Chapter 2

Matrices - Determinants - Systems of

Linear Equations
Department of Mathematics and Informatics
Hanoi University of Science and Technology
2025
MI1141 - Linear Algebra Chapter 2

Contents

Introduction 3

1 Matrices and Operations 3

1.1 Definition of Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.1 Special Types of Matrices . . . . . . . . . . . . . . . . . . . . . . 4
1.1.2 Special Square Matrices . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Matrix Equality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Matrix Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3.1 Matrix Addition . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3.2 Scalar Multiplication . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3.3 Matrix Multiplication . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3.4 Matrix Transpose . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3.5 Symmetric and Skew-Symmetric Matrices . . . . . . . . . . . . . 8
1.4 Elementary Row Operations . . . . . . . . . . . . . . . . . . . . . . . . . 9

2 Determinants 9
2.1 Permutations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2 Determinant of a Square Matrix . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Cofactor Expansion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4 Properties of Determinants . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.5 Computing Determinants Using Elementary Operations . . . . . . . . . . 13

3 Inverse Matrices 13
3.1 Definition and Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.2 Conditions for Invertibility . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.3 Methods for Finding Inverse Matrices . . . . . . . . . . . . . . . . . . . . 15
3.3.1 Method 1: Using Cofactors . . . . . . . . . . . . . . . . . . . . . . 15
3.3.2 Method 2: Solving Matrix Equations . . . . . . . . . . . . . . . . 15
3.3.3 Method 3: Elementary Row Operations . . . . . . . . . . . . . . . 16

4 Rank of a Matrix 17
4.1 Definition of Rank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.1.1 Properties of Rank . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.2 Row Echelon Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4.3 Computing Rank Using Elementary Operations . . . . . . . . . . . . . . 18

5 Systems of Linear Equations 19

5.1 Definition and Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
5.1.1 Matrix Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
5.2 Cramer Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
5.3 Conditions for Solutions and Solution Methods . . . . . . . . . . . . . . . 21
5.3.1 Elementary Operations on Systems . . . . . . . . . . . . . . . . . 21
5.3.2 Gaussian Elimination . . . . . . . . . . . . . . . . . . . . . . . . . 21
5.3.3 Gauss-Jordan Method . . . . . . . . . . . . . . . . . . . . . . . . 22
5.4 Homogeneous Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

2
MI1141 - Linear Algebra Chapter 2

Introduction
This chapter introduces matrices, determinants, and systems of linear equations. These
provide effective tools that help us understand the content of subsequent chapters.
Chapter Contents:
1. Matrices and Operations
2. Determinants
3. Inverse Matrices
4. Rank of a Matrix
5. Systems of Linear Equations

Convention: Throughout this chapter, K denotes either the set of real numbers R or
the set of complex numbers C.

1 Matrices and Operations

Matrices and their properties are central to linear algebra. Matrices are very useful
because they allow us to treat a table of many numbers as a single object, denote it by
a symbol, and represent computations with these symbols concisely and easily.

Learning Objectives
• Knowledge: Understand the concept of matrices, special types of matrices, matrix
equality, matrix operations, and elementary row operations on matrices.
• Skills: Practice matrix operations and elementary transformations proficiently.

1.1 Definition of Matrix

Definition 1.1 (Matrix). An m × n matrix is a rectangular array of numbers with m
rows and n columns of the form:
 
a11 a12 · · · a1n
 a21 a22 · · · a2n 
A =  ..
 
.. . . .. 
 . . . . 
am1 am2 · · · amn
where aij ∈ K. The number aij is called an element of matrix A, located at row i, column
j, for all 1 ≤ i ≤ m, 1 ≤ j ≤ n.
We write A = [aij ]m×n to indicate that A is a matrix with m rows, n columns, and
elements aij .
If K = R, then A is called a real matrix; if K = C, then A is called a complex matrix.
Example 1.1.  
1 2 3
A=
5 −4 6
is a 2 × 3 matrix with elements a11 = 1, a12 = 2, a13 = 3, a21 = 5, a22 = −4, a23 = 6.

3
MI1141 - Linear Algebra Chapter 2

1.1.1 Special Types of Matrices

• A matrix of size 1 × n is called a row matrix (or row vector).

• A matrix of size m × 1 is called a column matrix (or column vector).

• A matrix A = [aij ]m×n with aij = 0 for all i, j is called the zero matrix, denoted
θ.

• If the number of rows equals the number of columns (m = n), then A is called a
square matrix of order n.

Example 1.2.
 
1  
A = 4 is a column matrix, B = 1 2 3 4 is a row matrix
7
 
1 2 3
C = 4 5 6 is a square matrix of order 3
7 8 9

Note
Notation:

• Mm×n (K): the set of all m × n matrices with elements in K

• Mn (K): the set of all square matrices of order n with elements in K

1.1.2 Special Square Matrices

For a square matrix of order n:
 
a11 a12 · · · a1n
 a21 a22 · · · a2n 
A =  ..
 
.. .. . 
 . . . .. 
an1 an2 · · · ann

Definition 1.2 (Diagonal Elements). The elements a11 , a22 , . . . , ann are called diagonal
elements; they form the main diagonal of A.

Definition 1.3 (Special Square Matrices). 1. If aij = 0 for all i > j, then A is called
an upper triangular matrix.

2. If aij = 0 for all i < j, then A is called a lower triangular matrix.

3. If aij = 0 for all i ̸= j, then A is called a diagonal matrix.

4. If A is diagonal and all diagonal elements equal 1, then A is called an identity

matrix of order n, denoted In or En (or simply I or E).

4
MI1141 - Linear Algebra Chapter 2

 
1 2 3
Example 1.3. a) A = 4 5 6 is a square matrix of order 3 with diagonal elements
7 8 9
1, 5, 9.
 
10 0
b) B = 4  5 0 is a lower triangular matrix.
78 9
 
1 0 0
c) C = 0  −5 0 is a diagonal matrix.
0 0 3
 
  1 0 0
1 0
d) I2 = and I3 = 0 1 0 are identity matrices of orders 2 and 3.
0 1
0 0 1

1.2 Matrix Equality

Definition 1.4 (Matrix Equality). Two matrices A and B are equal if they have the
same size and their corresponding elements are equal.
   
2 1 2 1 5
Example 1.4. The matrices A = and B = are not equal because they
4 3 3 4 6
have different sizes.
   
1 5 x z 5 4
Example 1.5. Given A = and B = . Find x, y, z, t such that
−1 y 2 −1 5 t
A = B.
Solution: For A = B, we need: z = 1, y = 5, x = 4, t = 2.

1.3 Matrix Operations

1.3.1 Matrix Addition
Definition 1.5 (Matrix Addition). Given two matrices of the same size A = [aij ]m×n
and B = [bij ]m×n , their sum A + B is the m × n matrix defined by:

A + B = [aij + bij ]m×n

Thus, to add two matrices of the same size, we add their corresponding elements.

Example 1.6.      
1 2 3 2 −1 0 3 1 3
+ =
4 5 6 −3 1 4 1 6 10

Definition 1.6 (Negative and Subtraction). 1. The negative of matrix A = [aij ]m×n ,
denoted −A, is defined by −A = [−aij ]m×n .

2. The difference of two matrices A and B of the same size, denoted A−B, is defined
by:
A − B = A + (−B)

5
MI1141 - Linear Algebra Chapter 2

Proposition 1.1 (Properties of Matrix Addition). For all matrices A, B, C of the same
size:
1. Associativity: (A + B) + C = A + (B + C)
2. Commutativity: A + B = B + A
3. Zero element: A + θ = θ + A = A, where θ is the zero matrix
4. Inverse element: A + (−A) = (−A) + A = θ

1.3.2 Scalar Multiplication

Definition 1.7 (Scalar Multiplication). Given a matrix A = [aij ]m×n over K and a scalar
k ∈ K, the product of k and A is defined by:
kA = [kaij ]m×n
Thus, multiplying a scalar k by matrix A means multiplying k by each element of A.
Example 1.7.    
1 2 3 3 6 9
3 =
4 −5 6 12 −15 18
Proposition 1.2 (Properties of Scalar Multiplication). For all matrices A, B of the same
size and scalars k, l ∈ K:
1. k(A + B) = kA + kB
2. (k + l)A = kA + lA
3. k(lA) = (kl)A
4. 1A = A, (−1)A = −A
5. 0A = θ
6. kθ = θ
   
1 2 3 3 −1 4
Example 1.8. Given A = and B = . Find matrix X such that
4 5 0 2 1 −2
2X + A = B.
1 − 32 12
 
1
Solution: 2X = B − A ⇒ X = 2 (B − A) =
−1 −2 −1

1.3.3 Matrix Multiplication

Definition 1.8 (Matrix Multiplication). Suppose A = [aik ]m×p and B = [bkj ]p×n are
matrices of sizes m × p and p × n, respectively. The product AB is a matrix C = [cij ]m×n
of size m × n, where element cij (1 ≤ i ≤ m, 1 ≤ j ≤ n) is defined by:
p
X
cij = ai1 b1j + ai2 b2j + · · · + aip bpj = aik bkj
k=1

Element cij is computed by multiplying corresponding elements from row i of A with

elements from column j of B, then summing.

6
MI1141 - Linear Algebra Chapter 2

Note
The product AB is only defined when the number of columns of A equals the
number of rows of B. Moreover, matrix AB has the same number of rows as A and
the same number of columns as B.
 
  0 2
1 2 3
Example 1.9. Given A = and B = −2 1. Compute C = AB.
4 5 6
4 3
Solution:  
  0 2  
1 2 3  8 13
AB = −2 1 =

4 5 6 14 31
4 3

Proposition 1.3 (Properties of Matrix Multiplication). Suppose A, B, C are matrices

such that the operations in the following relations can be performed, and k ∈ K. Then:

1. IA = A, BI = B where I is an identity matrix of appropriate size

2. Associativity: (AB)C = A(BC)

3. Distributivity: A(B + C) = AB + AC, (B + C)A = BA + CA

4. k(AB) = (kA)B = A(kB)

Note
Important Notes

1. The product AB may exist while BA does not exist.

2. Matrix multiplication is not commutative: even if both AB and BA exist,

generally AB ̸= BA.

3. From AB = θ, we cannot conclude A = θ or B = θ.

Example 1.10. Given matrices:

 
1    
  −1 1 2 0 1
A= 1 2 3 , B = −2 , C= , D=
0 2 4 0 0
2

Then:
 
1 2 3
i) AB = [3] and BA = −2 −4 −6, so AB ̸= BA
2 4 6

ii) DC exists but CD does not exist

iii) DD = θ but D ̸= θ

7
MI1141 - Linear Algebra Chapter 2

Note
Matrix Powers and Polynomials

1. When A is a square matrix and m ∈ N∗ , we denote Am = |AA{z

· · · A}.
m times

2. Given polynomial p(x) = a0 xm + a1 xm−1 + · · · + am−1 x + am with ai ∈ K,

and A a square matrix of order n, then p(A) is defined by:

p(A) = a0 Am + a1 Am−1 + · · · + am−1 A + am In

where In is the identity matrix of order n.

 
1 2
Example 1.11. Given A = and polynomial p(x) = 2x2 − 3x − 1. Compute
−1 1
p(A).
Solution:  
2 −1 4
A =
−2 −1
       
2 −1 4 1 2 1 0 −6 2
p(A) = 2A − 3A − I = 2 −3 − =
−2 −1 −1 1 0 1 −1 −6

1.3.4 Matrix Transpose

Definition 1.9 (Transpose). The transpose of matrix A = [aij ]m×n , denoted At , is
defined by At = [bij ]n×m where bij = aji for all i = 1, 2, . . . , n and j = 1, 2, . . . , m.
We can obtain At from A by writing rows of A as columns of At (or columns of A as
rows of At ).
 
  1 4
1 2 3
Example 1.12. The transpose of A = is At = 2 5.
4 5 6
3 6
Proposition 1.4 (Properties of Transpose). Suppose A, B are matrices such that the
operations can be performed, and k ∈ K. Then:
1. (At )t = A

2. (A + B)t = At + B t

3. (kA)t = kAt

4. (AB)t = B t At

1.3.5 Symmetric and Skew-Symmetric Matrices

Definition 1.10 (Symmetric and Skew-Symmetric Matrices). A square matrix A of order
n is:
1. Symmetric if At = A

2. Skew-symmetric (or antisymmetric) if At = −A

8
MI1141 - Linear Algebra Chapter 2

Remark 1.1. If A = [aij ] is a symmetric (respectively skew-symmetric) square matrix

of order n, then aij = aji (respectively aij = −aji ) for all i, j = 1, 2, . . . , n. Moreover, all
diagonal elements of a skew-symmetric matrix equal zero.
Example 1.13.
   
1 2 −1 0 2 −1
A =  2 −2 3  is symmetric, B = −2 0 3  is skew-symmetric
−1 3 3 1 −3 0

1.4 Elementary Row Operations

Definition 1.11 (Elementary Row Operations). Given a matrix A, the following are
called elementary row operations:
1. Interchange two rows (or two columns)
2. Multiply a row (or column) by a nonzero scalar
3. Add a multiple of one row (or column) to another row (or column)
Note
Notation

• hi denotes row i, cj denotes column j

• hi ↔ hj (or ci ↔ cj ): interchange rows i and j (or columns i and j)

• λhi → hi (or λci → ci ): multiply row i (or column i) by λ ̸= 0

• hk + λhi → hk (or ck + λci → ck ): add λ times row i (or column i) to row k

(or column k)

Example 1.14.
     
4 5 6 1 2 3 1 2 3
h1 ↔h2 2h →h3
1 2 3 −−−−→ 4 5 6 −−3−−→ 4 5 6
7 8 9 7 8 9 14 16 18
 
1 2 3
−−−−−−−−−→ 0 −3 −6 
h2 +(−4)h1 →h2
h3 +(−14)h1 →h3 0 −12 −24

2 Determinants
Learning Objectives
• Knowledge: Understand the concept of permutations, determinants of square
matrices, and properties of determinants.
• Skills: Master computing determinants using definition, cofactor expansion, and
elementary operations.

9
MI1141 - Linear Algebra Chapter 2

2.1 Permutations
Definition 2.1 (Permutation). A permutation of degree n is a bijection f : {1, 2, . . . , n} →
{1, 2, . . . , n}.
Let Sn denote the set of all permutations of degree n.  
1 2 ··· n
A permutation of degree n is often written as f = , where
f (1) f (2) · · · f (n)
(f (1), f (2), . . . , f (n)) is a permutation of (1, 2, . . . , n). The number of permutations of
degree n is |Sn | = n!.

Definition 2.2 (Inversions and Sign). Given a permutation f of degree n:

1. A pair (f (i), f (j)) with 1 ≤ i < j ≤ n is called an inversion if f (i) > f (j).

2. If the number of inversions of f is even (or odd), then f is called an even permu-
tation (or odd permutation).

3. The sign of permutation f , denoted sign(f ), is defined by:

(
1 if f is even
sign(f ) =
−1 if f is odd
 
1 2 3
Example 2.1. Consider the permutation of degree 3: σ = , i.e., σ(1) = 3,
3 1 2
σ(2) = 1, σ(3) = 2.
The pair (σ(1), σ(2)) is an inversion since σ(1) = 3 > 1 = σ(2); but (σ(2), σ(3)) is not
an inversion since σ(2) = 1 < 2 = σ(3).
Permutation σ has 2 inversions: (σ(1), σ(2)) and (σ(1), σ(3)). Therefore, σ is an even
permutation and sign(σ) = 1.

Proposition 2.1. The number of inversions of f equals the number of inversions of f −1 ,

and:
sign(f −1 ) = sign(f )
Moreover, for permutations f and g of degree n:

sign(f ◦ g) = sign(f ) · sign(g)

2.2 Determinant of a Square Matrix

Definition 2.3 (Determinant). Given a square matrix of order n: A = [aij ]n×n . The
determinant of A, denoted det(A) or |A|, is defined by:
X
det(A) = |A| = sign(σ) · a1σ(1) a2σ(2) · · · anσ(n)
σ∈Sn

The determinant of a square matrix of order n is called a determinant of order n.

Note
The sum has n! terms; in each term, each row contributes one element, and each
column contributes one element.

10
MI1141 - Linear Algebra Chapter 2

Example 2.2 (Determinant of Order 2).

a11 a12
= a11 a22 − a12 a21
a21 a22

1 3
For instance: = 1 · 5 − (−2) · 3 = 11
−2 5
Example 2.3 (Determinant of Order 3).

a11 a12 a13

a21 a22 a23 = a11 a22 a33 + a12 a23 a31 + a13 a21 a32
a31 a32 a33
−a12 a21 a33 − a11 a23 a32 − a13 a22 a31
Sarrus’ Rule: Extend the first two columns to the right and compute products along
diagonals.
0 2 −1
Example 2.4. Compute: 1 4 5 = 0 + (−20) + 1 − 8 − 0 − 6 = −33
−2 −1 3

2.3 Cofactor Expansion

Definition 2.4 (Minor and Cofactor). Given a square matrix of order n: A = [aij ]n×n .
For each pair (i, j) with 1 ≤ i, j ≤ n:
• Let Mij denote the square matrix of order (n − 1) obtained from A by deleting row
i and column j.
• The cofactor of aij , denoted Aij , is defined by:

Aij = (−1)i+j det(Mij )

 
1 2 3
Example 2.5. Given A = 2 −1 0 . The cofactors are:
4 5 −3

−1 0 2 0
A11 = (−1)1+1 = 3, A12 = (−1)1+2 =6
5 −3 4 −3
2 −1
A13 = (−1)1+3 = 14, ...
4 5

Theorem 2.1 (Cofactor Expansion). Given a square matrix A = [aij ] of order n with
cofactors Aij , 1 ≤ i, j ≤ n:
1. For each fixed i (1 ≤ i ≤ n), expansion along row i:

det(A) = ai1 Ai1 + ai2 Ai2 + · · · + ain Ain

2. For each fixed j (1 ≤ j ≤ n), expansion along column j:

det(A) = a1j A1j + a2j A2j + · · · + anj Anj

11
MI1141 - Linear Algebra Chapter 2

Remark 2.1. These formulas allow computing a determinant of order n through de-
terminants of order (n − 1). In practice, choose a row or column with many zeros for
expansion.
 
1 2 3
Example 2.6. Compute det(A) for A = 2 −1 0  by expanding along row 2 (which
4 5 −3
has a zero):

det(A) = 2A21 + (−1)A22 + 0 · A23 = 2 · 21 + (−1) · (−15) = 57

2.4 Properties of Determinants

Theorem 2.2 (Properties of Determinants). For a square matrix A:
Property 1. det(A) = det(At )

Corollary 2.2.1. Any property of determinants true for rows is also true for columns.

Property 2. If two rows (or columns) are interchanged, the determinant changes
sign.

Corollary 2.2.2. If a matrix has two identical rows (or columns), its determinant is
zero.

Property 3. If all elements of a row (or column) are multiplied by scalar k, the
determinant is multiplied by k.

Corollary 2.2.3. 1. Common factors from a row (or column) can be factored out.

2. For a square matrix A of order n and scalar k: det(kA) = k n det(A)

3. If all elements of a row (or column) are zero, det(A) = 0.

Property 4. If two rows (or columns) are proportional, det(A) = 0.

Property 5. If row k (or column l) has elements akj = bj + cj for all j (or ail = ei + fi
for all i), then the determinant equals the sum of two determinants where row k (or
column l) is replaced by bj and cj (or ei and fi ) respectively.
Property 6. Adding a multiple of one row (or column) to another row (or column)
does not change the determinant.
Property 7. The determinant of a triangular matrix equals the product of its diag-
onal elements:
a11 a12 · · · a1n
0 a22 · · · a2n
.. .. . . . = a11 a22 · · · ann
. . . ..
0 0 · · · ann
Property 8. For square matrices A and B of the same order: det(AB) = det(A) ·
det(B)

12
MI1141 - Linear Algebra Chapter 2

2.5 Computing Determinants Using Elementary Operations

2 3 −97 −1
1 −1 0 1
Example 2.7. Compute D = .
4 5 −3 0
0 2 −52 1
Solution:
1 −1 0 1
h1 ↔h2 2 3 −97 −1
D −−−−→ −
4 5 −3 0
0 2 −52 1
1 −1 0 1
0 5 −97 −3
−−−−−−−→ −
h2 −2h1 →h2 0 9 −3 −4
h3 −4h1 →h3
0 2 −52 1
1 −1 0 1
h2 −2h4 →h20 1 7 −5
−−−−−−−→ −
0 9 −3 −4
0 2 −52 1
1 −1 0 1
0 1 7 −5
−−−−−−−→ −
h3 −9h2 →h3 0 0 −66 41
h4 −2h2 →h4
0 0 −66 11
1 −1 0 1
h4 −h3 →h4 0 1 7 −5
−− −−−−→ −
0 0 −66 41
0 0 0 −30
= −1 · 1 · (−66) · (−30) = −1980

Note
Combining Expansion and Elementary Operations When computing determinants,
combining elementary operations with cofactor expansion can reduce complexity.
For instance, use elementary operations to create zeros, then expand along that
row or column.

3 Inverse Matrices
Learning Objectives
• Knowledge: Understand invertible matrices, conditions for invertibility, properties
of inverse matrices, and methods for finding inverses.

• Skills: Master finding inverse matrices efficiently.

13
MI1141 - Linear Algebra Chapter 2

3.1 Definition and Properties

For n ∈ N∗ , Mn (K) denotes the set of square matrices of order n over K, and I (or E)
denotes the identity matrix of order n.

Definition 3.1 (Invertible Matrix). A matrix A ∈ Mn (K) is called invertible if there

exists a matrix B ∈ Mn (K) such that:

AB = BA = I

The matrix B is called the inverse matrix of A and is denoted B = A−1 .

Thus, AA−1 = A−1 A = I. Moreover, the inverse of A ∈ Mn (K), if it exists, is unique.
We denote by GLn (K) the set of all invertible matrices of order n.
   
1 2 −1 3 −2
Example 3.1. A = is invertible with inverse A = .
1 3 −1 1

Example 3.2. Suppose square matrix A satisfies A2 + A − I = θ. Then A2 + A = I, so

A(A + I) = (A + I)A = I. Therefore A−1 = A + I.

Proposition 3.1 (Properties of Inverse Matrices). 1. The identity matrix I is invert-

ible with I −1 = I.

2. If A ∈ GLn (K), then A−1 ∈ GLn (K) and (A−1 )−1 = A.

3. If A ∈ GLn (K), then At ∈ GLn (K) and (At )−1 = (A−1 )t .

4. If A ∈ GLn (K) and k ∈ K, k ̸= 0, then kA ∈ GLn (K) and (kA)−1 = k1 A−1 .

5. If A, B ∈ GLn (K), then AB ∈ GLn (K) and (AB)−1 = B −1 A−1 .

Note
To prove matrix X is invertible with inverse Y , we need only show XY = I and
Y X = I.

3.2 Conditions for Invertibility

Given A = [aij ] ∈ Mn (K) with cofactors Aij for i, j = 1, 2, . . . , n.
Let C = [Aij ]n×n .

Lemma 3.1. For all A ∈ Mn (K):

C t A = AC t = det(A) · I

Definition 3.2 (Nonsingular Matrix). A matrix A ∈ Mn (K) is called nonsingular if

det(A) ̸= 0.

Theorem 3.1 (Invertibility Criterion). A matrix A ∈ Mn (K) is invertible if and only if

A is nonsingular (i.e., det(A) ̸= 0).

Corollary 3.1.1. Matrix A ∈ Mn (K) is invertible if and only if there exists matrix B
such that AB = I or BA = I.

14
MI1141 - Linear Algebra Chapter 2

 
λ 1 1
Example 3.3. Find all values of parameter λ for which A =  1 λ 1  is invertible.
1 1 λ
Solution: A is invertible ⇔ det(A) ̸= 0. Computing:

det(A) = (λ − 1)2 (λ + 2)

So A is invertible ⇔ λ ̸= 1 and λ ̸= −2.

3.3 Methods for Finding Inverse Matrices

Problem: Given an invertible matrix A = [aij ] ∈ Mn (K). Find A−1 .

3.3.1 Method 1: Using Cofactors

1. Compute det(A).

2. Determine all cofactors Aij for all i, j.

3. Form matrix C = [Aij ]. Apply formula: A−1 = 1

det(A)
Ct
 
1 1 −1
Example 3.4. Find the inverse of A = 1 3 1 .
2 2 −1
Solution: det(A) = 2 ̸= 0, so A is invertible.
Computing cofactors and forming C:
 
−5 3 4
C t = −1 1 0
4 −2 2
 
−5 3 4
1
A−1 = −1 1 0
2
4 −2 2

Note
 
a b
Special Case: 2 × 2 Matrices For A = with ad − bc ̸= 0:
c d
 
−1 1 d −b
A =
ad − bc −c a

3.3.2 Method 2: Solving Matrix Equations

Consider the matrix equation AX = I where I is the identity matrix of order n. Since A
is invertible, the equation has unique solution X = A−1 .

15
MI1141 - Linear Algebra Chapter 2

 
1 2
Example 3.5. Find the inverse of A = .
3 4  
−1 a b
Solution: det(A) = −2 ̸= 0, so A is invertible. Let A = .
c d
Then:     
1 2 a b 1 0
=
3 4 c d 0 1
This gives systems:
( (
a + 2c = 1 b + 2d = 0
and
3a + 4c = 0 3b + 4d = 1

Solving: a = −2, c= 23 , b =1, d = − 21 .

−2 1
Therefore: A−1 = 3
2
− 12

3.3.3 Method 3: Elementary Row Operations

Steps:

1. Write the identity matrix of order n after matrix A to get an n × 2n matrix: [A|I].

2. Use elementary row operations to transform [A|I] into [I|B] (matrix A becomes
identity I, and I becomes matrix B). Then B = A−1 :
row operations
[A|I] −−−−−−−−→ [I|B] ⇒ B = A−1
 
1 1 −1
Example 3.6. Find the inverse of A = 1 3 1 .
2 2 −1
Solution:
 
1 1 −1 1 0 0
 1 3 1 0 1 0 
2 2 −1 0 0 1
 
1 1 −1 1 0 0
−−−−−−−→  0 2 2 −1 1 0 
h2 −h1 →h2
h3 −2h1 →h3 0 0 1 −2 0 1
1 0 0 − 52 32 2
 
continue
−−−−→  0 1 0 − 12 12 0 
0 0 1 2 −1 1
 
−5 3 4
1
A−1 = −1 1 0
2
4 −2 2

16
MI1141 - Linear Algebra Chapter 2

4 Rank of a Matrix
Learning Objectives
• Knowledge: Understand the concept of matrix rank and its properties.
• Skills: Master computing matrix rank using elementary operations.

4.1 Definition of Rank

Consider an m × n matrix:
 
a11 a12 · · · a1n
 a21 a22 · · · a2n 
A =  ..
 
.. ... .. 
 . . . 
am1 am2 · · · amn

Definition 4.1 (Submatrix and Minor). Let k be a positive integer with k ≤ min{m, n}.
A square matrix of order k obtained from A by deleting (m − k) rows and (n − k) columns
is called a submatrix of order k of A. Its determinant is called a minor of order k
of A.
Remark 4.1. A submatrix of order k is formed by elements at the intersection of k rows
k
and k columns of A. Therefore, A has Cm · Cnk submatrices of order k (not necessarily
distinct).
 
1 2 3 4
Example 4.1. Matrix A = 5 6 7 8  has size 3 × 4.
6 8 10 12
A has C33 · C43 = 4 minors of order 3.
A has C32 · C42 = 18 minors of order 2.
Definition 4.2 (Rank). The rank of matrix A, denoted rank(A) or r(A), is the highest
order of nonzero minors of A.
Convention: The rank of the zero matrix is 0.
Thus, rank(A) = r if and only if A has a nonzero minor of order r, and all minors of
order greater than r equal zero.
Example 4.2. For the matrix in the previous example, all minors of order 3 equal 0,
so rank(A) < 3. However, one minor of order 2 equals 4 (nonzero), so rank(A) ≥ 2.
Therefore rank(A) = 2.

4.1.1 Properties of Rank

1. If A is an m × n matrix, then 0 ≤ rank(A) ≤ min{m, n}.
2. For any matrix A: rank(At ) = rank(A).
3. For a square matrix A of order n:
• If det(A) ̸= 0, then rank(A) = n
• If det(A) = 0, then rank(A) < n

17
MI1141 - Linear Algebra Chapter 2

4.2 Row Echelon Form

A zero row of a matrix is a row whose elements are all zero.

Definition 4.3 (Row Echelon Form). A matrix is in row echelon form if it satisfies:

1. All zero rows (if any) are below all nonzero rows.

2. For any two nonzero rows, the leading nonzero element (pivot) of the upper row is
to the left of the leading element of the lower row.

Example 4.3.  
0 1 2 −3 4
0 0 0 3 5
 
A=0 0 0 0 4  is in row echelon form
0 0 0 0 0
0 0 0 0 0
 
0 1 2 −3 4
0 0 0 0 4
 
0 0
B= 0 3 5 is NOT in row echelon form
0 0 0 0 0
0 0 0 0 0
Proposition 4.1. The rank of a matrix in row echelon form equals the number of its
nonzero rows.
 
0 1 2 −3 4
0 0 0 3 5
 
Example 4.4. The rank of A = 0 0 0 0 4 is 3.

0 0 0 0 0
0 0 0 0 0

4.3 Computing Rank Using Elementary Operations

Since elementary operations do not change whether minors are zero or nonzero, they do
not change the rank of a matrix.

Proposition 4.2. The rank of a matrix is unchanged when elementary operations are
applied.

Method for finding rank:

elementary operations
A −−−−−−−−−−−−→ B (row echelon form) ⇒ rank(A) = rank(B) = # nonzero rows of B
 
2 3 4 −1 2
1 2 3 1 0
Example 4.5. Find the rank of A = 
3 5 7 0 2.


1 1 1 0 2

18
MI1141 - Linear Algebra Chapter 2

Solution:
 
1 2 3 1 0
h1 ↔h2 2 3 4 −1 2
A −−−−→  3

5 7 0 2
1 1 1 0 2
 
1 2 3 1 0
0 −1 −2 −3 2
−−−−−−−→  
h2 −2h1 →h2 0 −1 −2 −3 2
h3 −3h1 →h3
h4 −h1 →h4 0 −1 −2 −1 2
 
1 2 3 1 0
0 −1 −2 −3 2
−−−−−−→  
h3 −h2 →h3 0 0 0 0 0
h4 −h2 →h4
0 0 0 2 0
 
1 2 3 1 0
h ↔h4  0 −1 −2 −3 2

−−3−−→ 0 0 0 2 0
0 0 0 0 0
Therefore, rank(A) = 3.

5 Systems of Linear Equations

Learning Objectives
• Knowledge: Understand systems of linear equations, their forms, Cramer sys-
tems, homogeneous systems, conditions for existence of solutions, and Gaussian
elimination.
• Skills: Master solving linear systems using Gaussian elimination and Gauss-Jordan
methods, understand properties of special systems.

5.1 Definition and Forms

Definition 5.1 (System of Linear Equations). A system of m linear equations in n
unknowns has the form:


 a11 x1 + a12 x2 + · · · + a1n xn = b1

a21 x1 + a22 x2 + · · · + a2n xn = b2

..


 .

a x + a x + · · · + a x = b
m1 1 m2 2 mn n m

where aij , bi ∈ K for i = 1, 2, . . . , m, j = 1, 2, . . . , n are given constants, and xj for

j = 1, 2, . . . , n are unknowns.
The numbers aij are called coefficients, and bi are called constant terms.
If bi = 0 for all i = 1, 2, . . . , m, the system is called homogeneous.
Definition 5.2 (Solution). A solution of the system is an n-tuple (t1 , t2 , . . . , tn ) ∈ K n
such that substituting xj = tj for all j into the equations yields identities.

19
MI1141 - Linear Algebra Chapter 2

Example 5.1. The system


3x1 + 2x2 + 5x3 + 5x4 = 9

x1 + x2 + x3 − x4 = −2

−x1 + 3x3 + 2x4 = 5


is a nonhomogeneous linear system with 3 equations and 4 unknowns. The 4-tuple

(−1, 1, 0, 2) is a solution.

5.1.1 Matrix Form

For the system above:

• A = [aij ]m×n is called the coefficient matrix

 
b1
 b2 
• B =  ..  is the constant column
 
 . 
bm

• Ā = [A|B] is the augmented matrix

 
x1
 x2 
• X =  ..  is the unknown column
 
.
xn

The system can be written in matrix form:

AX = B

Remark 5.1. The elements of row i of the augmented matrix Ā are the coefficients of
equation i of system AX = B, and vice versa. Thus, fixing variable names, there is a
one-to-one correspondence between augmented matrices and linear systems.

5.2 Cramer Systems

Definition 5.3 (Cramer System). The system AX = B is called a Cramer system if
A is square and det(A) ̸= 0.

Theorem 5.1 (Cramer’s Rule). A Cramer system AX = B has a unique solution

(x1 , x2 , . . . , xn ) given by:

det(Aj )
xj = , j = 1, 2, . . . , n
det(A)

where Aj is the matrix obtained from A by replacing column j with the constant column
B.

20
MI1141 - Linear Algebra Chapter 2

Example 5.2. Show that the system


2x1 + 5x2 + x3 = 15

3x1 − x2 − x3 = −2

−x1 + 4x2 + 2x3 = 13


is a Cramer system and solve it.

Solution: det(A) = 57 ̸= 0, so it’s a Cramer system.
det(A1 ) 57 det(A2 ) 114 det(A3 ) 171
x1 = = = 1, x2 = = = 2, x3 = = =3
det(A) 57 det(A) 57 det(A) 57

5.3 Conditions for Solutions and Solution Methods

5.3.1 Elementary Operations on Systems
1. Interchange two equations

2. Multiply both sides of an equation by a nonzero scalar

3. Add a multiple of one equation to another equation

These correspond to elementary row operations on the augmented matrix.

Theorem 5.2 (Kronecker-Capelli Theorem). A system of linear equations has a solution
if and only if the rank of the coefficient matrix equals the rank of the augmented matrix:

rank(A) = rank(Ā)

Corollary 5.2.1. For a system of n unknowns AX = B:

1. rank(A) ̸= rank(Ā) ⇔ no solution (inconsistent)

2. rank(A) = rank(Ā) = n ⇔ unique solution

3. rank(A) = rank(Ā) = r < n ⇔ infinitely many solutions depending on (n − r)

parameters

5.3.2 Gaussian Elimination

Steps to solve AX = B:
Step 1: Form augmented matrix Ā = [A|B].
Step 2: Use elementary row operations to transform:
row operations
Ā = [A|B] −−−−−−−−→ Ā′ = [A′ |B ′ ]

Determine rank(A) = rank(A′ ) and rank(Ā) = rank(Ā′ ).

If rank(A) ̸= rank(Ā), the system has no solution.
If rank(A) = rank(Ā), continue to Step 3.
Step 3: Write the equivalent system A′ X = B ′ and solve it.
If rank(A) = rank(Ā) = r < n, the system has infinitely many solutions depending
on (n − r) parameters. Keep r unknowns corresponding to the first r nonzero elements
of the r rows of A′ , and treat the remaining (n − r) unknowns as parameters.

21
MI1141 - Linear Algebra Chapter 2

Example 5.3. Solve the system:


x1 + 2x2 + 3x3 − x4 = 3

2x1 + 3x2 − x3 = −1

3x1 + 5x2 + 2x3 − x4 = 4


Solution:
   
1 2 3 −1 3 1 2 3 −1 3
operations
 2 3 −1 0 −1  −−−−−→  0 −1 −7 2 −7 
3 5 2 −1 4 0 0 0 0 2
Since rank(A) = 2 ̸= 3 = rank(Ā), the system has no solution.

5.3.3 Gauss-Jordan Method

When the system is a Cramer system (rank(A) = rank(Ā) = n), we can continue elemen-
tary operations to transform the augmented matrix to:
 
1 0 · · · 0 t1
 0 1 · · · 0 t2 
 
 .. .. . . .. .. 
 . . . . . 
0 0 · · · 1 tn
The unique solution is (x1 , x2 , . . . , xn ) = (t1 , t2 , . . . , tn ).

5.4 Homogeneous Systems

A homogeneous system of n unknowns has the form:


 a11 x1 + a12 x2 + · · · + a1n xn = 0

a21 x1 + a22 x2 + · · · + a2n xn = 0

..


 .

a x + a x + · · · + a x = 0
m1 1 m2 2 mn n

• Always has the trivial solution (0, 0, . . . , 0)

• Always satisfies rank(A) = rank(Ā)
Proposition 5.1. For a homogeneous system:
1. Has nontrivial solutions ⇔ rank(A) < n
2. Has only trivial solution ⇔ rank(A) = n
Example 5.4. Find values of parameter m for which the system has nontrivial solutions:

(m − 1)x1 − x2 + 2x3 = 0

(2m + 1)x1 + mx2 + x3 = 0

−mx2 + (m + 1)x3 = 0


Solution: The system has nontrivial solutions ⇔ det(A) = 0.

Computing det(A) and setting it to zero gives the values of m.

22