Scribd
Upload
14 views

Lecture 20

Uploaded by

anutyo
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
14 views

Lecture 20

Uploaded by

anutyo
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

Lecture 20

Ajay Kumar
[email protected]

October 17, 2023


Rank and Nullity
Basis of the column space of A
 
0 0 1 ∗ 0 0 ∗ ∗ 0
0 0 0 0 1 0 ∗ ∗ 0
 
0 0 0 0 0 1 ∗ ∗ 0
 
Observe that if AE =   is a reduced
0 0 0 0 0 0 0 0 1
 
0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0
row-echelon matrix, then the first non-zero column of A is
 t
e1 = 1 0 0 · · · 0 which is a pivotal column, say, at j1 position.
Now if e1 , e2 , . . . , er are first r pivotal columns of A occurring at
j1 , j2 , . . . , jr positions, then the next column of A at (jr + 1)th position is
either the pivotal column er+1 or is a linear combination of the
preceding pivotal columns e1 , e2 , . . . , er . Clearly, pivotal columns forms
a basis of C(AE ). The columns of A corresponding to the pivotal
columns of AE are called the basic columns. The basic columns of A
form a basis for C(A).
Finding basis of a span of a set in Rm

Let S = {v1 , . . . , vn } be a set of vectors in Rm . To find a basis of


 
span(S), consider a m × n matrix A = v1 v2 · · · vn . To find a
basis of span(S), convert the matrix A into the reduced row echelon
form AE . Now, vector corresponding to basic columns of AE forms a
basis for span(S)
Definition
Given an m × n matrix A
• the rank of A is the dimension of the column space of A :
rank(A) = dim C(A).
• the nullity of A is the dimension of null space of A :
nullity(A) = dim N (A).
Example

       
1 2 2 3
Find a basis for span  2 , 4 , 1 , 3 .
     
3 6 1 4
   
1 2 2 3 1 2 0 1
Let A = 2 4 1 3 . Then AE = 0 0 1 1 with first and third
3 6 1 4 0 0 0 0
columns as pivotal columns. Thus a basis of column space of A is
 t  t
{ 1 2 3 , 2 1 1 }. Hence a basis for
           
1 2 2 3  1 2 
span 2 , 4 , 1 , 3 is 2 , 1 . Now, we can see that
 
3 6 1 4 3 1
rank(A) = 2 and nullity(A) = 2.
Basis of the null space of A
To find a basis of the null space of A, solve the homogeneous system
Ax = 0. Let rank A = r. Then we have the following observations:
• Since elementary row operations are invertible, the solution set
of Ax = 0 equals to the solution set of AE x = 0.
• The unknown variables corresponding the positions of basic
columns are called basic variables and other variables are called
free variables.
• There are exactly r basic variables and n − r free variables.
• Let x1 , x2 , . . . , xn−r denote the solutions obtained by sequentially
setting each free variable equal to 1 and other free variables
equal to zero.
• The set {x1 , x2 , . . . , xn−r } forms a basis of null space of A.
• rank(A)+ nullity(A)= number of columns of (A).
Theorem (rank nullity theorem)
Let A be an m × n matrix. Then rank(A) + nullity(A) = n.
Example

   
1 2 2 3 1 2 0 1
Let A = 2 4 1 3 . Then AE = 0 0 1 1 . Thus the original
3 6 1 4 0 0 0 0
homogeneous system is equivalent to the following reduced
homogeneous system: x1 + 2x2 + x4 = 0, x3 + x4 = 0, where x2 and x4
are free variables. By taking x2 = 1 and x4 = 0, we get a solution
 t
x1 = −2 1 0 0 . Similarly, by taking x2 = 0 and x4 = 1, we get a
 t
solution x2 = −1 0 −1 1 . Thus a basis of null space of A is
 t  t
{ −2 1 0 0 , −1 0 −1 1 } and a basis of column space of
 t  t
A is { 1 2 3 , 2 1 1 }.

You might also like