1

RANK OF A MATRIX
Let A be any m n matrix. Then A consists of n column vectors
a₁, a₂ ,....,a, which are m-vectors.

DEFINTION:
The rank of A is the maximal number of linearly
independent column vectors in A, i.e. the maximal number
of linearly independent vectors among {a₁, a₂,......, a}.
If A = 0, then the rank of A is 0.
We write rk(A) for the rank of A. Note that we may compute
the rank of any matrix-square or not
 2

RANK OF 2×2MATRIX
Let us see how to compute 2  2 matrix:
EXAMPLE :
a b 
The rank of a 2  2 matrix A =   is given by
c d 
rk ( A)  2 if det  A   ad  bc  0, since both column vectors are
independent in this case.
0 0
rk(A) = 1 if det(A) = 0 but A  0 =   ,since both column vectors
0 0
are not linearly independent, but there is a single column vector that is
linearly independent (i.e. non-zero).

rk(A) = 0 if A = 0
 3

COMPUTING RANK BY VARIOUS

METHODS
1. BY GAUSS ELIMINATION

2. BY DETERMINANTS

3. BY MINORS

4. BY NORMAL FORM
 4

1. USING GAUSS ELIMINATION

GAUSS ELIMINATION:
Use elementary row operations to reduce A to echelon form. The rank of A is the
number of pivots or leading coefficients in the echelon form. In fact, the pivot
columns (i.e. the columns with pivots in them) are linearly independent.

Note that it is not necessary to and the reduced echelon form –any
echelon form will do since only the pivots matter.

POSSIBLE RANKS:
Counting possible number of pivots, we see that rk(A)  m and rk(A)  n
for any m n matrix A.
 5

EXAMPLE

* Find the rank of a matrix

1 0 2 1
 
A=  0 2 4 2 
 
0 2 2 1
 6

SOLUTION
We use elementary row operations:

1 0 2 1 1 0 2 1 
A=  0 2 4 2    0 2 4 2 
   
0 2 2 1 0 0 2 1 

Since the echelon form has pivots in the first three columns,
A has rank, rk(A) = 3. The first three columns of A are linearly independent.
 7

2. USING DETERMINANTS
Definition:
Let A be an m n matrix. A minor of A of order k is a determinant of a k X k
sub-matrix of A.
We obtain the minors of order k from A by first deleting m rows and n -k
columns, and then computing the determinant. There are usually many
minors of A of a give order.

Find the minors of order 3 of the matrix

1 0 2 1
A =  0 2 4 2 
0 2 2 1
 
 8

COMPUTING MINORS: EXAMPLE

We obtain the determinants of order 3 by keeping all the rows and

deleting one column from A. So there are four different minors of order
3. We compute one of them to illustrate:
1 0 2
A = 0 2 4  1(  4)+ 0 =  4
0 2 2
The minors of order 3 are called the maximal minors of A, since there
are no 4 x 4 sub-matrices of A. There are 3.6 = 18 minors of order 2 and
3.4 = 12 minors of order 1.
 9

3. USING MINORS
Proposition:
Let A be an m X n matrix. The rank of A is the maximal order of a non-zero minor of A.

Idea of proof:
If a minor of order k is non-zero, then the corresponding columns of A are linearly
independent.

Computing the rank:

Start with the minors of maximal order k. If there is one that is non-zero, then rk(A) = k. If
all maximal minors are zero, then rk(A) < k, and we continue with the minors of order k-1
and so on, until we and a minor that is non-zero. If all minors of order 1 (i.e. all entries in
A) are zero, then rk(A) = 0.
 10

RANK:EXAMPLES USING MINOR

Find the rank of the matrix

1 0 2 1
 
A=  0 2 4 2 
0 2 2 1
 

Solution:
The maximal minors have order 3, and we found that the one obtained by
deleting the last column is  4  0. Hence rk(A) = 3.
 11

Solution:

The maximal minors have order 3, so we compute the 4 minors of order 3.

The first one is

1 2 1

9 5 2 = 7(-1) + (-1)(-7) =0

7 1 0

The other three are also zero. Since all minors of order 3 are zero, the rank must be
rk(A)<3. we continue to look at the minors or order two

1 2 = 5 – 18 = -13 ≠ 0

9 5

It is not necessary to compute any more minors, and we conclude that rk(A) =2. in fact, the
first two columns of A are linearly independent.
 12

4. USING NORMAL FORM

NORMAL:

A complex square matrix A is normal if

A*A=AA*

where A* is the conjugate transpose of A. That is, a matrix is

normal if it commutes with its conjugate transpose.

A matrix A with real entries satisfies A*=AT, and is therefore normal

if ATA = AAT.
 13

Find the rank of a matrix using normal form,

2 3 4 5 
 
3 4 5 6
A=  
4 5 6 7 
 
Solution:  9 10 11 12 
Reduce the matrix to echelon form,
2 3 4 5 1 0 0 0
   
3 4 5 6
    0 1 2 3
4 5 6 7 0 0 0 0
   
9 10 11 12  0 0 0 0
 14

ECHELON FORMS
A matrix with the following properties is in reduced row-echelon form, (RREF).

1. If a row does not consist entirely of zeros, then the first nonzero number in the row,
called its pivot, equals 1.

2. If there are any rows that consist entirely of zeros, then they are grouped together
at the bottom of the matrix.

3. In any two successive rows that do not consist entirely of zeros, the pivot in the
lower row occurs farther to the right than the pivot in the higher row.

4. Each column that contains a pivot has zeros everywhere else.

• A matrix that has the first three properties is said to be in row-echelon form.
• A matrix in reduced row-echelon form is of necessity in row-echelon form,
but not conversely.
 15

ROW-ECHELON & REDUCED ROW-

ECHELON FORM

• Reduced row-echelon form:

0 1 2 0 1
1 0 0 4  1 0 0 
0 1 0 7  ,  0 1 0  , 0 0 0 1 3 0 0
0 0 0
    0 ,
0 0 0
0 0 1  1 0 0 1  
0 0 0 0 0

• Row-echelon form:
1 4  3 7 1 1 0 0 1 2 6 0
0 1 6 2, 0 1 0, 0 0 1  1 0
   
0 0 1 5 0 0 0 0 0 0 0 1
 16

MORE ON ROW-ECHELON AND

REDUCED ROW-ECHELON FORM
• All matrices of the following types are in row-echelon form ( any real
numbers substituted for the *’s. ) :
0 1 * * * * * * * *
1 * 1 * 1 * 
*
* * * * * *
0 1 * * 0 1 * * 0 1 * * 
0 0 0 1 * * * * *
 , , , 0 0 0 0 1 * * * * *
0 0 1 * 0 0 1 *  0 0 0 0  
      0 0 0 0 0 1 * * * *
0 0 0 1 0 0 0 0  0 0 0 0 
0 0 0 0 0 0 0 0 1 *
• All matrices of the following types are in reduced row-echelon form ( any
real numbers substituted for the *’s. ) :
0 1 * 0 0 0 * * 0 *
1 0 1 * 1 * 
*
0 0 0 0 0 *
0 1 0 0 0 1 0 * 0 1 * * 
0 0 0 1 0 0 * * 0
 , , , 0 0 0 0 1 0 * * 0 *
0 0 1 0  0 0 1 *  0 0 0 
0  
      0 0 0 0 0 1 * * 0 *
0 0 0 1  0 0 0 0  0 0 0 0 
0 0 0 0 0 0 0 0 1 *
 17

ELIMINATION METHODS (1/7)

We shall give a step-by-step elimination procedure that
can be used to reduce any matrix to reduced row-echelon
form.
0 0  2 0 7 12 
2 4  10 6 12 28 
 
2 4  5 6  5  1
 18

ELIMINATION METHODS (2/7)

• Step1. Locate the leftmost column that does not consist entirely
of zeros.

0 0  2 0 7 12 
2 4  10 6 12 28 
 
2 4  5 6  5  1
Leftmost nonzero column
• Step2. Interchange the top row with another row, to bring a
nonzero entry to top of the column found in Step1.

2 4  10 6 12 28 
0 0  2 0 7 12  The 1st and 2nd rows in the
  preceding matrix were
2 4  5 6  5  1 interchanged.
 19

ELIMINATION METHODS (3/7)

• Step3. If the entry that is now at the top of the column found in Step1 is a,
multiply the first row by 1/a in order to introduce a pivot 1.

1 2  5 3 6 14 
0 0  2 0 7 12 
  The 1st row of the preceding
2 4  5 6  5  1 matrix was multiplied by 1/2.

• Step4. Add suitable multiples of the top row to the rows below so that all
entries below the pivot 1 become zeros.
1 2  5 3 6 14 
0 0  2 0 7  -2 times the 1st row of the
 12  preceding matrix was added to
0 0 5 0  17  29 the 3rd row.
 20

ELIMINATION METHODS (4/7)

Step 5. Now cover the top row in the matrix and begin again with Step1 applied to
the sub-matrix that remains. Continue in this way until the entire matrix is in row-
echelon form.

1 2  5 3 6 14 
0 0  2 0 7 12 
 
0 0  5 0  17  29 Leftmost nonzero
column in the submatrix

1 2  5 3 6 14 
0 0 1 0  7  6  The 1st row in the sub-matrix
 2  was multiplied by -1/2 to
0 0 5 0  17  29 introduce a pivot 1.
 21

ELIMINATION METHODS (5/7)

• Step5 (cont.)
-5 times the 1st row of the sub-
1 2  5 3 6 14  matrix was added to the 2nd row
0 0 1 0  7  6 of the sub-matrix to introduce a
 2  zero below the pivot 1.
0 0 0 0 21
1 
1 2  5 3 6 14  The top row in the sub-matrix was
0 0 1 0  7  6 covered, and we returned again
 2  Step1.
0 0 0 0 12 1 
Leftmost nonzero column in the
new sub-matrix
1 2  5 3 6 14 
0 0 1 0  7  6 The first (and only) row in the
 2  new sub-matrix was multiplied by
0 0 0 0 1 2  2 to introduce a pivot 1.

 The entire matrix is now in row-echelon form.

 22

ELIMINATION METHODS (6/7)

Step 6. Beginning with last nonzero row and working upward, add suitable
multiples of each row to the rows above to introduce zeros above the pivot 1’s.

1 2  5 3 6 14 7/2 times the 3rd row of the

0 0 1 0 0 1  preceding matrix was added to
  the 2nd row.
0 0 0 0 1 2 
1 2  5 3 0 2
0 0 1 0 0 1 -6 times the 3rd row was added
  to the 1st row.
0 0 0 0 1 2
1 2 0 3 0 7
0 0 1 0 0 1 5 times the 2nd row was added
 
0 0 0 0 1 2 to the 1st row.

The last matrix is in reduced row-echelon form.

 23

ELIMINATION METHODS (7/7)

• Step1~Step5: the above procedure produces a row-echelon form and
is called Gaussian elimination.

• Step1~Step6: the above procedure produces a reduced row-echelon

form and is called Gaussian-Jordan elimination.

• Every matrix has a unique reduced row-echelon form but a row-

echelon form of a given matrix is not unique.
 24

LEONTIEF INPUT-OUTPUT ANALYSIS

A way to analyze economics of interdependent sectors

Case study: Oil and Transportation

• Transportation requires
• gasoline from the oil industry
• transportation of equipment from the transportation industry

• Oil production requires

• transportation of gasoline from the transportation industry
• oil-based fuels for processing from the oil industry

• We will look at a single oil company and a single transportation company as a closed
system
 25

SOME BASIC CALCULATIONS

Oil Industry Transportation industry
Cost of producing $1 worth of gas: Cost of producing $1 worth of
– $.32 in oil costs transportation:
– $.12 in transportation costs – $.50 in gas costs
– $.20 in transportation costs
• Suppose that the demand from the outside sector of the economy (all
consumers outside of oil and transportation) is:
– $15 billion for oil
– $1.2 billion for transportation

• If the companies produce exactly this amount, then the amount used in
production is:
– .32(15) + .50(1.2) = $5.4 billion in oil
– .12(15) + .20(1.2) = $2.0 billion in transportation

• This leaves only $9.6 billion of oil and $1.0 billion of transportation to
meet the demand.
 26
LEONTIEF ANALYSIS ALLOWS US TO CALCULATE
HOW MUCH EACH COMPANY SHOULD PRODUCE
TO MEET A GIVEN DEMAND

Let x = the total output from oil company

Let y = the total output from transportation company

Internal External
From oil From
company transportation
company
Oil demand .32x .50y d1 = $15 b

Transportation .12x .20y d2 = $12 b

demand

The internal demand for each is the combined demand from the oil industry
and from the transportation industry
 27

SETTING UP THE DEMAND

EQUATIONS
The total output of each company will equal the sum of the
internal and external demands:

 x  .32 x  .50 y  d1

 y  .12 x  .20 y  d2
 x  .32 .50  x   d1 
 y   .12 .20  y    d 
       2
Expressed as a matrix equation:
X  MX  D
 28

SOLVING THE DEMAND EQUATIONS

Solve for X: X  MX  D
X  I  M  D
1

 x  .32 .50  x   d1 
 y   .12 .20  y    d 
       2

In our example:

.32 .50   .68 .50  1.32 .83 

I  M 
1
M   I M     
.12 .20   .12 .80   .20 1.13 
 29

THE SOLUTION

Putting it all together:

 d   15 
D   1   
 d 2  1.2

1.32 .83   15   20.8

X  I  M 
1
D  1.2    4.4 
 .20 1.13   

In order to meet the demand the companies need to produce

• $20.8 billion of oil

• $4.4 billion of transportation