Lecture 3: Solution Types and Matrix Operations

Shenghao Yang
Spring 2023

1 Solution Types of System of Linear Equations

Example 1. Find the solutions to the following system of equations by Gauss-Jordan elimination
 
 2x1 + x2 + 7x3 − 7x4 = 2
 x1 + x2 + 4x3 − 5x4 = 2

row equ
−3x1 + 4x2 − 5x3 − 6x4 = 3 ↔ x2 + x3 − 3x4 = 2

 

x1 + x2 + 4x3 − 5x4 = 2 0 = −5

   
2 1 7 −7 2 1 1 4 −5 2
row ops
 −3 4 −5 −6 3  −−−−−→  0 1 1 −3 2 
1 1 4 −5 2 0 0 0 0 −5

Form the augmented matrix and perform row operations:

   
2 1 7 −7 2 1 1 4 −5 2
R1 ↔R3
 −3 4 −5 −6 3  −−−−−→  −3 4 −5 −6 3 
1 1 4 −5 2 2 1 7 −7 2
 
R2 →3R1 +R2 1 1 4 −5
2
R →−2R1 +R3
−−3−−−−−− −−→  0 7 7 9 
−21
0 −1 −1 −23
 
1 1 4 −5 2
R ↔R3
−−2−−−→  0 −1 −1 3 −2 
0 7 7 −21 9
 
1 1 4 −5 2
R →−R3
−−2−−−−→  0 1 1 −3 2 
0 7 7 −21 9
 
1 1 4 −5 2
R →−7R2 +R3
−−3−−−−−− −−→  0 1 1 −3 2  .
0 0 0 0 −5

The equivalent system is

x1 + x2 + 4x3 − 5x4 = 2,
x2 + x3 − 3x4 = 2,
0 = −5.

As the last equality never holds, the system has no solution.

1
Example 2. Find the solutions to the following system of equations using Gauss-Jordan elimination

−7x1 − 6x2 − 12x3 = −33 x1 = −3

row equ
5x1 + 5x2 + 7x3 = 24 ↔ x2 = 5
x1 + x2 + 4x3 = 10 x3 = 2

     
−7 −6 −12 −33 1 1 4 10 1 0 0 −3
row ops   row ops  
 5 5 7 24  −−−−−→  0 1 16 37 −−−−−→  0 1 0 5 
1 1 4 10 0 0 -13 −26 0 0 1 2

Apply the elementary row operations to the augmented matrix:

   
−7 −6 −12 −33 1 1 4 10
R1 ↔R3
 5 5 7 24  −−−−−→  5 5 7 24 
1 1 4 10 −7 −6 −12 −33
 
R2 →−5R1 +R2 1 1 4 10
R →7R +R3
−−−3−−−−1−−− →  0 0 −13 −26 
0 1 16 37
 
1 1 4 10
R2 →R3
−−−−−→  0 1 16 37 
0 0 −13 −26
 
1 1 4 10
R3 →− 13
1
R3  
−−−−−− −−→  0 1 16 37 
0 0 1 2
 
R1 →−4R3 +R1 1 1 0 2
R →−16R3 +R2  
−−2−−−−−− −−→  0 1 0 5 
0 0 1 2
 
1 0 0 −3
R →−R2 +R1  
−−1−−−−− −−→  0 1 0 5 
0 0 1 2

The equivalent system is

x1 = −3
x2 = 5
x3 = 2
which is the unique solution.
Example 3. Find the solution set of the following system of equations using Gauss-Jordan elimination
 
x1 − x2 + 2x3 = 1
 
 x1 + x3 = 3
row equ
2x1 + x2 + x3 = 8 ↔ x2 − x3 = 2

 

x1 + x2 = 5 0=0

     
1 −1 2 1 1 −1 2 1 1 0 1 3
row ops row ops
 2 1 1 8  −−−−−→  0 1 −1 2  −
− −− −
→  0 1 −1 2 
1 1 0 5 0 0 0 0 0 0 0 0

2
 Apply the elementary row operations to the augmented matrix:
   
1 −1 2 1 R2 →−2R1 +R2 1 −1 2
1
R3 →−R1 +R3
 2 1 1 8  −−−−−−−−−−→  0 3 6 
−3
1 1 0 5 0 2 −2
4
 
R2 → 1 R2
1 −1 2 1
−−−−−
3
−→  0 1 −1 2 
0 2 −2 4
 
1 −1 2 1
R →−2R2 +R3
−−3−−−−−−−−→  0 1 −1 2 
0 0 0 0
 
1 0 1 3
R →R2 +R1
−−1−−−−−−→  0 1 −1 2  .
0 0 0 0
The equivalent system is
x1 + x3 = 3,
x2 − x3 = 2,
or
x1 = 3 − x3 ,
x2 = 2 + x3 .
Here x3 is called a free variable. x1 and x2 are called lead (dependent) variables. For each given value
of x3 , the free variable, the system becomes one of two variables with a unique solution. For any real
number α, the 3-tuple
(3 − α, 3 + α, α)
is a solution.

Free Variables and Dependent Variables

• Transform the coefficient matrix of the system to row echelon form.
• Each variable is associated with a column of the coefficient matrix, where its coefficients of all the
equations locate.
• A variable is said to be free if its associated column in the coefficient matrix in row echelon form
has no leading entry.
• A variable is said to be dependent if it is not free.
Check the free variables in the previous examples.

Three Possible Cases of Row Echolon Form

 
1 1 4 10
 
Unique solution:  0 1 16 37 
0 0 1 2
 
1 −1 2 1
Infinitely many solutions:  0 1 −1 2 
0 0 0 0
 
1 1 4 −5 2
No solution:  0 1 1 −3 2 
0 0 0 0 −5

3
Solution Types

• Check the last non-zero row of the augmented matrix in row echelon form.
• If it is of the form (0, . . . , 0, ∗), where ∗ is nonzero, then the system has no solution.
• Otherwise, the system has a solution, and is also called a consistent system.
– If the system has no free variables, then the system has a unique solution.
– If the system has at least one free variable, then the system has infinitely many solutions.

• The free variables of a system depend on the row echelon form of the augmented matrix.
• If we change the order of the variables, the set of free variables can be different.

Example 4. Consider the system

a+b=0
a + b + c = 1.

If we order the variables as a, b, c, then the augmented matrix is

   
1 1 0 0 1 1 0 0
→ .
1 1 1 1 0 0 1 1

So b is the free variable. But if we order the variables as b, a, c, then a is the free variable.

2 Consistent Systems
Consistent System

• A system of linear equations is consistent if it has at least one solution.

1. Form the augmented matrix of the system.
2. Transform the augmented matrix by Gaussian elimination to row echelon form.
3. If the last column has no leading entry, the system is consistent.
• Consider the augmented matrix in row echelon form of a consistent system.
– The number r of nonzero rows is the number of dependent variables.
– n − r is the number of free variables.
• For an m × n consistent system with n − r > 0 free variables, if we assign any value to the free
variables, the system becomes one with only r variables of a unique solution.

Example 5. We know b is a free variable of the following system,

a+b=0
a + b + c = 1,

Given any value of b, the system becomes

a = −b
a + c = 1 − b,

which has a unique solution.

Exercise 1. • The number of dependent variables is no more than for an m × n system.
• A consistent m × n system with m < n has solution(s).

4
 • If an m × n system has no free variables, then and the reduced row echelon form of the
coefficient matrix must be of the form
 
1 0 0 ··· 0
0 1 0 ··· 0
 
0 0 1 ··· 0
 
 .. .. .. .. .. 
. . . . .
 
0 0 0 ··· 1
 
0 0 0 ··· 0
0 0 0 ··· 0

Homogeneous Systems

• A system of linear equations is said to be homogeneous if the constants on the right-hand side are
all zero.
• Homogeneous systems are always consistent as the last column in the augmented matrix has only
zero.

• (0, 0, . . . , 0) is always a solution of a homogeneous system, and is call the trivial solution.

x1 + 2x2 + 2x3 = 0
x1 + 3x2 + 3x3 = 0
2x1 + 6x2 + 5x3 = 0
Exercise 2. • If n > m, an m × n homogeneous system has solution.

• If n ≤ m, is it possible that an m × n homogeneous system has a nontrivial solution?

Example 6. Consider the following system of linear equations:

2α1 x1 + x2 = β1
α2 x1 + x2 = β2 .

Which of the following conditions can guarantee the consistency of the above system?
1. 2α1 ̸= α2

2. β1 = β2
3. 2α1 = α2 and β1 = β2
4. All of the above.

3 Matrix Operations
Matrix Notation

• An m × n matrix A is an array of numbers with m rows and n columns in the form:

 
a11 a12 · · · a1n
 a21 a22 · · · a2n 
 
 .. .. .. ..  ,
 . . . . 
am1 am2 · · · amn

where aij is the entry of the matrix in the ith row and jth column, also known as the (i, j) entry.

• The above matrix can be also written as (aij ).

5
 • Matrices are usually denoted by uppercase letters, e.g., A, B, C, · · · .
• Denote by (A)ij the (i, j) entry of A.
• When m = n, the matrix is a square matrix.
• When all the entries of a matrix are zeros, the matrix is called zero matrix, denoted by 0.
We may use Rm×n to denote the collection of all m × n matrices.
Definition 1. The transpose of an m × n matrix A = (aij ) is the n × m matrix B = (bij ) such that
bji = aij , ∀i = 1, . . . , m, j = 1, . . . , n.
T
The transpose of A is denoted by A .
Example 7.  
  1 4
1 2 3 2
, 5 .
4 5 6
3 6
With the help of the transpose, we only need to focus on the row operations, row equivalence, etc.
The similar column operations/equivalence is just the row operations/equivalence on the transpose.

Vectors
• A column vector is a matrix of one column, which is usually denoted by a boldface lowercase
letter, e.g., x.
• The set of all column vectors of n real numbers is denoted by Rn .
• The set of all column vectors of n complex numbers is denoted by Cn .
• A row vector is a matrix of one row, which is the transpose of a column vector.
Example 8.  
1
2  
 , 1 2 3 4
3
4

Matrix Operations
Definition 2 (Equality). Two m × n matrices A = (aij ) and B = (bij ) are said to be equal (written as
A = B) if aij = bij for each i and j.
Exercise 3. Show that if A = B then AT = B T .
Definition 3 (Matrix Addition). If A = (aij ) and B = (bij ) are two m × n matrices, then the sum of A
and B (written as A + B) is the m × n matrix with the (i, j) entry aij + bij for each ordered pair (i, j),
i.e, A + B = (aij + bij ).
Example 9.      
1 2 3 2 3 4 3 5 7
+ =
4 5 6 −1 −2 −3 3 3 3
Definition 4 (Scalar Multiplication). Let A = (aij )m×n be a matrix and α be a scalar (any real
number). The scalar product of α and A (written as αA) is the m × n matrix with the (i, j) entry αaij ,
i.e., αA = (αaij ).
• We write −A = (−1)A, then A − B is A + (−1)B.
• By definition, 1A = A.
Example 10.    
1 2 3 2 4 6
2 =
4 5 6 8 10 12

6
Properties of Matrices Operations
Consider matrices A, B, C of the same size, and scalars α and β.
1. A + B = B + A.
2. A + (B + C) = (A + B) + C. (We will therefore write A + B + C.)

3. A + 0 = 0 + A = A.
4. α(βA) = (αβ)A.
5. α(A + B) = αA + αB.
6. (α + β)A = αA + βA.

Exercise 4. Simplify the calculation of

α(A + B) + β(B + C) + (α + β)(A + C),

where A, B, C are matrices of the same size and α, β, γ are scalars.

These properties can be proved using the similar properties of real number operations.
Proof of α(A + B) = αA + αB. Fix a row number i and a column number j.

(α(A + B))ij =α(A + B)ij (definition of scalar multiplication)

=α(aij + bij ) (definition of matrix addition).

Similarly,

(αA + αB)ij =(αA)ij + (αB)ij (definition of matrix addition)

=αaij + αbij (definition of scalar multiplication).

Now we are just comparing real (or complex) numbers, and you know the rules for comparing these:

α(aij + bij ) = αaij + αbij (distributive property).

We see that the (i, j)th entry of α(A + B) equals the (i, j)th entry of αA + αB for all i and j, which is
exactly the definition of two matrices being equal.
Exercise 5. Show that if we add a matrix A by itself for k times, the resulting matrix is kA, i.e.,

A + A + · · · + A = kA.
| {z }
k times