derivatives of each of its entries. Show that
d
(det(Φ(t))) = trace(A(t)) det(Φ(t))
dt
, for all t ∈ R.
126. (OSU W19)Let T : V → V be a linear operator on a complex finite
dimensional vector space. Let V ∗ denote the dual space of V , and T t the
transpose linear operator for T . Using component-wise operations, form the
external direct sum vector space V ⊕ V ∗ ; let S : V ⊕ V ∗ → V ⊕ V ∗ be given by
S(x, f ) = (T (v), T ∗ (f )). Determine, with proof, the minimal polynomial of S.
127. (Oregon F18)Let V be a 3-dimensional vector space over Q. Suppose T :
V → V is a linear transformation, and T x = y, T y = z, T z = x + y for certain
x, y, z ∈ V, x ̸= 0. Prove that x, y, z are linearly independent.

4.4 Uncategorized
128. (OSU W19)Let A be nilpotent matrix. Prove that the trace of A is zero.
129. (Denver W19)Let m be an integer. Let α1 , α2 , . . . , αm , and β1 , β2 , . . . , βm
be two lists of vectors in real inner-product vector space V . Prove that if
∀i, j ≤ m; ⟨αi , αj ⟩ = ⟨βi , βj ⟩
then the subspaces V1 = span(α1 , . . . , αm ) and V2 = span(β1 , . . . , βm ) are iso-
morphic.

5 Vector and Dual Spaces

This section will cover vector and dual space properties. Dimensionality will
come into play for some arguments, but this is mainly on how to decompose
vector spaces.

5.1 Appetizers
130. (KS 56)Let {v1 , v2 , . . . , vn } be a basis for a vector space V over R. Show
that if w is any vector in V , then for some choice of sign ±, {v1 ± w, v2 , . . . , vn }
is a basis for V .
131. (KS 60)Let V be a vector space, U, W be finite dimensional subspaces of
V . Show that dim(U + W ) = dim U + dim W − dim U ∩ W .
132. (Denver W19)Complete the following:
a Let V1 , V2 be two non-trivial subspaces of the vector space V over F. Show
that there exists v ∈ V such that v ̸∈ V1 and v ̸∈ V2 .
b Show the result holds for any s non-trivial subspaces.
c Scully addition The above is not always true; provide some conditions
to ensure it is true.

5.2 Entrees
133. (KS 57)Let V be a finite dimensional vector space over the field F. Let
V ∗ be the dual space of V . Show that V ∼
= V ∗.
134. (KS 58)Let V be a vector space of the field F. Let V ∗ be the dual space
of V and let V ∗∗ be the dual space of V ∗ . Show there is an injective linear
transformation φ : V → V ∗∗ .
135. (KS 59)Let V be a finite dimensional vector space and let W be a subspace
of V . Show that dim V W = dim V − dim W .


136. (KS 63)Let V be the vector space over the field R of functions from f :
R → R. Let U be the subspace of even functions, W be the subspace of odd
functions. Show that V = U ⊕ W .
137. (KS 65)Let A, B, C be subspaces of the nonzero vector space V satisfying:
V = A ⊕ B = B ⊕ C = A ⊕ C. Show that there exists a W ≤ V, dim W = 2
such that dim W ∩ A = dim W ∩ B = dim W ∩ C = 1.
138. (OSU S21)Let A, B ∈ Mm,n (C) such that Range(A∗ ) ∩ Range(B ∗ ) = {0}
and B ∗ A = 0.
 a Prove that Range(A) ⊥ Range(B).
b Prove that Rank(A + B) = Rank(A) + Rank(B).

139. (Denver W21)Suppose that V is an inner product space, real or complex.

a Prove that if u, v, w ∈ V then

1 ∥w − u∥2 + ∥w − v∥2 ∥u − v∥2

∥w − (u + v)∥2 = −
2 2 4

b Suppose that S is a subset of V such that u, v ∈ S implies 21 (u + v) ∈ S.

Let w ∈ V . Prove there is at most one u ∈ S that is closest to w. Namely,

∀y ∈ S, ∥w − u∥ ≤ ∥w − y∥

140. (OSU W21)Let V, W be finite dimensional vector spaces over F and p :

V → W be an F-linear map. Let V ∗ , W ∗ be the dual spaces for V, W re-
spectively. [Namely V ∗ = {ϕ : V → F : ϕ is linear }, and similarly for W ∗ .]
Moreover, consider the dual map p∗ : W ∗ → V ∗ defined by p∗ (ϕ)(v) = ϕ(p(v))
for any ϕ ∈ W ∗ , v ∈ V .

a Explain (show) why p∗ is well-defined linear transformation.

b Prove p is surjective if and only if p∗ is injective.

141. (OSU F20)Complete the following:

a The rank of a complex square matrix is defined by the dimension of its
column space. Give a direct proof that rank(A + B) ≤ rank(A) + rank(B).
b For a complex n × 1 column vector u with the Euclidean norm ∥u∥ = 1,
define Q = I − uu∗ , where u∗ is the conjugate transpose. Prove that
rank(Q) = n − 1.
142. (UMass F20)Let A be an arbitrary n × n matrix.
a If the characteristic polynomial is defined as pA (t) = det(tI − A) = tn +
c1 (A)tn−1 + · · · + cn (A), what does the constant term cn (A) represent?
[With some justification.]
b The Cayley-Hamilton theorem says that A satisifies its own characteristic
polynomial, namely pA (A) = 0. Assuming cn (A) ̸= 0, express A−1 as a
polynomial in A.
143. (UMass F20)Given a matrix M , let ker(M ) denote the kernal / nullspace
of the matrix; Im(M ) denote the image or range.
a Let M be a 15 × 15 matrix. Can ker(M ) = Im(M ) ?
 b If the rank of M 2 equals the rank of M , what do you know about Im(M )∩
ker(M )?
144. (Oregon F18)Suppose that V is a finite dimensional vector space over F,
V ∗ its dual space. Suppose W ⊆ V is a subspace; the annihilator of W , denoted
Ann(W ), is the subspace of V ∗ consisting of linear transformations that vanish
on W .
a Show that W is the intersection of the null spaces of the elements of its
annihilator.
b Suppose now that T : V → V is a linear operator. If V = RT ⊕ W and
W is T -invariant, show that W ⊆ NT .
c Let T t : V ∗ → V ∗ be given by T t (l) = l ◦ T . Prove that Z ⊆ V is a
T -invariant subspace if and only if Ann(Z) is T t -invariant.

5.3 Main Course

145. (KS 61)Let V be a finite dimensional vector space over a field F, U a
subspace of V . Show that there is a subspace of W of V such that V = U ⊕ W .
146. (KS 62, Denver F20, OSU F19)Let V be the vector space of n × n matrices
over the field R. Let U be a subspace of V consisting of symmetric matrices
and W the subspace of V consisting of skew-symmetric matrices. Show that
V = U ⊕ W.
147. (KS 64)Let U, V, W be vector spaces over a field F. Let S : U → V, T :
V → W be linear transformations such that T ◦ S = 0, the zero map. Show
that:

dim(W Im T ) − dim(ker T Im S ) + dim ker S = dim W − dim V + dim U.

 

148. (KS 66)Let

L−1 =0 0L 1 L Ln−1 L =0
V−1 = 0 −−−−→ V0 −−→ V1 −−→ · · · −−−→ Vn −−n−−→ Vn+1 = 0

be a sequence of finite dimensional vector spaces and linear transformations

with Li+1 ◦ Li = 0 for each 0 ≤ i ≤ n.
Therefore, the quotients Hi = ker(Li ) Im(L ) are defined for each 0 ≤ i ≤ n.

i−1
Prove that i (−1)i dim Vi = i (−1)i dim Hi .
P P
149. (KS 67)Let V be a finite dimensional vector space, V ∗ its dual space. If
(,̇)˙ is a non-degenerate bilinear form on V and W is a subspace of V . Define
W ⊥ = {v ∈ V | ∀w ∈ W, (v, w) = 0}. Show that if X, Y are subspaces of V
⊥
with Y ⊆ X then X ⊥ ⊆ Y ⊥ . Furthermore, Y ∼ X
 ∗
⊥ =( Y) .
X
150. (Denver F21, UCF F21) Let V, W be finite dimensional real inner product
spaces and let T : V → W be a linear map. Fix w ∈ W . Show that the
elements v ∈ V for which the Euclidean norm ∥T v − w∥ is minimal are exactly
the solutions to the equation T ∗ T v = T ∗ w.
151. (Denver F20)Let V be an inner product space. Recall a linear map P is
said to be a projection if and only if P 2 = P . Also, P is said to be an orthogonal
projection if and only if Im(P ) ⊥ Null(P ).
a Let P be a projection. Prove that P is an orthogonal projection if and
only if P is diagonalizable in an orthogonal basis.

b Let P be a projection. Prove that P is an orthogonal projection if and

only if ∥P v∥ ≤ ∥v∥ for every v ∈ V .
152. (Denver F20) Recall that trace(At B) defines an inner product on Mn (R).
Let S be the subspace of symmetric matrices, and N be the subspace of skew-
symmetric matrices.
a Prove that Mn (R) = S ⊕ N .
b Prove that S ⊥ N with respect to the inner product given above.
c Let A ∈ Mn (R). Prove that
1
min ∥A − S∥ = ∥A − At ∥
S∈S 2
.
153. (Denver W20)Let V, W be finite-dimensional inner product spaces over
C such that dim V ≤ dim W , prove that there is a linear map T : V → W
satisfying that

∀u, v ∈ V, ⟨T (u), T (v)⟩W = ⟨u, v⟩V

.
154. (OSU F19)Let M be the real vector space of real n × n matrices. Let S
be the subset of symmetric matrices; A the subset of skew-symmetric matrices.
a Show S, A are subspaces of M, and that M = S ⊕ A.

b Show that the map ⟨X, Y ⟩ = trace(XY t ) is an inner product on M.

c Show that S ⊥ A, and find orthonormal bases for each. Determine their
dimensions.
d Given X ∈ M determine a formula to compute the matrix Y ∈ A closest
to X.

155. (PSU W19)Let E, F be finite dimensional vector spaces. Let L : E →

E, T : F → F be linear maps. Assume that the minimal polynomials of L, T are
coprime. Consider the map H = (L, T ) : E × F → E × F given by H(e, f ) =
(L(e), T (f )). Show that if V ⊆ E × F is a subspace such that H(V ) ⊆ V ,
πE (V ) = E and πF (V ) = F , then V = E × F .
5.4 Uncategorized
156. (Denver W19)Let A, B be m × n and n × p matrices over R respetively.
a Prove that dim(Null(AB)) ≤ dim(Null(A)) + dim(Null(B)).

b Prove that rank(A) + rank(B) ≤ rank(AB) + n.

6 Spectral Theory
This section covers Spectral Theory, dealing with properties of the spectral
decomposition and spectral radius. Will also cover singular values. Using or-
thogonality of a linear functional will assist in many problems.

6.1 Appetizers
157. (OSU F21)Let n be a positive integer, let A = (aij )ni,j=1 be an n × b matrix
with aij = 2 when i = j, −1 when j = i ± 1 and else 0. Essentially,

−1
 
2 0 ... 0
 .. 
−1
 2 −1 . 0
A= 0
 .. 
 −1 2 . 0
 . .. .. ..
 ..

. . . −1
0 0 ... −1 2

Prove that every eigenvalue of A is a positive real number.

158. (Denver F21) For each of the following statements give either a counterex-
ample or a reason why it is true.
a For every real matrix A there is a real matrix B with B −1 AB diagonal.
b For every symmetric real matrix A there is real matrix B with B −1 AB
diagonal.

c For every complex matrix A there is complex matrix B with B −1 AB

diagonal.
d For every symmetric complex matrix A there is a complex matrix B with
B −1 AB diagonal.
159. (OSU S21)Let n be an odd positive integer. Let A be an n × n real
orthogonal matrix such that det(A) > 0. Prove that there is a nonzero vector v
such that Av = v.
160. (Denver W21, OSU W19)Let A be nonsingular real n × n matrix. Prove
that there exists a unique orthogonal matrix Q and a unique positive definite
symmetric matrix B such that A = QB.
161. (Denver W21)Let M2 (C) be the set of 2-by-2 matrices
 with coefficients in

A N
C and A ∈ M2 (C). Define S = {N ∈ M2 (C)|B = is diagonalizable
0 A
}.
 
λ1 0
a Suppose that A = , with λ1 ̸= λ2 . Prove that S is a two
0 λ2
dimensional subspace of M2 (C). [Hint: first consider B ∈ S and find
conditions on N such that B is diagonalizable.]

b Prove that for any A ∈ M2 (C) with two distinct eigenvalues λ1 , λ2 , S

is a two dimensional subspace of M2 (C). [Hint: transform this into the
previous case.]
162. (UMass W21)Let T be a real orthogonal n × n matrix with T t T = id,
where T t is the transpose matrix.

a Prove that det(T ) = ±1.

b Prove that for every x ∈ Rn , ∥x∥ = ∥T x∥.
c Prove that the only possible real eigenvalues of T are ±1.
d If n = 3 and det(T ) = 1, prove that 1 is an eigenvalue of T .

e For n = 4, write an orthogonal matrix without eigenvectors. Justify your

answer.
163. (Denver W20)Let A be the 3×3 matrix given by: A(1, 1, 1) = (−3, −3, −3), A(−1, 1, 0) =
(0, 0, 0), A(1, 1, −2) = (2, 2, −4).
a What are the eigenvalues and associated eigenvectors of A? Can we use
the set of eigenvectors as a basis for R3 ? Why or why not? If yes, does
this basis have any special properties?
b Calculate A2020 (0, 2, 1).
c Does the linear system Ax = b have a solution for any b ∈ R3 ? If so, why?
If not, for what kind of b ∈ R3 is Ax = b solvable ?
d Determine whether the matrix A has the following properties with expla-
nation.
(a) diagonalizable
(b) invertible
(c) orthogonal
(d) symmetric
164. (PSU W20)Assume that A, B are n × n real, symmetric, positive semidef-
inite matrices. Prove that det(A + B) ≥ max{det A, det B}.
6.2 Entrees
165. (KS 17)A square matrix A over C is Hermitian if A∗ = A. Prove the
eigenvalues of a Hermitian matrix are all real.
166. (KS 18)Solve the following:

a Prove that a 2 × 2 scalar matrix A over a field F has a square root (i.e.,
a matrix B such that B 2 = A).
b Prove that a real symmetric matrix having the property that every neg-
ative eigenvalue occurs with even multiplicity has a square root. [Hint:
Use (a).]

167. (OSU W21, Denver W19)Suppose A ∈ Mm,n (C), B ∈ Mn,m (C). [For (a)
assume m = n, for the rest assume n, m are arbitrary.]
a Prove that if AB, BA are similar matrices provided either A or B are
invertible. Show that if A and B are not invertible, then AB need not be
similar to BA.

b Show that if λ ̸= 0 is an eigenvalue of AB then it is an eigenvalue of BA.

c Prove that the eigenspaces of AB, BA corresponding to the eigenvalue
λ ̸= 0 have the same dimension.
d Prove that the characteristic polynomials for the matrices AB and BA
satisfy λn pAB (λ) = λm pBA (λ), ∀λ ∈ C.
168. (Denver W21)Define Rn×n to be the space of all real n-by-n matrices.
Suppose S ∈ Rn×n and define the linear mapping T : Rn×n → Rn×n by T (P ) =
P S + SP .

a Prove that if λ is an eigenvalue of S, u is the corresponding eigenvector

and u ∈ Null T P , then P u is also an eigenvector of S with eigenvalue −λ.
b Prove that if S is symmetric positive definite, then the mapping T is
injective.
169. (PSU W21)Let Mn (C) be the vector space of n × n matrices with entries
in C. Assume A, B ∈ Mn (C). Show that if 0 ̸= X ∈ Mn (C) and AX = XB,
then A, B have a common eigenvalue.
170. (Denver W19)Solve the following:
a Let x, y be distinct eigenvectors of a matrix A such that x + y is also an
eigenvector of A. Is x − y necessarily an eigenvector of A? Prove or give
a counterexample.
b Let S, T be linear operators on finite-dimensional vector space over C.
Prove that ST, T S have the same eigenvalues.
171. (PSU W21)Let A ∈ Mn (R) and let λ1 ≤ λ2 ≤ · · · ≤ λn be the eigenvalues
of AT A, where √
AT denotes the
√ transpose. Show that any eigenvalues µ ∈ C of
A must satisfy λ1 ≤ |µ| ≤ λn .
172. (OSU F20)Let A be an n × n real matrix.
a Briefly explain why there exist an orthogonal matrix B and an (upper)
triangular matrix C such that A = BC.
b Prove the Hadamard inequality:
n
Y
| det(A)| ≤ ∥aj ∥
j=1

, where aj denotes the j-th column of A.

c Show the equality in (b) holds if and only if A is an orthogonal matrix or

aj = 0 for some j.
173. (PSU F20)Let V be a finite-dimensional inner product space over C, let
T : V →√V be an invertible linear transformation. Prove that the operator
S = T −1 T ∗ T is unitary if and only if T is normal.
174. (PSU W19)Let A, B be symmetric real-valued matrices. Assume all eigen-
values of A are positive; and all eigenvalues of B are non-negative. Prove that
all the eigenvalues of the matrix AB are real and non-negative.

6.3 Main Course

175. (KS 70)Let ⟨x, y⟩ be a Hermitian inner product defined on complex vector
space V . Let φ : V → V be a normal operator. Prove that V contains an
orthonormal basis of eigenvectors for φ.
176. (OSU W21 + Denver W21)A matrix U ∈ Mn,n (C) is called unitary if
U U ∗ = In . Recall that unitary matrices are diagonalizable. Prove that if
unitary matrices U1 , U2 ∈ Mn,n (C) commute, then they are simultaneously
diagonalizable.
177. (Denver W21 + OSU W21)Let S, T be two normal operators in the complex
finite dimensional inner product space V such that ST = T S. Prove that these
is a basis for V consisting of vectors that are eigenvectors of both S and T .
178. (Denver S19)Let V be an n-dimensional inner product space over F.
a Suppose T ∈ L(V ) and U is a subspace of V . Prove or disprove: U ⊥ is
invariant under T ∗ if U is invariant under T .
b Let T1 , T2 be self-adjoint operators on V . Prove or disprove: T1 T2 + T2 T1
is also self-adjoint.

c let T be a self-adjoint operator on V . Show that T is a nonnegative self-

adjoint operator on V if and only if the eigenvalues of T are all nonnegative
real numbers.
179. (Denver S19)Solve the following:
2
a Let A ∈ Mn
(F) such
 that A = A. Show that A is similar to the diagonal
I 0
matrix C = r , for 0 ≤ r ≤ n.
0 0
b Let A, B ∈ Mn (F), such that A2 = A, B 2 =B, AB= BA. Suppose P0 is
Ir 0
an invertible matrix such that P0−1 AP0 = . Let B0 = P0−1 BP0 .
0 0
 
B1 0
Show that B0 is in the form of B0 = , where B1 is of order r
0 B2
and B12 = B1 , B22 = B2 .
c Furthermore, show there exists an invertible matrix P such that P −1 AP, P −1 BP
are both diagonal, and the diagonal entries are 0, 1 for both.[Hint, start
with P0 from (b). Let Q1 , Q2 be invertible matrices for B1 , B2 . Construct
P from these matrices.]
180. (OSU W21)Let A ∈ Mn,n (C) be a matrix with entrees aij , i, j ∈ {1, . . . , n}.
a Prove that if λ ∈ C is an eigenvalue of A, then there exists k ∈ {1, . . . , n}
such that X
|λ − akk | ≤ |akj |
j̸=k
.
b Use (a) to prove that any complex polynomial p(z) = z n + an−1 z n−1 +
· · · + a1 z + a0 , all roots are located in the disk D = {z ∈ C : |z| ≤ R}
where R = max{|a0 |, 1 + |a1 |, . . . , 1 + |an−1 |}. Hint, consider the block
matrix  
0 −a0
 −a1 
A=
 
.. 
In−1 . 
−an−1
181. (OSU F20)A real symmetric n × n matrix A is called a positive semi-
definite if xt Ax ≥ 0 for all x ∈ Rn . Prove that A is positive semi-definite if and
only if tr(AB) ≥ 0 for every real symmetric positive semi-definite n × n matrix
B.
182. (PSU F20)Let Mn (C) be the vector space of complex n × b matrices. Let
A ∈ Mn (C) be any matrix. Define T : Mn (C) → Mn (C) by T (B) = AB.
Prove that T is diagonalizable if and only if A is.
183. (Denver F20)Let V be a finite dimensional inner product space over C.
Let T : V → V be a self-adjoint operator on V . Suppose µ ∈ C, ε > 0 are given
and assume there is a unit vector v ∈ V such that ∥T (v) − µv∥ ≤ ε. Show that
there is an eigenvalue λ ∈ Spec(T ) such that |λ − µ| ≤ ε.
184. (Denver W20)Let V be a real finite-dimensional inner product space. Let
T : V → V be a linear transformation. Assume for all v, w ∈ V we have
⟨T v, w⟩ = ⟨v, T w⟩.
 a Prove that if λ, µ are distinct eigenvalues of T then the corresponding
eigenspaves Eλ , Eµ are orthogonal.
b If W is a subspace of V , prove that T (W ) ⊆ W implies that T (W ⊥ ) ⊆
W ⊥.
c Prove that these exists an eigenvector v1 ∈ V for T with associated (real)
eigenvalue λ1 . [Do not use a big theorem; prove directly. You may utilize
fundamental theorem of algebra.]
d Prove that there exists an orthonormal basis of V consisting of eigenvectors
for T .
185. (OSU W19)Let A ∈ Mn (C) and b a complex n × 1 vector.
a Suppose y is a nonzero complex 1 × n vector. Show that there is a nonzero
polynomial q(x) and eigenvalue λ of A such that yq(A) is nonzero, and
(yq(A))A = λ(yq(A)).
b Prove that Rank[b, Ab, A2 b . . . An−1 b] = n if and only if Rank[A−λIb] = n
for every eigenvalue λ of A.
186. (PSU W19)Assume that a square n × n matrix A has non-negative entries,
ie Aij ≥ 0. Prove that:
n
X
min Aij ≤ max |λi |
1≤i≤n 1≤i≤j
j=1

, where λi are the eigenvalues of A.

6.4 Uncategorized
187. (PSU W20)Does there exist a 3×3 matrix with entris in C which commutes
with  
5 0 0
A = 0 1 1
0 0 1
but is not of the form f (A) for some polynomial f ∈ C[x]?
188. (OSU F19)Let A be an m × n complex matrix, B a p × q complex matrix.
The tensor product of A and B is the mp × nq matrix given block form as
 
a11 B a12 B ... a1n B
 a21 B a22 B . . . @a2n B 
A⊗B = .
 
. .. .
 .. .. . .. 
am1 B am2 B ... amn B
.

a Let C, D be two complex matrices of size n × r and q × s respectively.

Show that (A ⊗ B)(C ⊗ D) = (AC) ⊗ (BD).
 b Suppose A, B are invertible square matrices. Show that (A ⊗ B)−1 =
A−1 ⊗ B −1 .
c Suppose now that A is an m × m complex matric of eigenvalues {λi }m i=1 ,
and that B is a p × p complex matrix of eigenvalues {µj }pj=1 . Prove that
A ⊗ B has the eigenvalues {λi µj }1≤i≤m,1≤j≤p .

189. (PSU S19)Let A ∈ Mn (R) and self-adjoint. Assume the diagonal entries
of A are n − 1, while off-diagonal elements satisfy |Aij | < 1 for i ̸= j. Denote
B = A + A−1 − 2I. Prove that the limit

P = lim exp(−tB)
t→∞

, exists and is equal to the orthogonal projector to the eigenspace

E1 = ker(A − I)