LINEAR ALGEBRA II

LINEAR MAPPINGS

Let U and V be vector spaces over the same field K . A mapping T :U →V is called a
linear mapping or a linear transformation if;
~~ ~ ~ ~ ~
1. For any u , v ∈U , T ( u + v )=T ( u )+T ( v ).
~ ~ ~
2. For any scalar k ∈ K and u ∈ U , T (k u )=kT ( u ).

Example 1: Let T : R → R be defined by T ( x , y )=( x + y , y ). Show that T is a linear

2 2

transformation.
~ ~
Solution: Let u =( x 1 , y 1 ) and v =( x 2 , y 2 ) in R and scalar k . Then
2

T (~
u +~
v )=T [ ( x1 , y 1 )+( x 2 , y 2 ) ]

=T ( x 1 +x 2 , y 1 + y 2 )

=( x 1 +x 2 + y 1 + y 2 , y 1 + y 2 )
And

T (~
u )+T ( ~
v )=T ( x 1 , y 1 )+T ( x 2 , y 2 )

=( x 1 + y 1 , y 1 )+( x 2 + y 2 , y 2 )

=( x 1 + y 1 + x2 + y 2 , y 1 + y 2 )

=( x 1 + x 2 + y1 + y 2 , y 1 + y 2 )=T ( ~
u +~
v ).

T (k ~
u )=T [ k ( x 1 , y 1 ) ]=T (kx 1 , ky 1 )

=(kx 1 + ky 1 , ky 1 )=k ( x 1 + y 1 , y 1 )

=kT (~
u)
Hence T is linear.

Example 2: Let T : R ⟶ R be defined by T ( x , y )=x+ y+1 . Show that T is not linear.

2

1
 ~ ~
Solution: Let u =(1 , 0 ) and v =(0 ,1 ) in R , then
2

T (~
u +~
v )=T [ (1 , 0 )+(0 , 1) ]

=T (1 ,1 )=1+1+1=3

But

T (~
u )+T ( ~
v )=T (1 , 0 )+ T ( 0 , 1 )

=( 1+1+0 )+( 0+1+1 )=2+2=4

~ ~ ~ ~
Thus T ( u + v )≠T ( u )+T ( v ) and hence T is not linear.

Example 3: Let T be a linear mapping T : R ⟶ R defined by T (3 ,1 )=(2 ,−4 ) and

2 2

T (1, 1 )=(0 , 2) . Find T ( x , y ). In particular find T (7 , 4 ).

Solution: Since { (3 ,1 ),(1 , 1) } is a basis for R , T exists. We first find ( x , y ) as a linear

combination of { (3 ,1 ),(1 , 1) } . Thus

3 λ+μ=x
( x , y )=λ (3 ,1 )+ μ ( 1, 1) ⇒ λ+μ= y . Solving these simultaneously
x y 3 x
λ= − , μ= y−
for λ and μ gives 2 2 2 2.

Therefore
x y
( ) 3 x
( x , y )= − (3,1)+ y− (1 , 1)
2 2 2 2 (.
)
Now

2
 T ( x , y )=T [ ( )
x y
2 2
3
2
x
− (3 ,1 )+ y − (1 , 1)]
2 ( )
( )
x y
2 2
3
2
x
= − T (3 ,1 )+ y− T (1 , 1)
2 ( )
( )
x y 3 x
= − (2 ,−4 )+ y− (0 ,2 )
2 2 2 2 ( )
=( x− y ,−2 x+2 y )+(0 , 3 y−x )=( x− y ,−3 x +5 y )

T (7 , 4 )=(7−4 ,−21+20)=(3 ,−1 ).

Exercise:

1. Show that the mapping defined by T ( x , y , z )=2 x+3 y −4 z where T : R ⟶ R is a

3

linear map.
2. Show that the mapping T : R ⟶ R defined by T ( x , y , z )=( x +1 , 2 y , x + y ) is not
2 3

linear.
3. Let T : R ⟶ R be defined by T (1,1,1,1)=(1,0,−1), T (0,1,1,1)=(2,1,1),
4 3

T (0,0 ,1 ,1)=(1,−1 ,3 ) and T (0 , 0 ,0 ,1 )=(−2 ,3 ,2 ). Find T ( x , y , z , w ) and hence determine

T (2,3,4 ,5).

MATRICES AND LINEAR TRANSFORMATIONS (OPERATORS)

Suppose
{e1 ,e2 ,e 3 ,⋯,en }is a basis for a vector space V over
K and for ~
v ∈ V suppose
~v=a e +a e +a e +⋯+a e
that 1 1 2 2 3 3 n n , then the coordinate vector of ~
v relative to the

basis
{ei }is the column vector

~[v] =¿(a1¿)(a2¿)(a3¿)(⋮¿) ¿
e
¿
3
 v=(4 , 3)∈ R and { f 1 =( 1 , 3) , f 2 =( 2 , 5 ) } be a basis of R2. Find the
Example 1: Let ~
2

~
coordinate vector of v relative to the basis
{f i }.

Solution: (4 ,3 )= λ(1 ,3 )+μ(2 ,5 ) so that

λ+2μ=4
3 λ+5 μ=3 solving these simultaneously gives μ=9 and λ=−14 .

[~v ]f =¿ (−14 ¿ ) ¿ ¿¿
Therefore, ( 4 ,3 )=−14 f 1 + 9 f 2 and ¿ .

Example 2: Let V be the vector space of polynomials of degree less or equal to 2,

V = {at 2 +bt+c: a,b,c ∈ R } . The polynomials e 1= 1,
2
e 2= t −1 and e 3 =(t−1 ) =t 2 −2t +1
~
form a basis for V . Let V =2t −5t +6 . Find [ v ] e , the coordinate vector of V relative
2

to the basis
{e1 ,e2 ,e 3}.

Solution: Set

2 t 2 −5 t+6= λ1 (1 )+ λ2 (t−1 )+λ3 (t 2−2 t+1 )

= λ1 (1 )+ λ2 t−λ 2 +λ 3 t 2 −2 λ3 t+λ 3

=λ3 t 2 +( λ2 −2 λ3 )t+ λ1 −λ2 +λ3

Equating the coefficients, we have

λ3=2 ¿} λ2−2λ3=−5 ¿}¿ ¿ ⇒λ1=3, λ2=−1, λ3=2¿

Therefore,
2 t 2 −5 t+6=3 e 1 −e 2 +2 e3 .

4
Exercise: Let V be the vector space of 2×2 matrices over R . Find coordinate vector

of the matrix A ∈V relative to the basis

1 1
,
1 0
,
{( ) (
1 1 0 −1 1 −1 1 0
0 0
,
0 0 )( ) ( )} where

(
A= 2 3
4 −7 ) .

MATRIX REPRESENTATION OF LINEAR OPERATORS

Let T be a linear operator on a vector space V over a field K , T :V →V . Suppose

{e1 ,e2 ,e 3 ,⋯,en }is a basis of V . Now T (e1 ), T( e2 ), T(e 3 ),⋯, T (e n) are vectors in V and

each is a linear combination of

{ei }. Therefore,

T (e1 )=a11 e 1+a 12 e 2 +a 13 e 3 +⋯+a1 n e n

T (e 2 )=a21 e1 +a22 e2 +a23 e 3 +⋯+a 2n en
T (e3 )=a31 e1 +a32 e2 +a33 e 3 +⋯+a 3n en
⋮

T (e n )=an 1 e 1+a n2 e2+an 3 e 3 +⋯+a nn en

Definition: The transpose of the above matrix of coefficients denoted by [ T ]e is

called the matrix representation of T relative to the basis

{ei } or simply the matrix

of T in the basis
{ei }. That is

( )
a 11 a21 ⋯ an 1
[T ]e = a 12 a22 ⋯ an 2
⋮ ⋮ ⋯ ⋮
a1n a2 n ⋯ a nn

Example 1: Find the matrix representation of T : R 2 ⟶ R2 defined by

T ( x , y )=(2 y ,2 x− y ) relative to the standard basis { e1 =( 1 , 0 ), e 2=( 0 ,1 ) } of R2 .

5
Solution: If (a ,b)∈ R , then ( a , b )=ae 1 + be 2 .
2

T ( e1 )=( 0 , 2)=0 e1 + 2 e2

T ( e 2 )=( 2 ,−1 )=2 e1 −e 2

Therefore,
[T ]e = 0 2
2 −1( )
Example 2: Let T : R ⟶ R be a linear mapping defined by T ( x , y )=(4 x−2 y , 2 x+ y ).
2 2

Compute the matrix of T in the basis { e1 =( 1 , 1) , e2 =(−1 , 0) }

Solution: Let (a ,b)∈ R , we write (a ,b) as a linear combination of e 1 and e 2 . Thus

2

( a , b )=λ 1 ( 1 ,1 )+ λ2 (−1 , 0 )

⇒ λ1 − λ2 =a , ⇒ λ 1= b , λ 2 =b− a , λ1 =b so

( a , b )=b ( 1, 1 )+( b−a )(−1 , 0 )=be 1 +( b−a) e 2

Now,

T (1, 1 )=(2 , 3)=3 e1 +e2

T (−1 , 0 )=(−4 ,−2 )=−2 e 1 +2 e 2

So
(
[T ]e = 3 −2
1 2 )
Example 3: Let V be a vector space of polynomials in t of deg≤3 and let D : V →V be
d
D( f (t ))= (f (t )).
the differential linear mapping defined by dt Compute the matrix of
2 3
D with respect to the basis {1, t , t , t }.

Solution:

D(1)=0=0⋅1+0⋅t+0⋅t 2 +0⋅t 3
D(t )=1=1⋅1+0⋅t +0⋅t2 +0⋅t 3
D(t 2 )=2t=0⋅1+2⋅t +0⋅t 2 +0⋅t 3
6
D(t 2 )=3 t 2 =0⋅1+0⋅t +3⋅t 2 +0⋅t3
So

( )
0 1 0 0
0 0 2 0
0 0 0 3
[D]=
0 0 0 0

Exercise:

1. Find the matrix representation of each of the following linear mappings

{
T : R ⟶ R relative to the standard basis e1 =(1,0,0),e 2=(0,1,0),e 3 =(0,0,1) .
3 3
}
i. T ( x , y , z )=(2 x−3 y +4 z , 5 x− y +2 z , 4 x +7 y )
ii. T ( x , y , z )=(2 y + z , x−4 y +2 z , 3 x )

2. Let
A= 1 2
3 4 ( ) and T be the linear mapping from
R
3

to
R
3

defined by
T (~ v where ~
v )= A ~ v is written as a column vector. Find the matrix of T in each
of the following basis.
i. { e1 =( 1 , 0 ), e 2=( 0 , 1) }
ii. { e1 =( 1 , 3) , e 2=( 2 ,5 ) }
3. Each of the sets

i. {1,t ,e t ,tet } and

ii. {e3 t ,te 3 t ,t 2 e3 t }
Is a basis of a vector space V of functions f : R ⟶ R . Let D be the differential
d
D( f (t ))= (f (t )).
mapping defined by dt Find the matrix of D in each of the
given basis.

Theorem: Let
{e1 ,e2 ,e 3 ,⋯,en }be a basis of V and let T be any operator on V. Then
~
for any vector v ∈ V ,
[T ]e [ ~v ]e=[ T (~v )]e .

Example: let T be the linear mapping on R defined by T ( x , y , z )=(2 y +z , x−4 y , 3 x )

3

7
 i. Find the matrix of T in the basis { f 1=(1,1,1),f 2=(1,1,0),f 3=(1,0,0)} .
ii. Verify that
[T ]f [ ~
v ]=[ T (~v )]f

Solution:

i. Let a , b , c ∈ R3 and set

(a,b,c)=λ1 (1,1,1)+λ2 (1,1,0)+λ 3(1,0,0)

So

λ1+λ2+λ3=a¿}λ1+λ2 =b¿}¿ ¿ ⇒ λ 1=c, λ 2=b−c , λ3 =a−b

Therfore,

(a,b,c)=c(1,1,1)+(b−c )(1,1,0)+(a−b)(1,0,0)
=cf 1 +(b−c )f 2 +(a−b) f 3

Since T ( x , y , z )=(2 y +z , x−4 y +2 z , 3 x )

T (f 1 )=T (1,1,1)=(3,−3,3)=3f 1−6 f 2+6 f 3

T (f 2 )=T (1,1,0)=(2,−3,3)=3 f 1−6f 2+5f 3
T (f 3 )=T(1,0,0)=(0,1,3)=3f 1−2 f 2−f 3

( )
3 3 3
[T ]f = −6 −6 −2
6 5 −1

~ ~v=(a,b,c) =cf +(b−c)f +(a−b)f

ii. Suppose v =(a , b , c ) then 1 2 3

( )
c
~
⇒[ v ]f = b−c
a−b also

8
 T (~
v )=(2b+c ,a−4 b , 3 a )

=3 af 1 +(a−4 b−3 a) f 2 +(2 b+c−a+4 b )f 3

=3 af 1 +(−2 a−4 b )f 2 +(−a+6 b+c )f 3

So that

( )
3a
~
[ T ( v )]f = −2 a−4 b
−a+6 b+c
Thus

( )( ) ( )
3 3 3 c 3a
[T ]f [ v ]= −6 −6 −2 b−c = −2 a−4 b =[ T ( ~
~ v ) ]f
6 5 −1 a−b −a+6 b+c
Exercise:

Let T be the linear mapping T : R ⟶ R defined by T ( x , y )=(5 x+ y , 3 x−2 y ). Find the

2 2

matrix of T with respect to the basis { f 1 =( 1 , 2) , f 2 =( 2 , 3) }. Verify that

[T ]f [ ~
v ]=[ T (~v )]f
~
for a vector v in R .
2

CHANGE OF BASIS

Let
{e1 ,e2 ,e 3 ,⋯,en }be a basis of V and let {f 1 , f 2 ,f 3 ,⋯,f n }be another basis. Suppose

f 1=a11 e 1 +a 12 e 2 +a 13 e 3 +⋯+a1n en
f 2=a21 e1 +a22 e2 +a23 e 3 +⋯+a 2n en
f 3 =a31 e1 +a32 e2 +a33 e 3 +⋯+a 3n en
⋮

f n =an 1 e 1 +a n2 e 2 +an 3 e 3 +⋯+a nn e n

9
Then the transpose P of the above matrix of coefficients is called the transitional

matrix from the ‘old’ basis

{ei } to the ‘new’ basis {f i }

( )
a 11 a21 ⋯ an 1
P= a12 a22 ⋯ an 2
⋮ ⋮ ⋯ ⋮
a 1n a2 n ⋯ a nn

Since the vectors are linearly independent, the matrix P is invertible. In fact, the the
−1
inverse P is the transitional matrix from the basis
{f i } back to the basis {ei }.

Example: Consider the following basis of R , { e1 =( 1 , 0 ), e 2=( 0 , 1) } and

2

{ f 1 =( 1 , 1) , f 2 =(−1 , 0) } . Then,
f 1=( 1 ,1 )=( 1 , 0 )+( 0 , 1)=e1 + e2

f 2=(−1, 0 )=−( 1, 0 )+0( 0 , 1)=−e 1 +0 e 2

Hence the transitional matrix P from the basis

{ei }to the basis {f i } is

P= (11 −10 )
Recall: If (a ,b)∈ R , ( a , b )=bf 1 +( b−a ) f 2 so,
2

e 1=( 1 , 0)=0( 1 ,1 )−(−1 , 0)=0 f 1 −f 2

e 2 =( 0 , 1)=( 1 ,1 )+(−1 ,0 )=f 1 + f 2

Hence the transitional matrix Q from

{f i } back to the basis {ei } is

(
Q= 0 1
−1 1 )
Observe that P and Q are inverses

( )(
PQ= 1 −1 0 1 = 1 0 =I
1 0 −1 1 0 1 )( )

10
Theorem: Let P be the transitional matrix from a basis
{ei } to a basis {f i } in a

vector space V. Then for any vector

~v ∈ V
,
P[ ~
v ]f =[~v ]e and hence
[~v ]f =P−1 [ ~
v ]e .

−1
Thus the effect of P is to transform the coordinate vector in the new basis
{f i }

back to the old basis

{ei }.
Example 1: Let ~
v ¿( a , b)∈ R2 and consider the following basis of R2 ,
{ e1 =( 1 , 0 ), e 2=( 0 , 1) } and { f 1 =( 1 , 1) , f 2 =(−1 , 0) } . Define T by T ( x , y )=(4 x−2 y , 2 x+ y ).

Recall: (a , b )=b (1, 1+(b−a)(−1 , 0)

=bf 1 +( b−a) f 2 , so
v ]f = b
[~
b−a( ) .

Also ( a , b )=a ( 1 ,0 )+b( 0 ,1 )=ae 1 +be 2 .

So
T ( e1 )=T ( 1 , 0)=( 4 , 2)=4 e 1 +2 e 2

T ( e 2 )=T ( 0 , 1)=(−2 ,1 )=−2 e 1 +e 2

Therefore,
[T ]e = 4 −2
2 1 ( ) .
Also
T ( f 1 )=T ( 1 ,1 )=( 2, 3 )=3 f 1 +f 2

T ( f 2 )=T (−1 ,0 )=(−4 ,−2)=−2 f 1 +2 f 2

So
[T ]f = (31 −22 ) .

From the previous example

P= ( 1 −1
1 0 ) and
P−1 = ( 0 1
−1 1 ) .

Now

11
 P−1 [T ]e P= ( 0
)(
1 4 −2 1 −1
−1 1 2 1 1 0 )( )
= ( 2
−2 3 1 0
= )(
1 1 −1 3 −2
1 2
=[T ]f )( )
Example 2: Consider the following basis of R3 ,

{ e1=(1,0,0),e 2=(0,1,0),e 3=(0,0,1)} and

{ f 1=(1,1,1),f 2=(1,1,0),f 3=(1,0,0)}
i.
~ 3
For v ¿( a , b , c) ∈ R , find
[~v ]e and
[~v ]f

ii. Find the transitional matrix P from the basis

{ei }to the basis {f i } and Q

from the basis

{f i } to the basis {ei }. Verify that Q=P−1 .

iii. Show that

[T ]f =P−1 [ T ]e P where T is defined by T ( x , y , z )=(2 y +z , x−4 y , 3 x ).

Solution:
~v=(a,,b,c)=a(1,0,0)+b(0,1,0)+c(0,0,1)
i.
=ae 1 +be 2 +ce 3

()
a
~
[ v ]= b
Therefore,
c
Also
~v=(a,b,c)=λ (1,1,1)+ λ (1,1,0)+λ (1,0,0)
⇒
1 2 3
λ1+λ+2 λ3=a¿} λ1+λ2 =b ¿}¿¿ ⇒λ1=c,λ2=b−c,λ3=a−b¿¿
So
¿
~v=(a,,b,c)=cf +(b−c)f +(a−b)f
1 2 3
Therefore,

( )
c
~
[ v ]f = b−c
a−b

12
ii.
f 1=(1,1,1)=e1 +e2 +e 3
f 2=(1,1,0)=e 1 +e 2 +0e 3
f 3=(1,0,0)=e1 +0e2 +0e 3

( )
1 1 1
P= 1 1 0
Therefore,
1 0 0

e 1=(1,0,0)=0 f 1 +0 f 2 +f 3
e 2=(0,1,0)=0f 1+f 2−f 3
e 1=(0,0,1)=f 1 −f 2 +0f 3

( )
0 0 1
Q= 0 1 −1
So that
1 −1 0

( )( )( )
1 1 1 0 0 1 1 0 0
PQ= 1 1 0 0 1 −1 = 0 1 0
1 0 0 1 −1 0 0 0 1 and so P−1 =Q .
iii. Since T ( x , y , z )=(2 y +z , x−4 y , 3 x ),
T (e1 )=T (1,0,0)=(0,1,3)=0e 1+e 2+3 e3
T (e 2 )=T(0,1,0)=(2,−4,0)=2e1−4 e 2+0e 3
T (e3 )=T(0 ,0,1)=(1,0,0)=e1+0e2 +0e 3

( )
0 2 1
[T ]e = 1 −4 0
Therefore,
3 0 0

T (f 1 )=T (1,1,1)=(3,−3,3)=3f 1−6 f 2+6 f 3

T (f 2 )=T (1,1,0)=(2,−3,3)=3 f 1−3 f 2+3 f 3
T (f 3 )=T(1,0,0)=(0,1,3)=3f 1−2 f 2−f 3

13
 ( )
3 3 3
[T ]f = 6 −6 −2
Therefore,
6 5 −1
Now,

( )( )( ) ( )
0 0 1 0 2 1 1 1 1 3 3 3
P[ T ]e P= 0 1 −1 1 −4 0 1 1 0 = −6 −6 −2
1 −1 0 3 0 0 1 0 0 6 5 −1

Hence
[T ]f =P−1 [ T ]e P
Exercise:

Consider the following basis of R ,

3
{ e1=(1,0,0),e 2=(0,1,0),e 3=(0,0,1)} and
{ f 1=(1,−1,0), f 2=(1,0,−1),f 3=(0,0,1)}.
~v ¿ ( a , b , c ) ∈ R3 , [ ~v ] [ ~v ] .
i. For find e and f

ii. Find the transitional matrix P from

{ei } to {f i } and Q from {f i } to {ei }.
−1
iii. Verify that P =Q

iv. Show that

P−1 [~v ]e=[~v ]f ~ 3
for any v ∈ R .
−1
v. Show that the matrix representation
[T ]f =P [ T ]e P where T is defined by
T ( x , y , z )=(−2 x+ y+ z , x−2 y+ z , x + y−2 z ).

MATRIX REPRESENTATION OF LINEAR MAPPINGS

So far we have been dealing with linear mappings from one vector space to the
same vector space i.e T :V →V . These linear mappings are normally called linear
operators.
Next we consider the general case of linear mappings from one vector space into
another. Let U and V be vector spaces over the same field K and dim U =m and

dim V =n . Furthermore let {e1 ,e2 ,e 3 ,⋯,em } be a basis of U and {f 1 , f 2 ,f 3 ,⋯,f n }be a

basis of V . Suppose T :U →V is a linear mapping. Then

T (e1 ), T( e2 ), ⋯, T (e m ) are
vectors in V and so each of them can be written as a linear combination of
{f 1 , f 2 ,f 3 ,⋯,f n };

14
 T (e1 )=a11 f 1 +a 12 f 2 +a13 f 3 +⋯+a 1n f n
T (e 2 )=a21 f 1 +a22 f 2 +a23 f 3 +⋯+a2 n f n
⋮

T (e m )=a m1 f 1 +am2 f 2 +am3 f 3 +⋯+amn f n

The transpose of the above matrix of coefficients denoted by

[ T ]ef is called the
{e ,e ,e ,⋯,em } and {f 1 , f 2 ,f 3 ,⋯,f n }.
matrix representation of T relative to the basis 1 2 3

( )
a11 a21 ⋯ an 1
[ T ]ef = a12 a22 ⋯ an 2
⋮ ⋮ ⋯ ⋮
a1 n a 2n ⋯ ann
That is, .

Theorem: For any vector v ∈ V ,

[ T ]ef [ v ]e =[ T ( v ) ] f .

That is if we multiply the coordinate vector of v in the basis { i } by

f
e [ T ]e we obtain
the coordinate vector of T (v ) with respect to the basis { f i} .

Example 1: Let T : R 3 ⟶ R2 be the linear mapping defined by

T ( x , y , z )=(3 x +2 y−4 z , x−5 y +3 z );

Fine the matrix representation of T relative to the following basis of R

3
i.

and R ,
2
{ f 1=(1,1,1),f 2=(1,1,0),f 3=(1,0,0)} and { g =( 1, 3 ), g =( 2 , 5) }.
1 2

ii. Verify that for any vector v ∈ R ,

3
[ T ] gf [ v ]f =[ T ( v )] g
Solution:

i. Let (a ,b) ∈ R2 , then

( a , b )=λ 1 ( 1 ,3 )+ λ 2 ( 2 ,5 )

⇒¿λ 1+2λ2=a ¿ }¿ ¿ ⇒λ2=3a−b and λ1=2b−5a¿

And so ( a , b )=( 2b−5 a ) g1 +( 3 a−b ) g 2
Since T ( x , y , z )=(3 x +2 y−4 z , x−5 y +3 z );
T ( f 1 )=( 1 ,−1)=−7 g 1 + 4 g2

15
 T ( f 2 )=( 5 ,−4 )=−33 g1 +19 g2

T (f 3 )=(3,1)=−13 g1 +8g 2

And so
(4
[ T ] gf = −7 −33 −13
19 8 )
ii. Let v=( a ,b , c ), then recall:
(a,b,c)=cf 1 +(b−c)f 2 +( a−b)f 3 so that

( )
c
[ v ] f = b−c .
a−b
T (v )=(3 a+2b−4 c , a−5 b+3 c )

= [ 2( a−5 b+3 c )−5(3 a+2b−4 c ) ] g1 + [ 3(3 a +2 b−4 c )−(a−5 b+3 c ) ] g2

=(−13 a−20 b+26 c )g 1 +(8 a+11b−15 c )g 2

So
( 8 a+11b−15 c )
[ T (v ) ]g = −7−20 b+26 c

Now

)( )(
c
g
f (
[ T ] [ v ]f = −7 −33 −13
4 19 8
b−c =
a−b
−7 c−33 b+33 c−13 a+13 b
4 c+19 b−19 c+8 a−8 b )
(
= −13 a−20 b+26 c
8 a+11b−15 c )
Example 2: Let T : R ⟶ R be defined by T ( x , y )=(2 x−3 y , x +4 y ). Find the matrix of
3 2

T in the basis { e1 =( 1 , 0 ), e 2=( 0 , 1) } and { f 1 =( 1 , 3) , f 2 =( 2 , 5 ) } of R2.

Solution: If (a ,b)∈ R , then ( a , b )=( 2b−5 a ) f 1 +( 3 a−b) f 2 and since

2

T ( x , y )=(2 x−3 y , x +4 y ),
T ( e1 )=( 2 ,1 )=−8 f 1 +5 f 2

T ( e 2 )=(−3 , 4 )=23 f 1−13 f 2

16
And so
(5
[ T ]ef = −8 23
−13 )

Example 3: Let
Pn denote the vector space of all polynomials in x of deg≤n over
R.

Let
Pn : P2 →P 3 be the linear mapping defined by T ( P( x ))=xP( x−3 ) , that is
T (c 0 +c1 x+c 2 x 2 )=x(c 0 +c2 ( x−3)+c3 ( x−3 )2 ). Find the matrix representation
[ T ] BB '
where B= {1, x ,x 2 } and B'= {1, x ,x 2 , x 3 } .

Solution:

T (1)=x=0⋅1+1⋅x+0⋅x 2 +0⋅x 3
T ( x)=x( x−3 )=x 2−3 x=0⋅1+3⋅x+1⋅x 2 +0⋅x 3
T ( x2 )=x( x−3)2 =x(x 2 −6 x+9)=x 3 −6 x 2 +9 x=0⋅1+9⋅x−6⋅x 2 +1⋅x 3

So

( )
0 0 0
[ T ] BB ' = 1 −3 9
0 1 −6
0 0 1

Example 4: Let V =P 2 [ x ] be the vector space of polynomials of deg≤2 . Define a

linear operator T : V →V by T [ P( x ) ] =P ( 2 x +3 ). Find the matrix of relative to the

basis {1+x+x 2 , x+x 2 , x 2} .

Solution:

T (1+ x+x 2 )=1+(2 x +3 )+(2 x+3 )2 =13+14 x +4 x 2

=λ1 (1+ x+x 2 )+λ 2 ( x+ x2 )+λ 3 ( x 2 )

17
 ⇒¿λ1=13 ¿ } λ1+λ2=14 ¿ }¿ ¿⇒λ1=13, λ2=1, λ3=−10¿
Therefore, T (1+ x+x 2 )=13(1+x+x 2 )+1( x+x 2 )−10(x 2 )
T ( x+x 2 )=2 x +3+(2 x +3)2 =12+14 x +4 x2

=λ1 (1+ x+x 2 )+λ 2 ( x+ x2 )+λ 3 ( x 2 )

⇒¿λ1=13 ¿ } λ1+λ2=14 ¿ }¿ ¿⇒λ1=12, λ2=2, λ3=−10¿

Therefore, T (1+ x+x 2 )=12(1+x+x 2 )+2(x+x 2 )−10( x 2 )
T ( x2 )=(2 x+3 )2 =9+12 x+ 4 x 2

=λ1 (1+ x+ x 2 )+ λ 2 ( x+ x2 )+ λ 3 ( x 2 )

⇒¿λ1=9 ¿ } λ1+λ2=12 ¿}¿ ¿⇒λ1=9, λ2=3, λ3=−8¿

Therefore, T ( x2 )=9(1+x+ x2 )+3( x+x 2 )−8( x 2 ) and

( )
13 12 9
[ T ]= 1 2 3
−10 −10 −8

Example 5: : Let
f :R ⟶R
3 2

be a linear mapping with

(
A= 2 5 −3
1 −4 7 ) as linear
matrix of f with respect to the standard of R and R . Find the matrix of f with
3 2

respect to the basis

B= {f 1=(1,1,1), f 2=(1,1,0), f 3=(1,0,0)} of R3 and
B' = {g 1=( 1 ,3 ) , g 2=( 2, 5 ) } of 2
R.

18
Solution: ( x , y , z)=x (1 , 0 ,0 )+ y (0 , 1 ,0 )+z(0 , 0 ,1 ) and therefore,

f ( x, y ,z )=f [ x(1,0,0)+ y(0,1,0)+z(0,0,1)]=xf (1,0,0)+ yf (0,1,0)+zf (0,0,1)

=x(2,1)+x(5,−4)+z(−3,7)

=(2x+5 y−3 z ,x−4 y+7 z)

Let (a ,b)∈ R2 , then

( a , b )=λ 1 ( 1 ,3 )+ λ 2 ( 2 ,5 )

⇒¿λ 1+2λ2=a ¿ }¿ ¿ ⇒λ2=3a−b and λ1=2b−5a¿

And so ( a , b )=( 2b−5 a ) g1 +( 3 a−b ) g 2

Now,

T ( 1, 1 , 1)=( 4 , 4 )=−12 g 1 +8 f 2

T ( 1, 1 , 0)=( 7 ,−3)=−41 g1 +24 f 2

T ( 1, 0 , 0 )=( 2 , 1)=−8 g 1 +5 f 2

Therefore,

[ T ] g=B
f =B = (8
' −12 −41 −8
24 5 )
Example 6: Let

( )
1 3 −1
A= 2 0 5
6 −2 4 be the matrix of T : P2 → P 2 with respect to the
B= {v 1 , v 2 , v 3 } v 1=3 x +3 x3
v 2=−1+3 x+2 x2
v 3=3+7 x−2 x 2
basis where , and . Find;

i. [ T (v 1 )]B , [ T (v 2 )] B , [ T (v 1 )]B ,
ii. T (v 1 ), T ( v 2 ) , T (v 3 ),

19
Solution:

() () ()
1 3 −1
[ T (v 1 )]B = 2 , [ T (v 2 )] B= 0 , [ T (v 3 )] B= 5 ,
i.
6 −2 4

ii.
T (v 1 )=1 v 1 +2 v 2 +6 v 3=1 (3 x+3 x 2 )+2 ((−1+3 x+2 x 2 )+6 (3+7 x−2 x 2 )

=16 +51 x−5 x 2

T (v 2 )=3 v 1 +0 v 2 −2 v 3 =3(3 x+3 x 2 )+0 ((−1+3 x+2 x 2 )−2(3+7 x−2 x 2 )

−6−5 x+13 x 2

T (v 3 )=−1 v 1 +5 v 2 +4 v 3=−1(3 x+3 x 2 )+5 ((−1+3 x+2 x 2 )+4 (3+7 x−2 x 2 )

=7 +40 x−x 2

EXERCISE
1. Find the matrix representation of each of the following linear mappings
relative to the standard basis of Rn ;
T : R ⟶ R defined by T ( x , y )=(3 x− y , 2 x +4 y ,5 x−6 y ).
2 3
i.
T : R ⟶ R defined by T ( x , y , s , t )=(3 x−4 y+2 s−5t , 5 x +7 y−s−2 t ).
4 2
ii.
2. Let T : P2 → P 2 be the linear mapping defined by T [ P( x ) ] =P ( 2 x +1 ) that is
T (c 0 +c1 x +c 2 x 2 )=c 0 +c 1 (2 x +1 )+c 2 (2 x+1 )2 ).
i. Find [ T ] B with respect to the basis B= 1, x ,x { 2
}
2
ii. Compute T (2−3 x+4 x )
3. Let f : R 3 ⟶ R2 be the linear mapping which is represented by the matrix

A= (−74 −33 −13

19 8 ) with respect to the following basis:
{
R : e1 =(1,1,1),e 2 =(1,1,0),e 3 =(1,0,0)
3
}
R : { g1 =( 1, 3 ), g2 =( 2 , 5) } . Find:
2

20
 i. [ T (e 1)] g , [ T (e 2) ]g , [ T (e 3) ]g ,
ii. T (e1 ), T ( e 2) , T (e3 ),
4. Let T : R ⟶ R be a linear mapping such that T (1, 0 , 0 , 0 )=(2 , 3 ,6 ),
4 3

T (0 ,1 , 0 , 0 )=(1 , 2 ,0 ) , T (0 , 0 ,1 , 0 )=(−1 ,2−3) and T (0 , 0 ,0 ,1 )=(0 , 2 ,−1 ). find the matrix

representation of T relative to the basis;
R : (1,1,0,0),(0,1,1,0),(0,0 ,1,1),(1,0,0,1) and
4

3
R : (1,0,0), (0,1,0), (0,0,1)

CONJUGATION OF MATRICES

Suppose A and B are square matrices for which there exists an invertible matrix
P such that B=P−1 AP , then B is said to be similar to A . Note that if B is similar to
A then A is similar to B.

Theorem: Similarity is an equivalent relation.

Proof:

i. The identity I is invertible and I =I

−1
. Since A=I −1 AI ,
A is similar to A .
ii. Suppose A is similar to B , then there exists an invertible matrix P such
−1
that A=P BP . Hence B=(P−1 )−1 AP−1 and P−1 is invertible. Thus B is
similar to A .
iii. Suppose A is similar to B and B is similar to C . Then there exists
−1
invertible matrices P and Q such that A=P BP and B=Q CQ . Hence
−1

A=P−1(Q−1 CQ) P=(QP)−1 CQP and QP is invertible. Thus A is similar to C .

Theorem: If A and B are similar, then they have the same determinant.

21
Proof: Since A and B are similar, there exists an invertible matrix P such that
A=P−1 BP so

det A=det (P−1 BP )=det (P−1 )⋅det B⋅det P

=det( P−1 )⋅det P⋅det B
1
= ⋅det P⋅det B
det P
=det B So A and B have the same determinant.

Definition: The determinant of a linear mapping T , det T is defined as the

determinant of any matrix representation of T .

Example: Let T : R ⟶ R be defined by T ( x , y )=( x + y ,−2 x+4 y ). Find the

2 2

determinant of T .

Solution: We choose any basis B of R and compute the determinant of [ T ] B . If we

2

choose B= {e1 =(1 , 0 ) ,e 2 =( 0 , 1) }

T (1, 0 )=(1 ,−2)=e1 −2 e 2

T ( 1, 0 )=(1 , 4 )=e 1 +42 e 2

So
(
[ T ] B= 1 1
−2 4 ) and det T =1×4−(−2 )( 1 )=6

Had we chosen B' = {(1,1),(1,2)} , let (a ,b) ∈ R2 then, ( a , b )=λ 1( 1 ,1 )+ λ2 ( 1 ,2 )

⇒¿λ 1+λ 2=a ¿ } ¿¿⇒λ 2=b−a, λ 1=2a−b¿

And (a , b )=(2 a−b )(1 ,1 )+(b−a )(1 ,2 )

T (1, 1 )=(2 , 2)=2 (1 , 1)+0 (1, 2 )

T (1, 2 )=(3 , 6 )=0(1 , 1)+3(1 , 2)

So
(0 3 )
[ T ] B '= 2 0
and det T =2×3−(0)(0 )=6 .

22
Definition: A linear mapping T is said to be diagonalizable if for some basis
{e1 ,e2 ,e 3 ,⋯,en } it can be represented as a diagonal matrix.

Theorem: Let A be a matrix representation of a linear mapping T . Then T is

diagonalizable if and only if there exists an invertible matrix P such that P AP is a
−1

diagonal matrix.

Example: Let T ( x , y )=( x +4 y , 2 x+3 y ) and let { e1 =( 1 , 0 ), e 2=( 0 , 1) } and

{ f 1 =( 1 , 1) , f 2 =( 2 ,−1) } be two basis of R2.

a. Find the transitional matrix P from { ei } to { f i } and the transitional matrix

Q=P−1 from f i to ei .{ } { }
−1
b. Show that
P [ T ] e P=[ T ] f is a diagonal matrix of T , that is T is
diagonalizable.

Solution:

a. f 1=( 1 ,1 )=e 1 +e 2

f 2=( 2 ,−1 )=2 e1 −e 2 . So

(
P= 1 2
1 −1 )
Next let (a ,b)∈ R and set ( a , b )=λ 1 ( 1 ,1 )+ λ2 ( 2 ,−1) which gives
2

a−b a+2b
λ1 +2 λ2 =a ¿ } ¿¿⇒ λ 2= , λ1 = ¿
3 3

Therefore
(a , b )= ( a+23 b ) f +( a−b3 ) f
1 2
.
Now

1 1
e 1=(1 , 0)= f 1 + f 2
3 3

( )
1 2
3 3
Q=P−1=
2 1 1 1
e 1=(0 , 1)= f 1 − f 2 −
3 3 . So
3 3

23
 b. Since T ( x , y )=( x + y ,−2 x+4 y ),
T ( e1 )=T ( 1 , 0)=( 1 ,2 )=e1 +2 e 2

T ( e 2 )=T ( 0 , 1)=( 4 , 3)=4 e 1 +3 e 2

So
(2 3 )
[ T ]e = 1 4

Also
T ( f 1 )=T ( 1 ,1 )=(5 , 5 )=5 f 1 +0 f 2

T ( f 2 )=T ( 2 ,−1 )=(−2 , 1)=0 f 1 −f 2

Therefore
[ T ]f = 5 ( 0
0 −1 ) a diagonal matrix.

( )( ( )(
1 2 1 2
P−1 [ T ] e P= 3
1
−
3
1 )(
1 4 1 2 = 3
2 3 1 −1 1 ) −
1 1 −1 0 −1 )(
3 1 2 = 5 0 = [T ]
f )
Now
3 3 3 3 . Hence T is
diagonalizable.

Exercise

Let V be the vector space of 2×2 matrices over

R
and let
M= 1 2 ,
3 4 ( ) be a linear
mapping on V defined by T ( A )=MA . Find the matrix of T with respect to the

standard basis of V . (Hint: Standard basis of V is

({ 10 00 ) ,( 00 10 ) ,( 01 00 ),( 00 01 )}
)

THEORY OF DETERMINANTS

Recall:

a. If
A=
( a11
a21
a 12
a 22 ) , a 2×2 matrix, then its determinant is given by;

24
 a a 1 2
|A|=| 11 12|=a 11 a22−a12 a 21 | |=(1×4)−(2×3)=−2
for example 3 4
a21 a22 .

( )
a11 a12 a 13
A= a21 a22 a 23
a31 a32 a 33 3×3
b. If ,a matrix, then its determinant is given by;
a11 a12 a 13
a a a a a a
|A|=|a21 a22 a 23|=a11| 22 23|−a12| 21 23|+a13| 21 22|
a32 a33 a31 a33 a 31 a32
a31 a32 a 33 .

Example:
1 2 3
1 1 2 1 2 1
|2 1 1 |=1| |−2| |+3| |=1(2−1)−2( 4−4)+3(2−4)=1−2(0)+3(−2)=−5
1 2 4 2 4 1
4 1 2
Exercise:

( )
t +3 −1 1
A= 5 t−3 1
Evaluate the determinant of
6 −6 t +4

Properties of Determinants:

Let A be a square matrix. Then

1. If A has a row or column of zeros only, then det A=0 .

2. If A has two identical rows or columns, then det A=0 . If one row or column is
a scalar multiple of another then det A=0 .
3. If a multiple of a row of A is added to another row to produce matrix B , then
det B=det A .
4. If two rows of A are interchanged to produce matrix B , then det B=−det A .
5. If one row of A is multiplied by a scalar k to produce a matrix B then
det B=k det A .
T
6. det A=det A .

25
 7. If B is another matrix of the same order as A , then det ( AB )=det A⋅det B .
8. If A is a diagonal or a triangular matrix, then det A is equal to the product of
the entries in the main diagonal.

Determinant of matrices of order 4×4 and more:

There are two main techniques of computing these determinants:
a) Cofactor Expansion

Let A be a matrix of order n×n and let

a ij be a general element of A . The

minor of element
a ij which we denote by
M ij is the determinants of the
(n−1)×(n−1 ) submatrix which is obtained after deleting the i th row and jth
column of A . The number
Aij =(−1)i+ j M ij a.
is called the cofactor of ij

Example:

( )
2 3 −4
A= 0 −4 2
Let
1 −1 5 a
, compute the minors and cofactors of a 11 , a 12 and 13
Solution:
−4 2 0 2 0 −4
M 11=| |=−18, M 12=| |=−2, M 13=| |=4 .
−1 5 1 5 and 1 −1

A11 =(−1)2 (−18)=−18, A12=(−1)3 (−2)=2 and A13=(−1)4 (4 )=4 .

26
Theorem:(Laplace’s expansion Theorem) The determinant of the matrix
A=aij is
equal to the sum of the products obtained by multiplying elements of a row (or
column) of A by their respective cofactors.
n n
|A|=a i1 Ai1 +ai 2 A i2 +⋯+ain Ain = ∑ aij Aij |A|=a 1 j A 1 j +a2 j A 2 j +⋯+anj A nj =∑ a ij A ij
j=1 and i=1

Note that the trick is to expand along a row or column that has a lot of zeros.

Example 1: Compute det A where

( )
4 0 0 0
7 −1 0 0
A=
2 6 3 0
5 −8 4 3
Solution: Expand along row 1 to get

−1 0 0 −1 0 0
2
|A|=4(−1) | 6 3 0 |−0+0−0=4| 6 3 0 |
−8 4 3 −8 4 3

{
4 (−1)|
3 0
|−0|
6 0
4 3 −8 3 −8 4
|+0|
6 3
}
| =4 {(−1)(9 )−0+0 }=−36

Or
Since A is a triangular matrix, the determinant can be obtained by multiplying
elements in the main diagonal;

|A|=4(−1)(3)(3)=−36

Example 2: Compute the determinant of A if

( )
3 −7 8 9 −6
0 2 −5 7 3
A= 0 0 1 5 0
0 0 2 4 −1
0 0 0 −2 0
Solution:

27
Expand along column 1 to get:

2 −5 7 3 2 −5 7 3
0 1 5 0 0 1 5 0
|A|=(−1)2(3)| |=3| |
0 2 4 −1 0 2 4 −1
0 0 −2 0 0 0 −2 0
Next expand the 4×4 submatrix obtained along the first column,

1 5 0 1 5 0
2
|A|=(3)(2)(−1) |2 4 −1 |=6|2 4 −1 |
0 −2 0 0 −2 0

{
6 1|
4 −1
−2 0
|−5|
2 −1
0 0
|+0|
2 4
0 −2 }
| =6 {−2−0+0 } =−12

Example 3: Compute the determinant of

( )
4 0 −7 3 −5
0 0 2 0 0
A= 7 3 −6 4 −8
5 0 5 2 −3
0 0 9 −1 2
Solution: Expand along column 2 to have;

4 −7 3 −5 4 −7 3 −5
0 2 0 0 0 2 0 0
|A|=(3)(−1)5| |=(−3)| |
5 5 2 −3 5 5 2 −3
0 9 −1 2 0 9 −1 2
Next expand along row 2 to get,

4 3 −5
4
|A|=(−3)(2)(−1) |5 2 −3 |
0 −1 2

{
=(−6) 4|
2 −3
−1 2
|−3|
5 −3
0 2
|−5|
5 2
0 −1
| =−6{4−30+25}=6 }
28
 b.Gauss Elimination Method:
In this technique multiples of rows or columns are added to other rows or
columns creating zeros which simplifies subsequent calculations. As we saw
earlier, this operation does not change the value of the determinant.

Example 1: Compute det A given that

( )
1 −4 2
A= −2 8 −9
−1 7 0
Solution: The strategy is to reduce A to echelon form and then use the fact that
the determinant of a triangular matrix is the product of the leading diagonal
entries.

1 −4 2 1 −4 2 1 −4 2
|A|=|−2 8 −9 |=|0 0 −5 |=(−1)|0 3 2 |=(−1)(1)(3)(−5)=15
−1 7 0 0 3 2 0 0 −5
Example 2: Compute det A where

( )
2 −8 6 8
3 −9 5 10
A=
−3 0 1 −2
1 −4 0 6
Solution: We first factor 2 from the top row,

1 −4 3 4
3 −9 5 10
|A|=2| |
−3 0 1 −2
1 −4 0 6
Next we perform elementary operations.

1 −4 3 4 1 −4 3 4 1 −4 3 4 1 −4 3 4
3 −9 5 10 0 3 −4 −2 0 3 −4 −2 0 3 −4 −2
|A|=2| |=2| |=2| |=2| |
−3 0 1 −2 0 −12 10 10 0 0 −6 2 0 0 −6 2
1 −4 0 6 0 0 −3 2 0 0 −3 2 0 0 0 1

29
 =(2)(1 )(3 )(−6 )(1 )=−36
Sometimes the two methods can be combined;
Example 3: Compute the determinant of

( )
5 4 2 1
2 3 1 −2
A=
−5 −7 −3 9
1 −2 −1 4
Solution: Make use of 1 in the third column to create zeros below and above it:

5 4 2 1 1 −2 0 5
2 3 1 −2 2 3 1 −2
|A|=| |=| |
−5 −7 −3 9 1 2 0 3
1 −2 −1 4 3 1 0 2
Next expand along column 3 to get,

1 −2 5
5

3 1 2
{
1 2 3 2 3 1 }
|A|=(1)(−1) |1 2 3|=−1 1|2 3 |+2|1 3 |+5|1 2| =−1 {(1)(1)+2(−7)+5(−5)} =38

Example 4: Compute the determinant of A given that

( )
0 1 2 −1
2 5 −7 3
|A|=
0 3 6 2
−2 −5 4 −2
Solution: Make use of 2 in column 1 to create a zero below it

0 1 2 −1 0 1 2 −1
2 5 −7 3 2 5 −7 3
|A|=| |=| |
0 3 6 2 0 3 6 2
−2 −5 4 −2 0 0 −3 1
Next expand along column 1 to obtain,

1 2 −1 1 2 −1
3
|A|=(2)(−1) |3 6 2 |=−2|3 6 2 |
0 −3 1 0 −3 1
30
 {
=−2 1|
6 2 3 2 3 6
|−2| |−1|
−3 1 1 0 0 −3 }
| =(−2) {(12)−(2)(3)−(1)(−9) }=−30

Example 5: Evaluate the determinant of,

( )
3 −2 −5 4
−5 2 8 −5
|A|=
−2 4 7 −3
2 −3 −5 8
Solution: First reduce A to a matrix which has 1 as an entry, such as adding twice
the first row to the second row. Then use this 1 to create zeros below and above it.

3 −2 −5 4 3 −2 −5 4 0 4 1 −5
−5 2 8 −5 1 −2 −2 3 1 −2 −2 3
|A|=| |=| |=| |
−2 4 7 −3 −2 4 7 −3 0 0 3 3
2 −3 −5 8 2 −3 −5 8 0 1 −1 2
Expand along column 1 to get,

4 1 −5 4 1 −5
3
|A|=(1)(−1) |0 3 3 |=(−1)|0 3 3 |
1 −1 2 1 −1 2

{ }
(−1) 4| 3 3 |−1|0 3 −5|0 3 || =−1 {4(9)−1(−3)−5(−3)}=−1{36+3+15}=−54
−1 2 1 2 1 −1

Exercise
Compute the determinant of each of the following Matrices;

( )
1 0 0 3
0 1 −2 0

( )
A= 3 1 −2 1
−2 3 −2 3
0 −3 3 3 1 1 −3 2
1. C=
2 0 2 3
3 3 1 −3

( )
2 5 −3 −1 3.
3 0 1 −3
B=
−6 0 −4 9
2.
4 10 −4 −1

31
 ( )
2 5 −3 −2
−2 −3 2 −5
D=
1 3 −2 2

( )
2 5 4 1
−1 −6 4 3
4. 4 7 6 2
G=
6 −2 −4 0
7.
1 3 1 0

( )
2 1 3 2
3 0 1 −2

( )
E= 1 −1 1 −1
1 −1 4 3
2 2 −1 1 1 −1 3 2
5. H=
4 2 1 3
0 1 3 0

( )
2 2 −2 0 8.
−4 2 0 1
F=
0 4 −1 −4
6.
3 1 3 −1

BLOCK MATRICES
Using a system of vertical and horizontal lines, we can partition a matrix A into
smaller matrices called blocks (or cells). The matrix A is then called a block matrix.
Clearly a given matrix can be divided into blocks in different ways as below;

( )( )( )
1 −2 0 1 3 1 −2 0 1 3 1 −2 0 1 3
A= 2 3 5 7 −2 = 2 3 5 7 −2 = 2 3 5 7 −2
3 1 4 5 9 3 1 4 5 9 3 1 4 5 9

32
The convenience of the partition into blocks is that the result of operations on block
matrices can be obtained by carrying out the computation with blocks just as if they
were the actual elements of the matrix.
Determinant of the block matrices:

Theorem: Consider the block matrix

M= A B
0 C ( ) where A and C are square

matrices. Then |M|=|A||C|. More precisely, if M is a triangular block matrix with

square matrices
A1 , A2 , A 3 ,⋯, Am on the leading diagonal,

( )
A1 B ⋮ C
0 A2 ⋮ D
M=
⋮ ⋮ ⋱ ⋮
0 0 ⋯ Am

Then
|M|=|A1||A 2|⋯|Am|.
Example: Find the determinant of the matrices

( )
2 5 0 0 0

( )
2 1 0 0 0 2 0 0 0
0 2 0 0 B= 0 0 4 2 0
A=
0 0 1 1 0 0 3 5 0
0 0 −2 4 0 0 0 0 7

Solution:
Since A is a triangular block matrix,

2 1 0 0
0 2 0 0 2 1 1 1
|A|=| |=| || |=4×(4+2)=24
0 0 1 1 0 2 2 4
0 0 −2 4

33
 2 5 0 0 0
0 2 0 0 0
|B|=|0 2 5 4 2 ||7|=4×(20−6)×7=392
0 4 2 0 |=| ||
0 2 3 5
0 0 3 5 0
0 0 0 0 7
Exercise:
Find the determinants of the following matrices:

( )(
2 8 0 0 0 0 0
0 2 0 0 0 0 0

)
0 0 4 2 0 0 0 3 1 0 0 0

( )
A= 0 0 1 3 0 0 0 0 3 0 0 0 2 1 0 0
0 0 0 0 0 3 0 B= 0 0 3 1 0 0 2 0 0
C=
0 0 0 0 0 1 0 0 0 0 3 1 0 0 2 0
0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 5

EIGEN VALUES AND EIGEN VECTORS

Polynomials of matrices and linear operators:

Consider the polynomial f (t ) over a field K;

f (t )=a n t n +a n−1 t n−1 +an−2 tn−2 +⋯+a 1 t+a0 .
If A is a square matrix over K, we define
f ( A )=an A n +an−1 A n−1 +an−2 A n−2 +⋯+a1 A+a0 I , where I is the identity matrix of the
same order as A.

34
In particular, we say that A is a zero or the root of the polynomial f (t ) if f ( A )=0 , a
zero matrix of the same order as A.

Example:

Let
A= (13 24 ) , f (t )=2t −3t+7
2
and g(t )=t2 −5t−2. Find f ( A ) and g( A ).

Solution:

f ( A )=2 A 2−3 A+7I

( ) ( ) ( ) ( )( )( )
2
=2 1 2 −3 1 2 +7 1 0 =2 7 10 − 3 6 + 7 0
3 4 3 4 0 1 15 22 9 12 0 7

( )( )( )(
= 14 20 − 3 6 + 7 0 = 18 14
30 44 9 12 0 7 21 39 )
g( A )= A2 −5 A−2I

( ) ( ) ( )( )( )( )( )
2
= 1 2 −5 1 2 −2 1 0 = 7 10 − 5 10 − 2 0 = 0 0
3 4 3 4 0 1 15 22 15 20 0 2 0 0

Exercise:

1. Find f ( A ) where
(A= 1 −2
4 5 ) and f (t )=t 2−3t +7.
A=(
2 3)
1 4
2. Show that is a zero of f (t )=t 2−4t−5.

35
Suppose T : V →V is a linear operator on a vector space V over a field K. If
f (t )=a n t n +a n−1 t n−1 +an−2 tn−2 +⋯+a 1 t+a0 , we define f ( T ) in the same way as we did

for matrices, that is

f (T )=a n T n +a n−1 T n−1 +an−2 T n−2 +⋯+a 1 T +a0 I where I is an
identity operator. We also say that T is the zero or root of the polynomial f (t ) if
f ( T )=0 , the zero operator on V .

Eigen Values and Eigen Vectors:

Let T : V →V be a linear operator on a vector space V over a field K. A scalar λ ∈ K

~
is called an eigen value of T if there exists a non-zero vector v ∈ V for which
T (~
v )=λ ~
v.
Every vector that satisfies this relation is then called an eigen vector of T belonging
to the eigen value λ .
~
Note that each scalar multiple k v is such an eigen vector:

T (k ~
v )=kT ( ~
v )=k ( λ ~
v )=λ (k ~
v)

The set of all such vectors is a subspace of V called the eigen space of λ . The terms
Characteristic value and Characteristic vector or proper value and proper vector are
frequently used instead of eigen value and eigen vector.

~ ~ ~ ~
Example1: Let I : V →V be the identity mapping. Then for every v ∈ V , I ( v )= v =1 v .
Hence 1 is the eigen value of I, and every vector in V is an eigen vector belonging to
I.

Example 2: Let D be the differential operator on the vector space V of

5t 5t
differentiable functions. We have, D(e )=5e . Hence 5 is an eigen value of D with
5t
eigen vector e .

36
If A is an n×n matrix over a field K, then λ ∈ K is an eigen value of A if for some non-
~ ~ ~ ~
zero vector v ∈ K then A v=λ v . In this case v is an eigen vector of A belonging to
the eigen value λ .

Theorem: Let T : V →V be a linear operator on a vector space over K. Then λ ∈ K is

an eigen value of T if and only if the operator λI −T is singular or is not invertible.
The eigen space of λ is then the kernel of λI −T .
~
Proof: λ is an eigen value of T if and only if there exists a nonzero vector v such
~ ~ ~ ~ ~ ~
that T ( v )=λ v or ( λI )( v )−T ( v )=0 or ( λI −T )( v )=0 ⇒ v ≠0 ∈ ker( λI −T ) . Hence
λI −T is singular.

We also have that is the eigen space of λ if and only if the above relations hold,
~
hence v is the kernel of λI −T .

Example 1: Let
( )
A= 1 4 .
2 3 Find the eigenvalues of A and a basis of each eigen
space.
Solution:

λI − A= ( 0λ 0λ )−(12 43 )=( λ−1

−2
−4
λ−3 )λ. is an eigen value if |λI−A|=0. That is
λ−1 −4
|λI− A|=| |=0 ⇒ λ 2−4 λ−5=0 ⇒ λ=5 or λ=−1
−2 λ−3
Therefore, the eigen values of A are λ=5 and λ=−1 .

To obtain the eigen space of λ=5 , we substitute λ=5 into the equation

( λI − A )~
v =0 to get (−24 2 )
−4 ¿ ( x ¿ ) ¿ ¿
¿ or
4x−4y=0¿} ¿¿ or x−y=0.¿ This system has 1
independent solution hence the dimension of the eigen space of λ=5 is 1. A basis
for this eigen space is {(1,1)}
~
The eigen space of λ=−1 , substitute λ=−1 in the equation ( λI − A ) v =0 to get

(
−2 −4 )
−2 −4 ( x ¿ ) ¿
¿
¿
¿
or
−2x−4 y=0¿} ¿ ¿ or x+2y=0.¿ This system has one independent

37
solution, hence the dimension of this eigen space is 1. A basis for this eigen space is
{(2 ,−1 )} .

Example 2: Find the eigen values and the basis for each of the eigen space of
T : R → R defined by T ( x , y , z )=(2 x + y , y −z , 2 y+ 4 z ).
3 3

Solution: First find the matrix of T, say to the usual basis of R3;

T (e1 )=T (1,0,0)=(2,0,0)=2e 1+0e 2+0 e3

T (e 2)=T(0,1,0)=(1,1,2)=e1 +e 2+2e 3
T (e3 )=T(0 ,0,1)=(0,−1, 4)=0e 1−e2+4e 3
Therefore,

( )
2 1 0
[ T ] =A= 0 1 −1
0 2 4

λ is an eigen value of A if |λI−A|=0 that is;

( )( )
λ 0 0 2 1 0
|λI − A|= 0 λ 0 − 0 1 −1 =¿ ( 0 ¿ ) ( 0 ¿ ) ¿ ¿ ¿
¿
0 0 λ 0 2 4

⇒ ( λ−2)|λ−1 1 |+|0 1 |+0=0

−2 λ−4 0 λ−4

⇒ ( λ−2) [ ( λ−1 )( λ−4 )+2 ]=0 ⇒( λ−2 )( λ2 −5 λ+6 )=0

⇒ ( λ−2)( λ−3 )( λ−2)=0 ⇒ λ=2 , 2 or λ=3 .

~
Therefore, the eigen values of λ=2 are obtained by solving ( λI − A ) v =0 . for λ=2 ,

38
 ( )
0 −1 0 − y =0
( λI− A )~v = 0 1 1 ¿ ( x ¿ ) ( y ¿ ) ¿ ¿ ⇔ y + z=0 y=0
¿
0 −2 −2 −2 y−2 z=0 or y+z=0
These equations have one independent solution, say (1,0,0) so the basis of this
eigen space is {(1,0,0)}.

Eigen space of λ=3 ,

( )
1 −1 0 x − y=0
( λI− A ) v = 0 2 1 ¿ ( x ¿ ) ( y ¿ ) ¿ ¿
~ ⇔ 2 y + z=0 x− y=0
¿ −2 y−z =0 2 y+z=0
0 −2 −1 or
The number of independent solutions in the above system is 1. So eigen values of
λ=3 has 1 dimension and its basis is {(1 ,1 ,−2)}.

Exercise:
1. For each Matrix below, find all the eigenvalues and a basis for each
eigenspace,

( )
−3 1 −1
C= −7 5 −1
i.
(
A= 3 −1
1 1 ii.
) 1 4(
B= 2 −1 ) iii.
−6 6 −2
2. Find the eigenvalues and a basis of each eigenspace of the operator of
T : R → R defined by T ( x , y , z )=( x + y + z ,2 y + z ,2 y +3 z ).
3 3

3. Suppose λ is an eigenvalue of an invertible operator T. Show that λ is an

−1

−1
eigenvalue of T .

Theorem: Non-zero eigenvectors of a matrix belonging to distinct eigenvalues are

linearly independent.

Diagonalization of Matrices

Let T : V →V be a linear mapping on a vector space V with finite dimension n. T can

be represented by a diagonal matrix

39
 ( )
λ1 0 0 ⋯ 0
0 λ2 0 ⋯ 0
0 0 λ3 ⋯ 0
⋮ ⋮ ⋮ ⋱ ⋮
0 0 0 ⋯ λn
if and only if there exists a basis
{~v 1 ,~v 2 , ~v 3 ,⋯,~v n } of V for which
T (~
v 1 )= λ1 ~
v1

T (~
v 2 )= λ 2 ~
v2

T (~
v 3 )= λ3 ~v 3
⋮

T (~
v n )= λn ~
vn
~v , ~v ,~v ,⋯, ~v
That is , the vectors 1 2 3 n are eigenvectors of T belonging to eigenvalues
λ 1 , λ2 , λ3 ,⋯, λ n .

Theorem: A linear operator T : V →V can be represented by a diagonal matrix D if

and only if V has a basis of eigenvectors of T. In this case the diagonal elements of
D, are the corresponding eigenvalues.

Alternatively: An n×n matrix A is similar to a diagonal matrix D if and only if A has

a linearly independent vectors. In this case the diagonal elements of D, are the
corresponding eigenvalues.
In the above theorem, if we let P to be the matrix whose columns are the linearly
−1
independent eigenvectors of A, then D=P AP .
If a linear operator or matrix can be represented by a diagonal matrix, we say that it
−1
is diagonalizable and if P AP=D is a diagonal matrix, then we say that P
diagonalizes A.

Example 1: Diagonalize
( )
A= 1 2
3 2

40
Solution:

|λI−A|=0 if λ is an eigenvalue of A.

λ−1 −2
|λI− A|=| |=( λ−1)( λ−2 )−6=0 ⇒ λ2 −3 λ−4=0 ⇒( λ−4 )( λ+1 )=0
−3 λ−2
⇒ λ=4 , λ=−1

Therefore, the eigenvalues of A are λ=4 , and λ=−1.

Eigen space of λ=4 ,

(−3
3 −2
2 )
¿ ( x ¿) ¿ ¿
¿ or 3 x−2 y=0

A basis of eigenspace of λ=4 is { ( 2, 3 ) } .

Eigenspace of λ=−1 ,

(−2
−3 −3 ¿)
−2 ¿ ( x ¿ ) ¿ ¿
or x + y=0

A basis for the eigenspace of λ=−1 is { (1 ,−1)} .

( )
1 1
P−1 =
(
1 −1 −1
= 5
) 5

Therefore
(
P= 2 1
3 −1 ) and
−5 −3 2 3
5
−
2
5

Now,

( ) ( )
1 1 4 4
P−1 AP=
5
3
−
( )(
5 1 2 2 1
2 3 2 3 −1
=
−
5
3 ) 5
2 (23 1
−1
= )(
4 0
0 −1 )
5 5 5 5 , a diagonal matrix.
Example 2: For each matrix find all the eigenvalues and a basis for each
eigenspace. Which matrix can be diagonalized and why?

( ) ( )
1 −3 3 −3 1 −1
A= 3 −5 3 B= −7 5 −1
a.
6 −6 4 b.
−6 6 −2

41
Solution:
a.

λ−1 3 −3
λ+5 −3 −3 −3 −3 λ+5
|λI− A|=| −3 λ+5 −3 |=( λ−1)| |−3| |−3| |=0
6 λ−4 −6 λ−4 −6 6
−6 6 λ−4

⇒( λ−1) [ ( λ+5 )( λ−4 )+18 ] −3 [ −3( λ−4 )−18 ] −3 [ −18+6 ( λ+5 ) ] =0

⇒ λ3 −12 λ−16=0

Using factor theorem, we get that λ=−2 is a root of the above

equation.

¿ − 8 λ 16 ¿ − 8 λ 16 ¿ 0 ¿
λ 2−2 λ−8=( λ+2)( λ−4) ⇒ λ3 −12 λ−16=( λ+2)2 ( λ−4)=0
⇒ λ=−2,−2 or λ=4

Therefore, the eigenvalues of A are λ=−2 ,−2 and λ=4 .

Eigenspace of λ=−2 ;

( )
−3 3 −3
−3 3 −3 ¿ ( x ¿ ) ( y ¿ ) ¿ ¿
¿
−6 6 −6 or x− y + z=0

Hence the dimension of the eigenspace of λ=−2 is 2 and a basis for

this eigenspace is

{ (1 , 1 , 0),(0 , 1 ,1 )}

42
 Eigenspace of λ=4 ,

( )
3 3 −3 x+ y−z=0
−3 9 −3 ¿ ( x ¿ ) ( y ¿ ) ¿ ¿ x−3 y+z=0
¿
−6 6 0 or x− y=0

x+ y−z=0
Or 2 y−z=0 .

A basis for the eigenspace of λ=4 is { (1 , 1 ,2 ) } ’

Since { (1 , 1 , 0),(0 , 1 ,1 ),(1 , 1 ,2 ) } span R then A is diagonalizable. Infact

3

( ) ( )
1 0 1 −2 0 0
−1
P= 1 1 1 D=P AP= 0 −2 0
0 1 2 and
0 0 4
b.

λ+3 −1 1
|λI−B|=| 7 λ−5 1 |=0 ⇒( λ+3)|λ−5 1 |+1|7 1 |+1|7 λ−5 |=0
−6 λ+2 6 λ+2 6 −6
6 −6 λ+2
⇒( λ+3 ) [ ( λ−5)( λ +2)+6 ] +1 [ 7( λ+2)−6 ] +1 [ −42−6( λ−5) ] =0

⇒ λ3 −12 λ−16=0 ⇒( λ+2 )2 ( λ−4 )=0 ⇒ λ=−2 , −2 or λ=4

Therefore the eigenvalues are λ=−2 , −2 and λ=4 .

Eigenspace of λ=−2 ,

( )
1 −1 1
7 −7 1 ¿ ( x ¿) ( y ¿ ) ¿ ¿ ¿
¿
6 −6 0
¿
There is only one independent solution. A basis for this eigenspace is { (1 , 1 , 0) } .

Eigenspace of λ=4 ,

43
( )
7 −1 1
7 −1 1 ¿ ( x ¿) ( y ¿ ) ¿ ¿
¿
6 −6 6

There is only one independent solution. A basis for this eigenspace is { (0 ,1 , 1) } .

Now { (1 , 1 , 0),(0 , 1 ,1 ) } are not enough to span R and so B is not diagonalizable.

3

Exercise:

1. Show that
P= 1 1
0 1 ( )
is not diagonalizable.
2. For each matrix find all the eigenvalues and a basis for each eigenspace.

( ) ( ) ( )
3 1 1 1 2 2 1 1 0
A= 2 4 2 B= 1 2 −1 C= 0 1 0
i.
1 1 3 ii.
−1 1 4 iii.
0 0 1

Where possible, find invertible matrices P1 , P2 and

P3 such that
P −1 AP1 ,
1

P −1 BP 2 P −1 CP3
2 and 3 are diagonal.

CHARACTERISTIC POLYNOMIAL AND CAYLEY-HAMILTON

THEOREM

D( x)=|xI−A| is called a
Let A be a square matrix, the determinant

characteristic polynomial of A. We also call |xI−A|=0 the characteristic

equation of A.
Example: Find the characteristic polynomial of the matrix

( )
1 3 0
A= −2 2 −1
4 0 −2
Solution:

44
 x−1 −3 0
x−2 −1 2 1
D( x)=|xI− A|=| 2 x−2 1 |=( x−1)| |+2| |
0 x+2 −4 x+2
−4 0 x+2

=( x−1) [ ( x−2 )( x+2 ) ] +3 [ 2( x +2 )+4 ]

=( x−1)( x 2 −4)+6 x+12+12=x 3 −4 x−x 2 +4+6 x+24

=x 3 −x 2 +2 x+ 28

CAYLEY-HAMILTON THEOREM
Every matrix is a zero of its characteristic polynomial.

Example: Verify the Cayley-Hamilton theorem given that

A= (13 22 )
Solution:

x−1 −2
|xI− A|=| |=(x−1)( x−2)−6=x2 −3 x−4
−3 x−2

Thus f ( x)=x 2 −3 x−4 is the characteristic polynomial of A. Now,

( ) ( ) ( )
2
f ( A )=A 2 −3 A−4 I= 1 2 −3 1 2 −4 1 0
3 2 3 2 0 1

45
 ( )( )( )( )
=7 6 −3 6 −4 0 =0 0
9 10 9 6 0 4 0 0 Thus A is a zero of its characteristic
polynomial.

Theorem: Similar matrices have the same characteristic polynomial.

Proof: Let A and B be similar matrices. Then there exists an invertible matrix P such
−1
that B=P AP . So

|xI−B|=|xI−P−1 AP|=|xP−1 P−P−1 AP|

=|P−1( xI− A)P|
=|P−1||xI−A||P|
1
=|P−1||P||xI− A|= |P||xI− A|
|p|

=|xI−A|

Theorem: Let A be a square matrix over a field K . A scalar λ ∈ K is an eigenvalue of

A if and only if λ is a root of the characteristic polynomial of A.
Theorem:

Suppose
M=
( A1
0
B
A2 )
where A1 and A2 are square matrices. The characteristic
polynomial of M is the product of the characteristic polynomial of A1 and A2 .

Example: Find the characteristic polynomial of

( )
2 1 0 0
0 2 0 0
A=
0 0 1 1
0 0 −2 4
Solution:

46
 x−2 −1 0 0
0 x−2 0 0 x−2 −1 x−1 −1
D( x)=|xI− A|=| |=| || |
0 0 x−1 −1 0 x−2 2 x−4
0 0 2 x−4
( x−2 )2 [ ( x−1)( x−4 )+2 ]=(x−2)2 ( x 2 −5 x+6)

¿( x−2 )2 ( x−2)( x−3 )

( x−2 )3 ( x−3 )
The eigenvalues of A are λ=2 , 2, 2 and λ=3 . We can extend the above theorem to
triangular block matrices. If

( )
A1 B ⋯ C
M= 0 A2 ⋯ D
⋮ ⋮ ⋱ ⋮
0 0 0 An

where
Ai are square matrices, then the characteristic polynomial of M is the product

of characteristic polynomials of the

Ai .

Example: Find the characteristic polynomial of

( )
2 5 0 0 0
0 2 0 0 0
A= 0 0 4 2 0
0 0 3 5 0
0 0 0 0 7
Solution:

x−2 −5 0 0 0
0 x−2 0 0 0
|xI− A|=| 0 x−2 −5 || x−4 −2 |( x−7)
0 x−4 −2 0 |=|
0 x−2 −3 x−5
0 0 −3 x−5 0
0 0 0 0 x−7

47
 =( x−2)2 [ ( x−4 )( x−5 )−6 ] (x−7)

¿( x−2)2 ( x 2 −9 x+14 )(x−7)=( x−2 )3 ( x−7 )2

Exercise: Find the characteristic polynomial of each of the following matrices:

( )
3 1 0 0 0
1 4 0 0 0

( ) ( )
3 −5 5 1 1 0 C= 0 0 3 1 0
A= 5 −7 5 B= 0 2 0 0 0 1 3 0
5 −5 3 0 0 1 0 0 0 0 1

MINIMAL POLYNOMIAL

Definition: Let
f (a )=a n x n +a n−1 x n−1 +an−2 x n−2 +⋯+a2 x 2 +a 1 x+a0 be a polynomial of

degree n . Then
a n is called the leading coefficient and f ( x ) is monic if a n=1.

Example: x 3 +x 2 +1 , x 2 +1 and x 4 +2 x3 +3 x+1 are all monic polynomials.

Definition: Let A be a square matrix. Then the minimal polynomial of A is a monic

polynomial M ( x ) of lowest degree such that M ( A )=0 .
Theorem: The minimal polynomial of a matrix divides the characteristic
polynomial.
Theorem: The minimal polynomial and the characteristic polynomial have the same
irreducible factors (factors that cannot be factorized further)
eg:

48
 If D( x)=( x−1)2 ( x−2)( x+3)

M ( x)=( x−1)( x−2)( x+3)

or
M ( x)=( x−1)2 (x−2)( x+3) are the min imal polynomials.
Example 1: Find the characteristic polynomial, minimal polynomial and
eigenvalues of the matrix:

( )
3 −5 5
A= 5 −7 5
5 −5 3
Solution:

x−5 5 −5
|xI− A|=| −5 x+7 −5 |=( x−3)|x+7 −5 |−5|−5 −5 |−5|−5 x+7 |
5 x−3 −5 x−3 −5 5
−5 5 x−3
=( x−3 ) [ ( x +7 )( x−3 )+ 25 ] −5 [ −5( x−3 )−25 ] −5 [−25+5( x+ 7) ]

=( x−3)(x+2)2

Therefore, the characteristic polynomial of A is D( x)=( x−3)( x+2)2 .

The eigenvalues of A are λ=−2 ,−2 and λ=3 .
The possibilities of minimal polynomial are:

M 1 ( x )=( x−3 )( x+2)

2
M 2 ( x )=( x−3 )( x+2)

( )( )( )
0 −5 5 5 −5 5 0 0 0
M 1 ( A )=( A−3 I )( A+2 I )= 5 −10 5 5 −5 5 = 0 0 0 =0
5 −5 0 5 −5 5 0 0 0
M 2 ( A )=0 by Cayley-Hamilton theorem. ⇒ M 1 ( x )=( x −3)( x +2) is the minimal
polynomial.
Example 2: Find the characteristic polynomial, minimal polynomial and the
eigenvalues of

49
 ( )
2 1 0 0
0 2 0 0
A=
0 0 2 0
0 0 0 5
Solution:

x−2 −1 0 0
0 x−2 0 0 x−2 −1 x−2 0
|xI− A|=| |=| || |
0 0 x−2 0 0 x−2 0 x−5
0 0 0 x−5
=( x−2)2 ( x−2)( x−5)=( x−2)3 ( x−5)

Therefore the characteristic polynomial of A is f ( x)=(x−2)3 ( x−5) .

The eigenvalues of A are λ=2 , 2, 2 and λ=5 .
Possibilities of minimal polynomial are

M 1 ( x )=( x−2 )(x −5)

M 2 ( x )=( x−2 )2 ( x−5 )
M 3 ( x )=( x−2 )3 ( x−5 )

( )( )( )
0 1 0 0 −3 1 0 0 0 −3 0 0
0 0 0 0 0 −3 0 0 0 0 0 0
M 1 ( A )=( A−2 I )( A−5 I )= = ≠0
0 0 0 0 0 0 −3 0 0 0 0 0
0 0 0 3 0 0 0 0 0 0 0 0

Hence M 1 ( x ) is not a minimal polynomial of A.

( )( ) ( )( )
0 1 0 0 2 −3 1 0 0 0 0 0 0 −3 1 0 0
0 0 0 0 0 −3 0 0 0 0 0 0 0 −3 0 0
M 2 ( A )=( A−2 I )2 ( A−5 I )= =
0 0 0 0 0 0 −3 0 0 0 0 0 0 0 −3 0
0 0 0 3 0 0 0 0 0 0 0 9 0 0 0 0

50
 ( )
0 0 0 0
0 0 0 0
= =0
0 0 0 0
0 0 0 0

M 3( A )=0 by Cayley-Hamilton theorem. Hence M 2 ( x)=( x−2 )2 ( x−5 ) is the minimal

polynomial of A.

Exercise:
1. For each of the following matrix find the polynomial having the matrix as a
root,

( )
1 4 −3
C= 0 3 1
i.
(
A= 2 5
1 −3 ) ii.
(
B= 2 −3
7 −4
iii.
) 0 2 −1
2. Find the characteristic polynomial, minimal polynomial and eigenvalues of
each matrix below;

( )
2 1 0 0

( ) ( )
1 1 0 0 2 0 0 1 2 2
B=
A= 0 2 0 0 0 1 1 C= 1 2 −1
i.
0 0 1 ii.
0 0 −2 4 iii.
−1 1 4

MINIMAL POLYNOMIAL THEOREMS

Theorem: The minimal polynomial M ( x ) of a matrix A divides every polynomial

which has A as a zero. In particular, M ( x ) divides the characteristic polynomial of A.

Proof: Suppose f ( x ) is a polynomial for which f ( A )=0 . By division algorithm there

exists polynomials q ( x ) and r ( x ) for which f ( x )=M ( x )q( x )+ r ( x ) and r ( x )=0 or
deg r ( x )<deg M ( x ). Substituting x= A in this equation and using the fact that
f ( A )=0 and M ( A )=0 , we obtain r ( A )=0 . If r ( A )≠0 , then r ( x ) is a polynomial of
degree less than M ( x ) which has A as a zero; this contradicts the definition of a
minimal polynomial. Thus r ( x )=0 and so f ( x )=M ( x )q( x ) i.e M ( x ) divides f ( x ).

51
Theorem: Let M ( x ) be the minimal polynomial of an n-square matrix A. Then the
characteristic polynomial of A divides M ( x ) .

Theorem: The characteristic polynomial D( x ) and the minimal polynomial M ( x ) of

a matrix A have the same irreducible polynomials.

Proof: Suppose f ( x ) is an irreducible polynomial. If f ( x ) divides M ( x ) then since

M ( x ) divides D( x ), f ( x ) divides D( x ) . On the other hand if f ( x ) divides D( x ) then

by the preceding theorem, f ( x ) divides ( M ( x))n . But f ( x ) is irreducible, hence f ( x )

also divides M ( x ) . Thus M ( x ) and D( x ) have the same irreducible factors.

TRACE

A=( a ij )
The trace of a square matrix written tr ( A ) is defined to be the sum of its

diagonal elements i.e

tr ( A )=a11+a22 +a33 +⋯+a nn
Theorem: Similar matrices have the same trace.

Thus the trace of a linear operator T is the trace of any one of its matrix
representations i.e tr ( T )=tr ( [ T ] e ) .

Example: Let V be the space of 2×2 matrices over

R
, and let
( )
M= 1 2 .
3 4 Let T be
the linear operator on V defined by T ( A )=MA . Find the trace of T.

Solution: We must find a matrix of representation of T. Choose the usual basis of V

{ ( ) ( ) ( ) ( )}
E1 =
1
0
0
0
, E2 =
0 1
0 0
, E3 =
0 0
1 0
, E 4=
0 0
0 1 , then

( )( ) ( )
T ( E1 )= 1 2 1 0 = 1 0 =1 E 1 +0 E2 +3 E 3 +0 E 4
3 4 0 0 3 0

T ( E )=( 1 2 )( 0 1 )=( 0 1 )=0 E +1 E +0 E +3 E

2 1 2 3 4
3 4 0 0 0 3

52
 ( )( ) ( )
T ( E3 )= 1 2 0 0 = 2 0 =2 E 1 +0 E2 +4 E 3 +0 E 4
3 4 1 0 4 0

T ( E )=(
3 4 )(0 4 ) (0 4 )
1 2 0 0 0 2
4 = =0 E +2 E +0 E +4 E
1 2 3 4

Hence

( )
1 0 2 0
[ T ] E= 0 1 0 2
3 0 4 0
0 3 0 4 and tr ( T )=1+1+4 +4=10

53