You are on page 1of 7

Math 415 - Lecture 2

Echelon Forms, General Solution.

Wednesday August 24 2016


Textbook: Chapter 1.3, Chapter 2.2 (just the pages 78 and 79)
Suggested Practice Exercise: in Chapter 1.3, Exercise 17, 23, 24, in Chapter 2.2, Exercise 2 (just
reduce A, B to echelon form), 8
Khan Academy Video: Matrices: Reduced Row Echelon Form 1

Row Reduction and Echelon Forms

Definition. A matrix is of Echelon form (or row echelon form) if


1. All nonzero rows (rows with at least one nonzero element) are above any rows of all zeros.
2. the leading entry (the first nonzero number from the left) of a nonzero row is always strictly to
the right of the leading coefficient of the row above it.
3. All entries in a column below a leading entry are zero.
A leading entry of an echelon form matrix is also called a pivot.
Remark. It can be shown that 1. and 2. imply 3. Can you see why?
Example 1. Are the following matrices in Echelon form? Circle the leading entries in each row!

3
1 2 0 5
0
2 0 1 4
Echelon form ? 1. X 2. X 3. X
(a)
0
0 0 0 0
0
0 0 0 0

0 2 0 1 4
Not echelon form.
3
1 2 0 5
1. X 2. Fails 3. Fails
(b)
0
0 0 0 0
Would be after R1 R2
0
0 0 0 0

-3
2
1

0 6
0 Echelon form.

(c)
0 32
1. X 2. X 3. X
0
0
0
0

(d)

0
0
0
0
0

4
0
0
0
0

2
0
0
0
0

0
3
0
0
0

0
0
2
0
0

1
0
1
0
0

2
0
0
0
0

3
6
5
1
0

1
0
0
0
-3

0
5
9
0
0
0
0 2
0
0

Echelon form.

1. X 2. X 3. X

Leading column of 0s is OK.

Why Echelon Form?


The echelon form formalizes our idea of a triangular matrix. The echelon form of an augmented
matrix is very useful if you want to know if a system is consistent, and if so if there are infinitely
many solutions. We will see why later in this lecture. If you want to find the actual solutions (if any),
we will see that it will be helpful to consider an even nicer form of the augmented matrix.
Definition. A matrix is of the row reduced echelon form (or: reduced echelon form, or: RREF)
if in addition to conditions 1, 2, and 3 above it also satisfies
4. The leading entry in each nonzero row is 1.
5. Each leading 1 is the only nonzero entry in its column.
Example 2.

0
0

(a) 0

0
0

1
0

(b)
0
0

(c) 0
0

Are the following matrices in reduced echelon form? Circle the leading entries!

1 3
0
0
2 5
0
0 6 7
0
1 12
0
0 2 1
0 0 1

0 0
0 1 3 4
0
0 5 0 Reduced row echelon form.
1. X 2. X 3. X 4. X 5. X
0 0
0
0
0 0 1
0 0 1
0 0
0
0
0 0
0 1 1 1

0 5
0 7
2 4
0 6
No:

0 4. Fails 5. X
0 0 -5
0 0
0
0
0
1
0

2 3
2 2
0 0

2
0
1

24

7 No:
4. X 5. Fails
4

Theorem 1 (Uniqueness of The Reduced Echelon Form). Each matrix is row-equivalent to one and
only one reduced echelon matrix.
Definition. We say a matrix B is an (reduced) echelon form of a matrix A if A and B are rowequivalent and B is (reduced) echelon form.
Question: Is the Theorem also true for Echelon from?
Solution.
No:


1
0



2
1

1
0

Both are row-equivalent and in echelon form.


0
1

Example 3. Find an echelon form and the reduced echelon form of the matrix


3 9 12 9 6 15
3 7 8 5 8 9
Solution.
"

3
0

9
2


1

1
R1 3 R1 0

3
1

R2R2R1

9
4

12
4

6 15
2 6

This is an echelon form!


3
2

4
2


2 5
1 3

R2 21 R2


1 0

R1R1+3R2 0 1

2 2 5
2 2 1

4
3

This is reduced row echelon form!


Remark. The method we used above is called Gaussian elimination (or: Gauss-Jordan elimination).

Pivots

Definition. A pivot position is the position of a leading entry in an echelon form of a matrix. A
pivot column is a column that contains a pivot position.
Example 4. In this example, circle the pivot positions and determine the pivot columns.

1
0

0
0

0
2
0
0

5
4
0
0

0
0
5
0

7
6

0
0

Solution.

1
0

0
0

0
2
0
0

5
4
0
0

0
0
5
0

7
6

0
0

Columns 1, 2 and 4 are pivot columns. Note that in this case it was easy to determine the pivot
columns, because the matrix was already in echelon form. In general, whenever you want to determine
the pivot columns of matrix, you have to find an echelon form first.
Example 5. Locate the pivot columns of the following matrix.

0 3 6
4
9
1 2 1
3
1
1
4
5 9 7

Solution.

1
4
5 9 7
3
1
1 2 1
R1R3
0 3 6
4
9

1
4
5 9 7
0
2
4 6 6

R2R2+R1
0 3 6
4
9

1
4 5 9 7
0 2 4 6 6

R3R3+1.5R2
0
0 0 -5
0
So column 1,2 and 4 are pivot columns. Note that we couldnt have seen that from the original
matrix. We had to bring the matrix to echelon form.

Remark. There is no more than one pivot in any row. There is no more than one pivot in any
column.

Free variables

Why do we care about pivots and pivot columns? Recall: each column of a coefficient matrix
corresponds to one of the variables.
Definition. A pivot variable (or basic variable) is a variable that corresponds to a pivot column
in the coefficient matrix of a system.
Definition. A free variable is variable that is not a pivot variable.
Example 6. Consider the following

1 6 0
3 0
0 0 1 8 0
0 0 0
0 1

system of linear equations:

x1 +6x2
0
5
x3
7

+3x4
8x4
x5

=0
=5
=7

What are the pivot columns? 1st, 3rd, and 5th columns.
What are the pivot variables? x1 , x3 , and x5 .
What are the free variables? x2 and x4 .

General Solution of linear systems

Goal. Solve each equation for the pivot variable in terms of the free variables (if any) in the equation.
We call such a solution general solution of a linear system.
Example 7. Find the general solution of
3x1
3x1

7x2
9x2

5x4
9x4

+8x3
+12x3

+8x5
+6x5

=9
= 15

Solution. Step 1: Write down the augmented matrix:



3 7 8 5 8
3 9 12 9 6

9
15

Step 2: Determine the reduced echelon form. We already did this earlier:


1 0 2 2 5 4
0 1 2 2 1 3
Step 3: Write down the equation form of the RREF matrix:
x1
x2

2x3
2x3

+2x4
+2x4

+5x5
+ x5

= 4
= 3

Step 4: Write down pivot and free variables and the general solution:
Pivot variables: x1 , x2
Free variables: x3 , x4 , x5

x1

x2
x3
General solution:

x5

= 2x3 2x4 5x5 4


= 2x3 2x4 x5 3
= free
= free
= free

Remark. The general solution of the system provides a parametric description of the solution
set.
The free variables act as parameters.
The above system has infinitely many solutions. Why?
Because you can pick any value of x3 , x4 and x5 .
Summing up: the algorithm to find the general solution of linear system is as follows.
Algorithm 1. Given a linear system,
(1) Write down the augmented matrix.
(2) Using elementary row operations, find the reduced echelon form of the matrix.
(3) Write down the equations corresponding to the reduced echelon form.
(4) Write down pivot and free variables and the general solution.
Warning. Use only the reduced echelon form to solve a system. If you use an echelon form that is
not in reduced echelon form, you might get unwanted dependencies among variables.

Existence And Uniqueness

We use the reduced echelon form to find the complete solution of a linear system. The question
whether a system has solution and whether it is unique, is much easier to answer than to find the
complete solution. For that an echelon form is sufficient.
Example 8. Consider the following system:
3x2

6x3

+6x4

+4x5

= 5

3x1

7x2

+8x3

5x4

+8x5

=9

3x1

9x2

+12x3

9x4

+6x5

= 15

Is the system Consistent? Is the solution Unique? Are there free variables?
Solution. To answer these question we need just an echelon form. The augmented matrix is

0
3 6
6 4 5
3 7
8 5 8
9
3 9 12 9 6 15
and suppose that an echelon form of this matrix is

3 9 12 9 6
0
2 4
4 2
0
0
0
0 1

15
6 .
4

(Dont believe me? Do the calculation yourself!)


1. What are the free variables? x3 , x4 .
2. Is the system consistent? Yes /No Why?
We can set all free variables = 0. Thenwe can solve the third
 equation for x5 , the second for x2 and
the first for x1 . Only a row of the form 0 0 0 0 0 b , where b 6= 0, would be a problem!
3. How many solutions?
Infinitely many solutions - even if we dont see the free variable, we still get a solution.

Theorem 2 (Existence and Uniqueness Theorem). A linear system is consistent if and only if an
echelon form of the augmented matrix has no row of the form


0 ... 0 b ,
where b is nonzero.If a linear system is consistent, then the solution contains either
a unique solution (when there are no free variables) or
infinitely many solutions (when there is at least one free variable).
Warning. You have to use an echelon form of the augmented matrix. If you just use the augmented
matrix, you will get wrong results.
6

Example 9. The (reduced) echelon form of

3
4 3
2
5
5
2 3
1

1
0
0

is

0
1
0

5
3
0

Is the system consistent? How many pivots? How many free variables? How many solutions?
Solution.


The system is consistent, because there is no row of the form 0 0 b , where b is nonzero. There
are 2 pivots and hence 0 free variables. Therefore there is exactly one solution.
Example 10. An echelon form of

3 4
3 4
6 8

3
3
6

is

3 4
0 0
0 0

3
0
0

What can you say about the number of solutions?


Solution.


The system is consistent, because there is no row of the form 0 0 b , where b is nonzero. There
is 1 pivots and hence there is 1 free variables. Therefore there are infinitely many solution.