Linear Operators as Matrices

Sushmitha P

1 Basics of Vector spaces

1. (V, +, .) is a vector space over a field F, if for x, y ∈ V , there is a unique x + y ∈ V
and for α ∈ F, αx ∈ V such that

• (V, +) is an abelian group.

• 1.x = x for all x ∈ V .
• (αβ)x = α(βx).
• α(x + y) = αx + αy.
• (α + β)x = αx + βx.

2. A subset W of V is a called a subspace if (W, +, .) is a vector space.

3. W is a subspace of V if and only if 0 ∈ W and αx + y ∈ W for all α ∈ F, x, y ∈ W .

4. W1 ∩ W2 , W1 + W2 are subspaces, whenever W1 and W2 are subspaces.

5. If S ⊆ V , then W = span(S) is a subspace. In this case, S is said to generate W .

6. A set of vectors is said to be linearly dependent if the zero vector can be written
as a linear combination of those vectors with at least one of the co-efficients being
non-zero. A set which is not linearly dependent is linearly independent.

7. A set B is said to be a basis for V if B is linearly independent and generates V .

8. The number of elements in any two bases of V are the same and is called the
dimension of the vector space V .

2 Linear Transformations
1. Let V and W be two vector spaces over the same field. Then T : V −→ W is said
to be a linear transformation if T (αx + y) = αT (x) + T (y), for x, y ∈ V and α ∈ F.

2. Two important subspaces that arise from a linear transformation T are the range of
T : R(T ) = {T (x) ∈ W : x ∈ V } and the null space of T : N (T ) = {x ∈ V : T (x) =
0}.

3. If R(T ) and N (T ) are finite dimensional, then their dimensions are called the rank
and nullity of T respectively.

1
3 Rank-Nullity-Dimension Theorem
Let V and W be vector spaces and let T : V −→ W be linear. If V is finite dimensional,
then
dim(V ) = rank(T ) + nullity(T ).

1. T is one-one if and only if N (T ) = {0}.

2. If V and W are of the same dimension, T is one-one if and only if T is onto.

3. Two linear transformations that agree on a basis must be equal.

4. Suppose V and W are vector spaces with ordered bases {β1 , β2 , ..., βn } and {γ1 , γ2 , ..., γm },
then there are scalars aij such that the matrix representation of T : V −→ W is
given by T (βj ) = m
P
k=1 akj γk , 1 ≤ j ≤ n. We say A = [aij ] to be the matrix of T
with respect to {β1 , β2 , ..., βn } and {γ1 , γ2 , ..., γm }.

3.1 Some Problems

1. Let A be a 3 × 4 matrix and B be a 4 × 3 matrix with real entries such that AB is
non-singular. Which of the following are true?

• Nullity of A is 0.
• BA is non-singular.

2
2. Let T : R3 −→ R3 be such that T (1, 1, 1) = (1, −1, 1), T 2 (1, 1, 1) = (1, 1, 1) and
T 2 (1, 1, 2) = (1, 1, 1). What is the rank(T )?

3. Let T1 , T2 : R5 −→ R3 be such that rank(T1 ) = 3 and nullity(T2 ) = 2. Let T3 :

R3 −→ R3 be such that T3 ◦ T1 = T2 . What is rank(T3 )?

3
4. Choose the right options.

• There exists a 2 × 5 matrix A such that nullity(A) = 2.

• There exists a 2 × 5 matrix A such that nullity(A) = 0.
• There exists a 2 × 5 matrix A and a 5 × 2 matrix B such that AB = I2×2 .
• There exists a 2 × 5 matrix A such that N (A) = {x ∈ R5 : x1 = x2 , x3 = x4 =
x5 }.

5. Let V be a finite dimensional vector space over R and T : V −→ V be a linear map.

Can T be written as T2 ◦ T1 such that T1 : V −→ W and T2 : W −→ V and

• both T1 and T2 are onto

• both T1 and T2 are one-one
• T1 is onto and T2 is one-one
• T1 is one-one and T2 is onto.

4
4 Diagonalization
1. Why should we diagonalize a matrix?

2. T : V −→ V is a linear operator on V .

3. T is said to be diagonalizable if there exists an ordered basis with respect to which

the matrix of T is a diagonal matrix. (A matrix A is diagonalizable if it is similar
to a diagonal matrix.)

4. If there is a basis {x1 , x2 , ..., xn } such that T xj = λj xj , j = 1, 2, ..., n then matrix of

T is diagonal.

5. λ ∈ F is said to be an eigenvalue of T if there exists a v(6= 0) ∈ V such that T v = λv.

v is called the eigenvector corresponding to λ.

6. λ is an eigenvalue of T if and only if det(T − λI) = 0.

7. det(A − λI) is the characteristic polynomial of A.

8. T satisfies its characteristic polynomial. (Cayley-Hamilton Theorem)

9. The minimal polynomial of T is the monic polynomial of least degree satisfied by T .

10. Minimal polynomial divides the characteristic polynomial.

4.1 Problems on Eigenvalues

1. Suppose A is a 9 × 9 matrix and f (x) is the minimal polynomial of A. What is the
rank of f (A)?

2. Let A be an n × n complex matrix and B denote the inverse of (A + iI). Then all
the eigenvalues of (A − iI)B are

• purely imaginary
• of modulus one
• real
• of modulus less than one

5
3. Let T : Rn −→ Rn be such that T 2 = T − I. Which of the following is/are true?

• T is invertible
• T − I is not invertible
• T has a real eigenvalue
• T 3 = −I

4. For which of the following matrices is PA (x) − PA−1 (x) a constant?

!
3 3
•
2 4
!
4 3
•
2 3
!
3 2
•
4 3
!
2 3
•
3 4

5. T is an operator on the set of all n × n matrices such that T (A) = AT . The

eigenvalues of T are

6
  
0 0 0 1 0 0
0 0 0 0 1 0
 
0 0 0 0 0 1
6. What are the eigenvalues of A =  ?
 
1 0 0 0 0 0
 
0 1 0 0 0 0
0 0 1 0 0 0
i, −i, 1, −1

7. Let A be a 2 × 2 matrix over R such that det(A + I) = 1 + det(A). Which of the

following is true?(Part B)

• A=0
• det(A) = 0
• tr(A) = 0
• A is non-singular.

8. Let A be an n × n matrix, n ≥ 2, with characteristic polynomial xn−2 (x2 − 1). Then

• An = An−2
• rank(A) = 2
• rank(A) ≥ 2
• there exist non-zero vectors x and y such that A(x + y) = x − y.

7
 9. Let A and B be complex n × n matrices. Which of the following are true?

• If A, B and A + B are invertible, then A−1 + B −1 is invertible.

• If A, B and A + B are invertible, then A−1 − B −1 is invertible.
• If AB is nilpotent, then BA is nilpotent.
• Characteristic polynomials of AB and BA are equal if A is invertible.

4.2 Diagonalizability(Contd.)
1. T is diagonalizable if and only if there exists a basis of V consisting of eigenvectors
of T .

2. Eigenvectors corresponding to distinct eigenvalues are linearly independent. Hence

any matrix with distinct eigenvalues is diagonalizable.

3. The characteristic polynomial of any diagonalizable operator splits.

4. T is diagonalizable if and only if algebraic multiplicity of λ is equal to the geometric

multiplicity of λ for all λ an eigenvalue of T .

5. T is diagonalizable if and only if the minimal polynomial of T is a product of distinct

linear factors.

8
4.3 Problems on Diagonalizability
!
0 1
1. Is diagonalizable over R?
−1 0

2. Which of the following matrices are diagonalizable?

 
1 2 3
• 0 4 5
 
0 0 6
 
0 1 0
• −1 0 0
 
0 0 1
 
1 2 3
• 2 1 4
 
3 4 1
 
0 1 2
• 0 0 1
 
0 0 0

9
3. Which of the following are not diagonalizable? (Part B)
 
2 0 1
• 0 3 2
 
0 0 2
 
2 1 0
• 0 3 0
 
0 0 3
!
1 1
•
1 1
!
1 −1
•
2 4

4. A is an n×n matrix with one eigenvalue of multiplicity n. When is A diagonalizable?

(non-zero nilpotent matrices are not diagonalizable)

5. If A is such that A2 = A, then rank(A)=trace(A). Does there exist a 3 × 3 matrix

A such that A2 = A and trace(A)=4?

10
 !
a 0
6. Let A = , a, c ∈ R and c 6= 0. Then A is diagonalizable
c a

• for all values of a

• for no values of a
• if and only if a = c
• if and only if a = 0.

7. Which of the following are true for n ≥ 2?

• There exist A, B ∈ Mn (R) such that AB − BA = I.

• If AB = BA, then A is diagonalizable over R if and only if B is diagonalizable
over R.
• AB and BA have the same minimal polynomial over R.
• AB and BA have the same eigenvalues in R.

11
  
1 −1 1
8. Let M =  2 1 4 . Given 1 is an eigenvalue of M . Which of the following
 
−2 1 −4
are correct?

• (x − 1)(x + 4) is the minimal polynomial of M .

• (x − 1)2 (x + 4) is the minimal polynomial of M .
• M is not diagonalizable.
• M −1 = 41 (M + 3I).

9. Let A and B be n × n real matrices such that det(A) < 0 and det(B) > 0. For
0 ≤ t ≤ 1, consider C(t) = tA + (1 − t)B. Then

• C(t) is invertible for each t ∈ [0, 1].

• there exists a t0 ∈ (0, 1) such that C(t0 ) is singular.
• C(t) is not invertible for t ∈ [0, 1].
• C(t) is invertible only for finitely many values of t.

12
[]

13