MATH 304

Linear Algebra
Lecture 18:
Rank and nullity of a matrix.
Basis and coordinates.
Change of coordinates.
Rank of a matrix
Definition. The row space of an m×n matrix A is
the subspace of Rn spanned by rows of A. The
column space of A is a subspace of Rm spanned by
columns of A.
The row space and the column space of A have the
same dimension, which is called the rank of A.
Theorem 1 Elementary row operations do not change the
row space of a matrix.
Theorem 2 If a matrix A is in row echelon form, then the
nonzero rows of A form a basis for the row space.
Theorem 3 The rank of a matrix is equal to the number of
nonzero rows in its row echelon form.
Nullspace of a matrix
Let A = (aij ) be an m×n matrix.
Definition. The nullspace of the matrix A,
denoted N(A), is the set of all n-dimensional
column vectors x such that Ax = 0.
 

a11 a12 a13 . . . a1n
 x 1 0
 
 a21 a22 a23 . . . a2n   x2   0 
 
 . .. .. .. ..   x3  =  . 
 ..   .. 
. . . .  ... 
am1 am2 am3 . . . amn xn 0

The nullspace N(A) is the solution set of a system

of linear homogeneous equations (with A as the
coefficient matrix).
Let A be an m×n matrix. Then the nullspace
N(A) is the solution set of a system of linear
homogeneous equations in n variables.
Theorem The nullspace N(A) is a subspace of the
vector space Rn .

Definition. The dimension of the nullspace N(A) is

called the nullity of the matrix A.
Problem. Find the nullity of the matrix
 
1 1 1 1
A= .
2 3 4 5
Elementary row operations do not change the nullspace.
Let us convert A to reduced row echelon form:
     
1 1 1 1 1 1 1 1 1 0 −1 −2
→ →
2 3 4 5 0 1 2 3 0 1 2 3
 
x1 − x3 − 2x4 = 0 x1 = x3 + 2x4
⇐⇒
x2 + 2x3 + 3x4 = 0 x2 = −2x3 − 3x4
General element of N(A):
(x1 , x2 , x3 , x4 ) = (t + 2s, −2t − 3s, t, s)
= t(1, −2, 1, 0) + s(2, −3, 0, 1), t, s ∈ R.
Vectors (1, −2, 1, 0) and (2, −3, 0, 1) form a basis for N(A).
Thus the nullity of the matrix A is 2.
rank + nullity

Theorem The rank of a matrix A plus the nullity

of A equals the number of columns in A.
Sketch of the proof: The rank of A equals the number of
nonzero rows in the row echelon form, which equals the
number of leading entries.
The nullity of A equals the number of free variables in the
corresponding system, which equals the number of columns
without leading entries in the row echelon form.
Consequently, rank+nullity is the number of all columns in the
matrix A.
Problem. Find the nullity of the matrix
 
1 1 1 1
A= .
2 3 4 5

Alternative solution: Clearly, the rows of A are

linearly independent. Therefore the rank of A is 2.
Since
(rank of A) + (nullity of A) = 4,
it follows that the nullity of A is 2.
Basis and dimension

Definition. Let V be a vector space. A linearly

independent spanning set for V is called a basis.

Theorem Any vector space V has a basis. If V

has a finite basis, then all bases for V are finite and
have the same number of elements (called the
dimension of V ).

Example. Vectors e1 = (1, 0, 0, . . . , 0, 0),

e2 = (0, 1, 0, . . . , 0, 0),. . . , en = (0, 0, 0, . . . , 0, 1)
form a basis for Rn (called standard) since
(x1 , x2 , . . . , xn ) = x1 e1 + x2 e2 + · · · + xn en .
Basis and coordinates
If {v1 , v2 , . . . , vn } is a basis for a vector space V ,
then any vector v ∈ V has a unique representation
v = x1 v1 + x2 v2 + · · · + xn vn ,
where xi ∈ R. The coefficients x1 , x2 , . . . , xn are
called the coordinates of v with respect to the
ordered basis v1 , v2 , . . . , vn .

The mapping
vector v 7→ its coordinates (x1 , x2 , . . . , xn )
is a one-to-one correspondence between V and Rn .
This correspondence respects linear operations in V
and in Rn .
Examples. • Coordinates of a vector
v = (x1 , x2 , . . . , xn ) ∈ Rn relative to the standard
basis e1 = (1, 0, . . . , 0, 0), e2 = (0, 1, . . . , 0, 0),. . . ,
en = (0, 0, . . . , 0, 1) are (x1 , x2 , . . . , xn ).
 
• Coordinates of a matrix ca b
d
∈ M2,2 (R)
     
1 0 0 0 0 1
relative to the basis 0 0
,
1 0
,
0 0
,
 
0 0
0 1
are (a, c, b, d).

• Coordinates of a polynomial
p(x) = a0 + a1 x + · · · + an−1 x n−1 ∈ Pn relative to
the basis 1, x, x 2 , . . . , x n−1 are (a0 , a1 , . . . , an−1 ).
Vectors u1 =(3, 1) and u2 =(2, 1) form a basis for R2 .
Problem 1. Find coordinates of the vector
v = (7, 4) with respect to the basis u1 , u2 .

The desired coordinates x, y satisfy

 
3x + 2y = 7 x = −1
v = xu1 +y u2 ⇐⇒ ⇐⇒
x +y =4 y =5

Problem 2. Find the vector w whose coordinates

with respect to the basis u1 , u2 are (7, 4).
w = 7u1 + 4u2 = 7(3, 1) + 4(2, 1) = (29, 11)
Change of coordinates
Given a vector v ∈ R2 , let (x, y ) be its standard
coordinates, i.e., coordinates with respect to the
standard basis e1 = (1, 0), e2 = (0, 1), and let
(x ′ , y ′ ) be its coordinates with respect to the basis
u1 = (3, 1), u2 = (2, 1).
Problem. Find a relation between (x, y ) and (x ′ , y ′ ).
By definition, v = xe1 + y e2 = x ′ u1 + y ′ u2 .
In standard coordinates,
        ′
x 3 2 3 2 x
=x ′
+y ′
=
y 1 1 1 1 y′
 ′  −1     
x 3 2 x 1 −2 x
=⇒ = =
y ′
1 1 y −1 3 y
Change of coordinates in Rn
The usual (standard) coordinates of a vector
v = (x1 , x2 , . . . , xn ) ∈ Rn are coordinates relative to the
standard basis e1 = (1, 0, . . . , 0, 0), e2 = (0, 1, . . . , 0, 0),. . . ,
en = (0, 0, . . . , 0, 1).
Let u1 , u2 , . . . , un be another basis for Rn and (x1′ , x2′ , . . . , xn′ )
be the coordinates of the same vector v with respect to this
basis.

Problem 1. Given the standard coordinates

(x1 , x2 , . . . , xn ), find the nonstandard coordinates
(x1′ , x2′ , . . . , xn′ ).
Problem 2. Given the nonstandard coordinates
(x1′ , x2′ , . . . , xn′ ), find the standard coordinates
(x1 , x2 , . . . , xn ).
 It turns out that
 ′ 
x1 u11 u12 . . . u1n x1
  
 x2   u21 u22 . . . u2n  x2′ 
 . = . .. . . .  . .
 ..   .. . . ..  .. 
xn un1 un2 . . . unn xn′

The matrix U = (uij ) does not depend on the vector v.

Columns of U are coordinates of vectors
u1 , u2 , . . . , un with respect to the standard basis.
U is called the transition matrix from the basis
u1 , u2 , . . . , un to the standard basis e1 , e2 , . . . , en .
This solves Problem 2. To solve Problem 1, we have
to use the inverse matrix U −1 , which is the
transition matrix from e1 , . . . , en to u1 , . . . , un .
Problem. Find coordinates of the vector
x = (1, 2, 3) with respect to the basis
u1 = (1, 1, 0), u2 = (0, 1, 1), u3 = (1, 1, 1).
The nonstandard coordinates (x ′ , y ′ , z ′ ) of x satisfy
 ′  
x 1
 y ′  = U  2 ,
z′ 3
where U is the transition matrix from the standard basis
e1 , e2 , e3 to the basis u1 , u2 , u3 .
The transition matrix from u1 , u2 , u3 to e1 , e2 , e3 is
 
1 0 1
U0 = (u1 , u2 , u3 ) =  1 1 1 .
0 1 1
The transition matrix from e1 , e2 , e3 to u1 , u2 , u3 is the
inverse matrix: U = U0−1 .
The inverse matrix can be computed using row reduction.
 
1 0 1 1 0 0
(U0 | I ) =  1 1 1 0 1 0 
0 1 1 0 0 1
   
1 0 1 1 0 0 1 0 1 1 0 0
→  0 1 0 −1 1 0  →  0 1 0 −1 1 0
0 1 1 0 0 1 0 0 1 1 −1 1
 
1 0 0 0 1 −1
→  0 1 0 −1 1 0  = (I | U0−1 )
0 0 1 1 −1 1

Thus  ′     
x 0 1 −1 1 −1
 y ′  =  −1 1 0   2 =
  1 .
z ′
1 −1 1 3 2