Scribd
Upload
962 views2 pages

Row Echelon Form

This document discusses transforming a matrix into row echelon form through elementary row operations. It provides an example of transforming the matrix A into row echelon form G through a series of row operations, including row interchanges, row multiplications, and adding multiples of rows. It then generalizes the process, describing how to find the pivot column and entry and use row operations to transform any matrix into row echelon form. The document also discusses reduced row echelon form and how it uniquely determines a matrix through elementary row operations.

Uploaded by

mkpradyumna
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
962 views2 pages

Row Echelon Form

This document discusses transforming a matrix into row echelon form through elementary row operations. It provides an example of transforming the matrix A into row echelon form G through a series of row operations, including row interchanges, row multiplications, and adding multiples of rows. It then generalizes the process, describing how to find the pivot column and entry and use row operations to transform any matrix into row echelon form. The document also discusses reduced row echelon form and how it uniquely determines a matrix through elementary row operations.

Uploaded by

mkpradyumna
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Transformation of a matrix to a row echelon form

Example. Using elementary row transformations, produce a row echelon form A

of
the matrix
A =

0 2 8 7
2 2 4 0
3 4 2 5

.
We know that the rst nonzero column of A

must be of view

1
0
0

. We cant
achieve this from matrix A unless interchange the rst row with a row having a
nonzero number in the rst place. So lets transpose rst and second rows:
B = A
r
1
r
2
=

2 2 4 0
0 2 8 7
3 4 2 5

.
Multiply the rst row by
1
2
so that the rst entry becomes 1:
C = B
2r
1
r
2
=

1 1 2 0
0 2 8 7
3 4 2 5

.
Now add 3 times row 1 of C to its third row to get a matrix with the rst column
of desired form:
D = C
3r
1
+r
3
r
3
=

1 1 2 0
0 2 8 7
0 1 4 5

.
Now we can forget about the rst row and concentrate on submatrix of D consist-
ing of the other rows. Again, we need to transform this submatrix to that with the
rst nonzero column

1
0

. To this end, we procede in a similar way. Now we dont


need to interchange second and third rows: just multiply the second one by
1
2
and
then subtract it from row 3, and get the desired view of the second column. But in
order to avoid fractions, lets swap rows 2 and 3 and then replace row 3 by its sum
with a multiple of row 2.
E = D
r
2
r
3
=

1 1 2 0
0 1 4 5
0 2 8 7

and
1
F = E
2r
2
+r
3
r
3
=

1 1 2 0
0 1 4 5
0 0 0 3

.
It is left to multiply the last row by
1
3
:
G = F
3r
3
r
3
=

1 1 2 0
0 1 4 5
0 0 0 1

.
Thus we obtained a matrix A

= G in a row echelon form.


This example suggests a general way to produce a row echelon form of an arbitrary
matrix A = [a
ij
]. We nd the rst nonzero column (pivot column of A) and the rst
nonzero entry in it (it is called pivot; in the previous example, pivot of A is a
21
= 2).
Let the pivot be a
ij
for some i, j. Interchange rows 1 and i getting matrix B = [b
ij
]
with the pivot b
1j
. Multiply the rst row of B by
1
b
1j
and obtain a matrix C with the
pivot c
1j
= 1. Then add multiples of row 1 of C to the rest of rows, getting matrix
D with the pivot column of view

1
0
.
.
.
0

. Now focus on the submatrix of D consisting


of all rows except the rst. Proceding in the same way with this submatrix we will
nally arrive at a matrix A

having row echelon form.


In general, matrix A

is not uniquely determined by A. In the previous example, if


we add a multiple of the last row of A

= G to its second row, we get another echelon


form of A. If we add 5 times row 3 of G to row 2, we will have
H = G
r
2
+5r
3
r
2
=

1 1 2 0
0 1 4 0
0 0 0 1

.
Now lets add the second row to the rst one:
I = H
r
1
+r
2
r
1
=

1 0 6 0
0 1 4 0
0 0 0 1

.
We have matrix I in a reduced echelon form, i. e. an echelon form plus the
condition that all entries above leading ones are 0. Like above, any matrix can be
transformed to that in a reduced echelon form. Unlike echelon form, reduced echelon
form is unique for any matrix. In other words, if matrices A

and A

are obtained
from A by a sequences of elementary row transformations, and both A

, A

are in a
reduced echelon form, then A

= A

. This is a theorem which needs a proof, but we


wont give it now.
2

You might also like