Lecture 6

Linear Transformations

6.1 Introduction to Linear Transformations

6.2 The Kernel and Range of a Linear Transformation
6.3 Matrices for Linear Transformations
6.4 Transition Matrices and Similarity

6.1 Introduction to Linear Transformations

n Function T that maps a vector space V into a vector space W:

T : V ¾mapping
¾ ¾®W , V , W : vector space

V: the domain of T
W: the codomain of T

1
 n Image of v under T:

If v is in V and w is in W such that

T ( v) = w
Then w is called the image of v under T .

n the range of T:
The set of all images of vectors in V.

n the preimage of w:
The set of all v in V such that T(v)=w.

n Ex 1: (A function from R2 into R2 )

T : R 2 ® R 2 v = (v1 , v2 ) Î R 2
T (v1 , v2 ) = (v1 - v2 , v1 + 2v2 )
(a) Find the image of v=(-1,2). (b) Find the preimage of w=(-1,11)
Sol:
(a ) v = (-1, 2)
Þ T ( v ) = T (-1, 2) = (-1 - 2, - 1 + 2(2)) = (-3, 3)
(b) T ( v ) = w = (-1, 11)
T (v1 , v2 ) = (v1 - v2 , v1 + 2v2 ) = (-1, 11)
Þ v1 - v2 = -1
v1 + 2v2 = 11
Þ v1 = 3, v2 = 4 Thus {(3, 4)} is the preimage of w=(-1, 11).

3
 n Linear Transformation (L.T.):
V, W – vector spaces
T : V → W, a mapping from V to W, is linear transformation if it
satisfies
(1) T (u + v ) = T (u) + T ( v ), "u, v Î V
(2) T (cu) = cT (u), ∀c ∈R

n Notes:
(1) A linear transformation is said to be operation preserving.

T (u + v ) = T (u) + T ( v ) T (cu) = cT (u)

Addition Addition Scalar Scalar

in V in W multiplication multiplication
in V in W

(2) A linear transformation T : V ® V from a vector space into

itself is called a linear operator.

5
 n Ex 2: (Verifying a linear transformation T from R2 into R2)
T (v1 , v2 ) = (v1 - v2 , v1 + 2v2 )
Pf:
u = (u1 , u2 ), v = (v1 , v2 ) : vector in R 2 , c : any real number
(1)Vector addition :
u + v = (u1 , u2 ) + (v1 , v2 ) = (u1 + v1 , u2 + v2 )

T (u + v ) = T (u1 + v1 , u2 + v2 )
= ((u1 + v1 ) - (u2 + v2 ), (u1 + v1 ) + 2(u2 + v2 ))
= ((u1 - u2 ) + (v1 - v2 ), (u1 + 2u2 ) + (v1 + 2v2 ))
= (u1 - u2 , u1 + 2u2 ) + (v1 - v2 , v1 + 2v2 )
= T (u) + T ( v )

(2) Scalar multiplication

cu = c(u1 , u2 ) = (cu1 , cu2 )
T (cu) = T (cu1 , cu 2 ) = (cu1 - cu 2 , cu1 + 2cu 2 )
= c(u1 - u 2 , u1 + 2u 2 )
= cT (u)
Therefore, T is a linear transformation.

7
 n Ex 3: (Functions that are not linear transformations)
(a ) f ( x) = sin x
sin( x1 + x2 ) ¹ sin( x1 ) + sin( x2 )
sin( p2 + p3 ) ¹ sin( p2 ) + sin( p3 )

(b) f ( x) = x 2
( x1 + x2 ) 2 ¹ x12 + x22
(1 + 2) 2 ¹ 12 + 2 2

(c ) f ( x ) = x + 1
f ( x1 + x2 ) = x1 + x2 + 1
f ( x1 ) + f ( x2 ) = ( x1 + 1) + ( x2 + 1) = x1 + x2 + 2
f ( x1 + x2 ) ¹ f ( x1 ) + f ( x2 )

n Notes: Two uses of the term “linear”.

(1) f ( x) = x + 1 is called a linear function because its graph

is a line.

(2) f ( x) = x + 1 is not a linear transformation from a vector

space R into R because it preserves neither vector
addition nor scalar multiplication.

9
 n Zero transformation:
T :V ® W T ( v ) = 0, "v Î V
n Identity transformation:
T :V ® V T ( v ) = v, "v Î V
n Thm 6.1: (Properties of linear transformations)
T : V ® W , u, v Î V
(1) T (0) = 0
(2) T (- v ) = -T ( v )
(3) T (u - v ) = T (u) - T ( v )
(4) If v = c1v1 + c2 v2 + ! + cn vn
Then T ( v ) = T (c1v1 + c2 v2 + ! + cn vn )
= c1T (v1 ) + c2T (v2 ) + ! + cnT (vn )

10

n Ex 4: (Linear transformations and bases)

Let T : R 3 ® R 3 be a linear transformation such that
T (1,0,0) = (2,-1,4) Recall:
T (0,1,0) = (1,5,-2) e1=(1,0,0); e2=(0,1,0); e3=(0,0,1)
T (0,0,1) = (0,3,1) is the standard basis of R3

Find T(2, 3, -2).

Sol:
(2,3,-2) = 2(1,0,0) + 3(0,1,0) - 2(0,0,1)
T (2,3,-2) = 2T (1,0,0) + 3T (0,1,0) - 2T (0,0,1)
= 2(2,-1,4) + 3(1,5,-2) - 2T (0,3,1)
= (7,7,0)

11
 n Ex 5: (A linear transformation defined by a matrix)
é3 0ù
ê év ù
The function T : R ® R is defined as T ( v ) = Av = 2
2 3 1 úê 1 ú
ê ú ëv2 û
(a) Find T ( v ) , where v = (2,-1) ë - 1 - 2û
(b) Show that T is a linear transformation form R 2 into R 3

Sol: (a ) v = (2,-1) R 2 vector R 3 vector

é3 0ù é6 ù
ê ú é2ù ê ú
T ( v ) = Av = 2 1 ê ú= 3
ê ú -1 ê ú
ë - 1 - 2 û ë û ë0 û
\T (2,-1) = (6,3,0)
(b) T (u + v ) = A(u + v ) = Au + Av = T (u) + T ( v ) (vector addition)

T (cu) = A(cu) = c( Au) = cT (u) (scalar multiplication)

12

n Thm 6.2: (The linear transformation given by a matrix)

Let A be an mÍn matrix. The function T defined by
T ( v ) = Av

n Note: R n vector R m vector

é a11 a12 ! a1n ù é v1 ù é a11v1 + a12 v2 + ! + a1n vn ù
êa a22 ! a2 n ú êv2 ú ê a21v1 + a22 v2 + ! + a2 n vn ú
Av = ê 21 úê ú = ê ú
ê " " " úê " ú ê " ú
êëam1 am 2 ! amn úû êëvn úû êëam1v1 + am 2 v2 + ! + amn vn úû

T ( v ) = Av

13
 n Ex 7: (Rotation in the plane)
Show that the L.T. T : R 2 ® R 2 given by the matrix

écos q - sin q ù
A=ê
ë sin q cos q úû
has the property that it rotates every vector in R2
counterclockwise about the origin through the angle θ.
Sol:
v = ( x, y ) = (r cos a , r sin a ) (polar coordinates)
r： the length of v
α：the angle from the positive
x-axis counterclockwise to
the vector v

14

⎡ cosθ −sin θ ⎤ ⎡ x ⎤ ⎡ cosθ −sin θ ⎤ ⎡ r cos α ⎤

T (v) = Av = ⎢ ⎥⎢ ⎥=⎢ ⎥⎢ ⎥
⎣ sin θ cosθ ⎦ ⎢⎣ y ⎥⎦ ⎣ sin θ cosθ ⎦ ⎣ r sin α ⎦
⎡ r cosθ cos α − r sin θ sin α ⎤
=⎢ ⎥
⎣ r sin θ cos α + r cosθ sin α ⎦
⎡ r cos(θ + α ) ⎤
=⎢ ⎥
⎢⎣ r sin(θ + α ) ⎥⎦
r : the length of T(v)
θ +α：the angle from the positive x-axis counterclockwise to
the vector T(v)
Thus, T(v) is the vector that results from rotating the vector v
counterclockwise through the angle θ.

15
 n Ex 8: (A projection in R3)
The linear transformation T : R 3 ® R 3 is given by
é1 0 0ù
A = ê0 1 0 ú
ê ú
ë0 0 0 û
is called a projection in R3.

16

n Ex 9: (A linear transformation from Mmxn into Mn xm )

T ( A) = AT (T : M m´n ® M n´m )
Show that T is a linear transformation.
Sol:
A, B Î M m´n
T ( A + B) = ( A + B)T = AT + B T = T ( A) + T ( B)
T (cA) = (cA)T = cAT = cT ( A)

Therefore, T is a linear transformation from Mmxn into Mn xm.

17
 6.2 The Kernel and Range of a Linear Transformation
n Kernel of a linear transformation T:
Let T : V ® W be a linear transformation
Then the set of all vectors v in V that satisfy T ( v) = 0 is
called the kernel of T and is denoted by ker(T).
ker(T ) = {v | T ( v ) = 0, "v Î V }

n Ex 1: (Finding the kernel of a linear transformation)

T ( A) = AT (T : M 3´2 ® M 2´3 )
Sol:
ì é0 0 ù ü
ï ï
ker(T ) = íêê0 0úú ý
ï ê0 0 ú ï
îë ûþ

18

n Ex 2: (The kernel of the zero and identity transformations)

(a) T(v)=0 (the zero transformation T :V ® W )
ker(T ) = V

(b) T(v)=v (the identity transformation T : V ® V )

ker(T ) = {0}

n Ex 3: (Finding the kernel of a linear transformation)

T ( x, y, z ) = ( x, y,0) (T : R 3 ® R 3 )
ker(T ) = ?
Sol:
ker(T ) = {(0,0, z ) | z is a real number}

19
 n Ex 5: (Finding the kernel of a linear transformation)
éx ù
é 1 - 1 - 2ù ê 1 ú
T (x) = Ax = ê x (T : R 3 ® R 2 )
ë- 1 2 3 úû ê 2 ú
ë x3 û
ker(T ) = ?
Sol:
ker(T ) = {( x1 , x2 , x3 ) | T ( x1 , x2 , x3 ) = (0,0), x = ( x1 , x2 , x3 ) Î R 3 }

T ( x1 , x2 , x3 ) = (0,0)
éx ù
é 1 - 1 - 2 ù ê 1 ú é0 ù
êë- 1 2 x =
3 úû ê 2 ú êë0úû
ë x3 û

20

é 1 - 1 - 2 0ù G . J . E é1 0 - 1 0ù
êë- 1 2 ¾¾¾®ê
3 0úû ë0 1 1 0úû

é x1 ù é t ù é 1 ù
Þ ê x2 ú = ê- t ú = t ê- 1ú
ê ú ê ú ê ú
ë x3 û ë t û ë 1 û

Þ ker(T ) = {t (1,-1,1) | t is a real number}

= span{(1,-1,1)}

21
 n Thm 6.3: (The kernel is a subspace of V)
The kernel of a linear transformation T : V ® W is a
subspace of the domain V.
Pf: ∵T (0) = 0 (Theorem 4.1)
\ ker(T ) is a nonempty subset of V
Let u and v be vectors in the kernel of T . then
T (u + v ) = T (u) + T ( v ) = 0 + 0 = 0 Þ u + v Î ker(T )

T (cu) = cT (u) = c0 = 0 Þ cu Î ker(T )

Thus, ker(T ) is a subspace of V .
n Note:
The kernel of T is sometimes called the nullspace of T.

22

n Ex 6: (Finding a basis for the kernel)

Let T : R 5 ® R 4 be defined by T (x) = Ax, where x is in R 5 and
é1 2 1 - 1ù
0
ê2 1 3 1 0 úú
A=ê
ê- 1 0 -2 0 1 ú
ê ú
ë0 0 0 2 8û

Find a basis for ker(T) as a subspace of R5.

23
 Sol:
[A 0] =
é1 2 0 1 -1 0ù é1 0 2 0 -1 0ù
ê2 1 3 1 0 0 ú G . J . E ê0 1 -1 0 -2 0ú
ê ú ¾¾¾®ê ú
ê- 1 0 -2 0 1 0ú ê0 0 0 1 4 0ú
ëê 0 0 0 2 8 0ûú ëê0 0 0 0 0 0ûú
s t
é x1 ù é- 2 s + t ù é - 2ù é 1 ù
ê x2 ú ê s + 2t ú ê1ú ê2ú
ê ú ê
x = ê x3 ú = ê s ú = s êê 1 úú + t êê 0 úú
ú
ê x4 ú ê - 4t ú ê 0 ú ê - 4ú
êë x5 úû êë t úû êë 0 úû êë 1 úû

B = {(-2, 1, 1, 0, 0), (1, 2, 0, - 4, 1)}: one basis for the kernel of T

24

n Corollary to Thm 6.3:

Let T : R n ® R m be the L.T given by T (x) = Ax
Then the kernel of T is equal to the solution space of Ax = 0
T (x) = Ax (a linear transformation T : R n ® R m )
Þ Ker (T ) = NS ( A) = {x | Ax = 0, "x Î R m } (subspace of R m )

n Range of a linear transformation T:

Or, equivalently range(T ) = {T ( v ) | "v Î V }

25
 n Thm 6.4: (The range of T is a subspace of W)
The range of a linear transformation T : V ® W is a subspace of W .
Pf:
! T (0) = 0 (Thm.6.1)
\ range(T ) is a nonempty subset of W
Let T (u) and T ( v ) be vector in the range of T

T (u + v ) = T (u) + T ( v ) Î range(T ) ( u Î V , v Î V Þ u + v Î V )
T (cu) = cT (u) Î range(T ) (u Î V Þ c u Î V )

Therefore, range(T ) is W subspace.

26

n Notes:
T : V ® W is a L.T.
(1) Ker (T ) is subspace of V
(2)range(T ) is subspace of W

n Corollary to Thm 6.4:

Let T : R n ® R m be the L.T. given by T (x) = Ax
Then the range of T is equal to the column space of A
Þ range(T ) = CS ( A)

27
 n Ex 7: (Finding a basis for the range of a linear transformation)

Let T : R 5 ® R 4 be defined by T (x) = Ax, where x is R 5 and

é1 2 01 - 1ù
ê2 1 3 1 0 úú
A=ê
ê- 1 0 -2 0 1 ú
ê ú
ë0 0 0 2 8û

Find a basis for the range of T.

28

Sol:
é1 2 0 1 - 1ù é1 0 2 0 - 1ù
ê2 1 3 1 0 ú G . J . E ê0 1 -1 0 2ú
A=ê ú ¾¾¾®ê ú=B
ê- 1 0 -2 0 1ú ê0 0 0 1 4ú
êë 0 0 0 2 8 úû êë0 0 0 0 0 úû
c1 c 2 c3 c 4 c5 w1 w2 w3 w 4 w5
Þ {w1 , w2 , w4 }is a basis for CS ( B)
{c , c , c }is a basis for CS ( A)
1 2 4

Þ {(1, 2, - 1, 0), (2, 1, 0, 0), (1, 1, 0, 2)}is a basis for the range of T

29
 n Rank of a linear transformation T:V→W:
rank (T ) = the dimension of the range of T

n Nullity of a linear transformation T:V→W:

nullity (T ) = the dimension of the kernel of T

n Note:
Let T : R n ® R m be the L.T. given by T (x) = Ax, then
rank (T ) = rank ( A)
nullity (T ) = nullity ( A)

30

nThm 6.5: (Sum of rank and nullity)

Let T : V ® W be a L.T. form an n - dimensional vector space V
into a vector space W . Then
rank (T ) + nullity (T ) = n
dim(range of T ) + dim(kernel of T ) = dim(domain of T )
Pf:
Let T is represented by an m ´ n matrix A
Assume rank ( A) = r
(1)rank (T ) = dim(range of T ) = dim(column space of A)
= rank ( A) = r
(2)nullity (T ) = dim(kernel of T ) = dim(solution space of A)
= n-r
Þ rank (T ) + nullity (T ) = r + (n - r ) = n

31
 n Ex 8: (Finding the rank and nullity of a linear transformation)
Find the rank and nullity of the L.T. T : R 3 ® R 3 define by
é1 0 - 2ù
A = êê0 1 1 úú
êë0 0 0 úû
Sol:
rank (T ) = rank ( A) = 2
nullity (T ) = dim(domain of T ) - rank (T ) = 3 - 2 = 1

32

n Ex 9: (Finding the rank and nullity of a linear transformation)

Let T : R 5 ® R 7 be a linear transformation.
(a ) Find the dimension of the kernel of T if the dimension
of the range is 2
(b) Find the rank of T if the nullity of T is 4
(c) Find the rank of T if Ker (T ) = {0}
Sol:
(a ) dim(domain of T ) = 5
dim(kernel of T ) = n - dim(range of T ) = 5 - 2 = 3
(b)rank (T ) = n - nullity (T ) = 5 - 4 = 1

(c)rank (T ) = n - nullity (T ) = 5 - 0 = 5

33
 n One-to-one:
A function T : V ® W is called one - to - one if the preimage of
every w in the range consists of a single vector.
In other words,
T is one - to - one iff for all u and v inV, T (u) = T ( v )
implies that u = v.

34

n Onto:
A function T : V ® W is said to be onto if every element
in w has a preimage in V
(T is onto W when W is equal to the range of T.)

35
 n Thm 6.6: (One-to-one linear transformation)
Let T : V ® W be a L.T.
Then T is 1 - 1 iff Ker (T ) = {0}
Pf:
Suppose T is 1 - 1
Then T (v) = 0 can have only one solution : v = 0
i.e. Ker (T ) = {0}
Suppose Ker (T ) = {0} and T (u ) = T (v)
T (u - v) = T (u ) - T (v) = 0
T is a L.T.
! u - v Î Ker (T ) Þ u - v = 0
Þ T is 1 - 1

36

n Ex 10: (One-to-one and not one-to-one linear transformation)

(a ) The L.T. T : M m´n ® M n´m given by T ( A) = AT
is one - to - one.
Because its kernel consists of only the m ´ n zero matrix.

(b) The zero transformation T : R 3 ® R 3 is not one - to - one.

Because its kernel is all of R 3 .

37
 n Thm 6.7: (Onto linear transformation)
Let T : V ® W be a L.T., where W is finite dimensional.
Then T is onto iff the rank of T is equal to the dimension of W .
n Thm 6.8: (One-to-one and onto linear transformation)
Let T : V ® W be a L.T. with vector space V and W both of
dimension n. Then T is one - to - one if and only if it is onto.
Pf:
If T is one - to - one, then Ker (T ) = {0} and dim( Ker (T )) = 0
dim(range(T )) = n - dim( Ker (T )) = n = dim(W )
Consequently, T is onto.
If T is onto, then dim(range of T ) = dim(W ) = n
dim( Ker (T )) = n - dim(range of T ) = n - n = 0
Therefore, T is one - to - one.

38

nEx 11:
The L.T. T : R n ® R m is given by T (x) = Ax, Find the nullity and rank
of T and determine whether T is one - to - one, onto, or neither.
é1 2 0ù é1 2ù
( a ) A = ê0 1 1 ú (b) A = ê0 1 ú
ê ú ê ú
ë0 0 1 û ë0 0 û
é1 2 0ù
é1 2 0 ù
(c ) A = ê ( d ) A = ê0 1 1ú
ë0 1 - 1úû ê ú
ë0 0 0û
Sol:
T:Rn→Rm dim(domain of T) rank(T) nullity(T) 1-1 onto

(a)T:R3→R3 3 3 0 Yes Yes

(b)T:R2→R3 2 2 0 Yes No
(c)T:R3→R2 3 2 1 No Yes
(d)T:R3→R3 3 2 1 No No

39
 n Isomorphism:
A linear transformation T : V ® W that is one to one and onto
is called an isomorphism. Moreover, if V and W are vector spaces
such that there exists an isomorphism from V to W , then V and W
are said to be isomorphic to each other.
n Thm 6.9: (Isomorphic spaces and dimension)
Two finite-dimensional vector space V and W are isomorphic
if and only if they are of the same dimension.
Pf:
Assume that V is isomorphic to W , where V has dimension n.
Þ There exists a L.T. T : V ® W that is one to one and onto.
! T is one - to - one
Þ dim( Ker (T )) = 0
Þ dim(range of T ) = dim(domain of T ) - dim( Ker (T )) = n - 0 = n

40

! T is onto.
Þ dim(range of T ) = dim(W ) = n
Thus dim(V ) = dim(W ) = n
Assume that V and W both have dimension n.
Let {v1 , v2 , ! , vn } be a basis of V, and
let {w1 , w2 , ! , wn } be a basis of W .
Then an arbitrary vector in V can be represented as
v = c1v1 + c2 v2 + ! + cn vn
and you can define a L.T. T : V ® W as follows.
T ( v ) = c1w1 + c2 w2 + ! + cn wn
It can be shown that this L.T. is both 1-1 and onto.
Thus V and W are isomorphic.

41
42

n Ex 12: (Isomorphic vector spaces)

The following vector spaces are isomorphic to each other.
(a ) R 4 = 4 - space
(b) M 4´1 = space of all 4 ´1 matrices
(c) M 2´2 = space of all 2 ´ 2 matrices
(d ) P3 ( x) = space of all polynomials of degree 3 or less
(e)V = {( x1 , x2 , x3 , x4 , 0), xi is a real number}(subspace of R 5 )

43
 6.3 Matrices for Linear Transformations
n Two representations of the linear transformation T:R3→R3 :

(1)T ( x1 , x2 , x3 ) = (2 x1 + x2 - x3 ,- x1 + 3 x2 - 2 x3 ,3 x2 + 4 x3 )

é 2 1 - 1 ù é x1 ù
(2)T (x) = Ax = ê- 1 3 - 2ú ê x2 ú
ê úê ú
ë 0 3 4 û ë x3 û
n Three reasons for matrix representation of a linear transformation:
n It is simpler to write.
n It is simpler to read.
n It is more easily adapted for computer use.

44

n Thm 6.10: (Standard matrix for a linear transformation)

Let T : R n → R m be a linear transformation such that
é a11 ù é a12 ù é a1n ù
ê a21 ú ê a22 ú êa ú
T (e1 ) = ê ú, T (e2 ) = ê ú, " , T (en ) = ê 2 n ú,
ê ! ú ê ! ú ê ! ú
ëê m1 ûú
a ëê m 2 ûú
a ëêamn ûú
Where {e1, e2, …,en} is the standard basis of Rn.
Then the m ´ n matrix whose n columns correspond to T (ei )
é a11 a12 ! a1n ù
êa a22 ! a2 n ú
A = [T (e1 ) T (e2 ) ! T (en )] = ê 21 ú
ê " " # " ú
ëêam1 am 2 ! amn ûú
is such that T ( v ) = Av for every v in R n .
A is called the standard matrix for T .

45
 Pf:
é v1 ù
êv ú
v = ê 2 ú = v1e1 + v2 e2 + ! + vn en
ê"ú
ëêvn ûú
T is a L.T. Þ T (v ) = T (v1e1 + v2 e2 + ! + vn en )
= T (v1e1 ) + T (v2 e2 ) + ! + T (vn en )
= v1T (e1 ) + v2T (e2 ) + ! + vnT (en )

é a11 a12 ! a1n ù é v1 ù é a11v1 + a12 v2 + ! + a1n vn ù

êa a22 ! a2 n ú êv2 ú ê a21v1 + a22 v2 + ! + a2 n vn ú
Av = ê 21 úê ú = ê ú
ê " " # " úê " ú ê " ú
ëêam1 am 2 ! amn ûú ëêvn ûú ëêam1v1 + am 2 v2 + ! + amn vn ûú

46

é a11 ù é a12 ù é a1n ù

ê a21 ú ê a22 ú êa ú
= v1 ê ú + v2 ê ú + ! + vn ê 2 n ú
ê " ú ê " ú ê " ú
êëam1 úû êëam 2 úû êëamn úû
= v1T (e1 ) + v2T (e2 ) + ! + vnT (en )

Therefore, T ( v ) = Av for each v in R n

47
 n Ex 1: (Finding the standard matrix of a linear transformation)
Find the standard matrix for the L.T. T : R 3 ® R 2 define by
T ( x, y , z ) = ( x - 2 y , 2 x + y )
Sol:
Vector Notation Matrix Notation
é1ù
é1 ù
T (e1 ) = T (1, 0, 0) = (1, 2) T (e1 ) = T ( ê0ú ) = ê ú
ê ú ë 2û
ë0 û
é0 ù
é - 2ù
T (e2 ) = T (0, 1, 0) = (-2, 1) T (e2 ) = T ( ê1ú ) = ê ú
ê ú ë1û
ë0 û
é0 ù
é0 ù
T (e3 ) = T (0, 0, 1) = (0, 0) T (e3 ) = T ( ê0ú ) = ê ú
ê ú ë0 û
ë1û

48

A = [T (e1 ) T (e2 ) T (e3 )]

é1 - 2 0 ù
=ê
ë2 1 0úû
n Check:
é xù é xù
ê ú é1 - 2 0 ù ê ú é x - 2 y ù
A y =ê y =
ê ú ë2 1 0úû ê ú êë2 x + y úû
z
ë û ëzû
i.e. T ( x, y, z ) = ( x - 2 y,2 x + y )
n Note:
é1 - 2 0ù ¬ 1x - 2 y + 0 z
A=ê
ë2 1 0úû ¬ 2 x + 1 y + 0 z

49
 n Ex 2: (Finding the standard matrix of a linear transformation)
The linear transformation T : R 2 ® R 2 is given by projecting
each point in R 2 onto the x - axis. Find the standard matrix for T .
Sol:
T ( x, y ) = ( x, 0)
é1 0ù
A = [T (e1 ) T (e2 )] = [T (1, 0) T (0, 1)] = ê
ë0 0úû
n Notes:

(1) The standard matrix for the zero transformation from Rn into Rm
is the mxn zero matrix.
(2) The standard matrix for the identity transformation from Rn into
Rn is the nxn identity matrix In

50

51
 n Composition of T1:Rn→Rm with T2:Rm→Rp :

T ( v ) = T2 (T1 ( v )), v Î R n

T = T2 ! T1 , domain of T = domain of T1

n Thm 6.11: (Composition of linear transformations)

Let T1 : R n ® R m and T2 : R m ® R p be L.T.
with standard matrices A1 and A2 , then

(1)The composition T : R n ® R p , defined by T ( v ) = T2 (T1 ( v )), is a L.T.

(2) The standard matrix A for T is given by the matrix product A = A2 A1

52

Pf:
(1)( T is a L.T.)
Let u and v be vectors in R n and let c be any scalar then
T (u + v ) = T2 (T1 (u + v )) = T2 (T1 (u) + T1 ( v ))
= T2 (T1 (u)) + T2 (T1 ( v )) = T (u) + T ( v )
T (cv ) = T2 (T1 (cv )) = T2 (cT1 ( v )) = cT2 (T1 ( v )) = cT ( v )

(2)( A2 A1 is the standard matrix for T )

T ( v ) = T2 (T1 ( v )) = T2 ( A1 v ) = A2 A1 v = ( A2 A1 ) v
n Note:
T1 ! T2 ¹ T2 ! T1

53
 n Ex 3: (The standard matrix of a composition)
Let T1 and T2 be L.T. from R 3 into R 3 s.t.
T1 ( x, y, z ) = (2 x + y, 0, x + z )
T2 ( x, y, z ) = ( x - y, z, y )
Find the standard matrices for the compositions
T = T2 ! T1 and T ' = T1 ! T2 ,
Sol:
é2 1 0ù
A1 = êê0 0 0úú (standard matrix for T1 )
êë1 0 1úû
é1 - 1 0ù
A2 = êê0 0 1úú (standard matrix for T2 )
êë0 1 0úû

54

The standard matrix for T = T2 ! T1

é1 - 1 0ù é2 1 0ù é2 1 0ù
A = A2 A1 = ê0 0 1ú ê0 0 0ú = ê1 0 1ú
ê úê ú ê ú
ë0 1 0 û ë1 0 1 û ë 0 0 0 û
The standard matrix for T ' = T1 ! T2

é2 1 0ù é1 - 1 0ù é2 - 2 1ù
A' = A1 A2 = êê0 0 0úú êê0 0 1úú = êê0 0 0úú
êë1 0 1úû êë0 1 0úû êë1 0 0úû

55
 n Inverse linear transformation:

If T1 : R n ® R n and T2 : R n ® R n are L.T. s.t. for every v inR n

T2 (T1 ( v )) = v and T1 (T2 ( v )) = v

Then T2 is called the inverse of T1 and T1 is said to be invertible

n Note:
If the transformation T is invertible, then the inverse is
unique and denoted by T–1 .

56

n Thm 6.12: (Existence of an inverse transformation)

Let T : R n ® R n be a L.T. with standard matrix A,
Then the following condition are equivalent.

(1) T is invertible.
(2) T is an isomorphism.
(3) A is invertible.
n Note:
If T is invertible with standard matrix A, then the standard
matrix for T–1 is A–1 .

57
 n Ex 4: (Finding the inverse of a linear transformation)
The L.T. T：R 3 ® R 3 is defined by
T ( x1 , x2 , x3 ) = (2 x1 + 3 x2 + x3 , 3 x1 + 3 x2 + x3 , 2 x1 + 4 x2 + x3 )
Show that T is invertible, and find its inverse.
Sol:
The standard matrix for T
é2 3 1ù ¬ 2 x1 + 3 x2 + x3
A = êê3 3 1úú ¬ 3 x1 + 3 x2 + x3
êë2 4 1úû ¬ 2 x1 + 4 x2 + x3

é 2 3 1 1 0 0ù
[A I 3 ] = ê 3 3 1 0 1 0ú
ê ú
ë2 4 1 0 0 1û

58

é1 0 0 - 1 1 0ù
ê
¾¾¾® 0 1 0 - 1 0
G. J .E

ê
1 ú= I
ú
[ A-1 ]
ë0 0 1 6 - 2 - 3û
Therefore T is invertible and the standard matrix for T -1 is A-1
é- 1 1 0ù
ê
-1
A = -1 0 1ú
ê ú
ë 6 - 2 - 3û
é- 1 1 0 ù é x1 ù é - x1 + x2 ù
-1 ê
-1
T ( v) = A v = - 1 0 1 ú ê x2 ú = ê - x1 + x3 ú
ê úê ú ê ú
ë 6 - 2 - 3û ë x3 û ë6 x1 - 2 x2 - 3 x3 û
In other words,
T -1 ( x1 , x2 , x3 ) = (- x1 + x2 , - x1 + x3 , 6 x1 - 2 x2 - 3 x3 )

59
 Matrix of a LT in general case
n Transformation matrix for nonstandard bases:
Let V and W be finite - dimensional vector spaces with basis B and B',
respectively,where B = {v1, v2 ,!, vn }

If T : V ® W is a L.T. s.t.
é a11 ù é a12 ù é a1n ù
ê a21 ú ê a22 ú êa ú
[T (v1 )]B ' = ê ú, [T (v2 )]B ' = ê ú, " , [T (vn )]B ' = ê 2 n ú
ê ! ú ê ! ú ê ! ú
ëêam1 ûú êëam 2 ûú ëêamn ûú

then the m ´ n matrix whose n columns correspond to [T (vi ) ]B '

60

⎡ a11 a12 ! a1n ⎤

⎢ ⎥
⎢ a a22 ! a2n ⎥
A = ⎡ T (v1 ) T (v2 ) ! T (vn ) ⎤ = ⎢ 21
⎣ ⎦ " " # " ⎥
⎢ ⎥
⎢⎣ m1
a am2 ! amn ⎥
⎦
is such that [T ( v )]B ' = A[ v ]B for every v in V .

61
 n Ex 5: (Finding a matrix relative to nonstandard bases)
Let T：R 2 ® R 2 be a L.T. defined by
T ( x1 , x2 ) = ( x1 + x2 , 2 x1 - x2 )
Find the matrix of T relative to the basis
B = {(1, 2), (-1, 1)} and B' = {(1, 0), (0, 1)}
Sol:
T (1, 2) = (3, 0) = 3(1, 0) + 0(0, 1)
T (-1, 1) = (0, - 3) = 0(1, 0) - 3(0, 1)

[T (1, 2)]B ' = éê03ùú, [T (-1, 1)]B ' = éê-03ùú

ë û ë û
the matrix for T relative to B and B'
é3 0 ù
A = [[T (1, 2)]B ' [T (1, 2)]B ' ] = ê
ë0 - 3úû

62

n Ex 6:
For the L.T. T：R 2 ® R 2 given in Example 5, use the matrix A
to find T ( v ), where v = (2, 1)
Sol:
v = (2, 1) = 1(1, 2) - 1(-1, 1) B = {(1, 2 ), ( - 1, 1)}
é1ù
Þ [v ]B = ê ú
ë- 1û
é3 0 ù é 1 ù é3ù
Þ [T ( v )]B ' = A[v ]B = ê =
ë0 - 3úû êë- 1úû êë3úû
Þ T ( v ) = 3(1, 0) + 3(0, 1) = (3, 3) B ' = {(1, 0 ), ( 0 , 1)}

n Check:
T (2, 1) = (2 + 1, 2(2) - 1) = (3, 3)

63
 n Summary: the matrix of T relative to the bases B and B':
T :V ® W (a L.T.)
B = {v1 , v2 , ! , vn } (a basis for V )
B' = {w1 , w2 , ! , wm } (a basis for W )

Thus, the matrix of T relative to the bases B and B' is

A = [[T (v1 )]B ' , [T (v2 )]B ' , !, [T (vn )]B ' ]Î M m´n

and we have [T(v)]B’ = A[v]B for every v in V.

64

n Notes:
(1)In the special case where V = W and B = B',
the matrix A is called the matrix of T relative to the basis B
(2)T : V ® V : the identity transformation
B = {v1 , v2 , ! , vn } : a basis for V
Þ the matrix of T relative to the basis B
é1 0 ! 0ù
ê0 1 ! 0 ú
A = [[T (v1 )]B , [T (v2 )]B , ! , [T (vn )]B ] = ê ú=I
ê" " # "ú n
ê ú
ë0 0 ! 1 û

65
 6.4 Transition Matrices and Similarity

A = [[T (v1 )]B , [T (v2 )]B , !, [T (vn )]B ] ( matrix of T relative to B)

A' = [[T ( w1 )]B ' , [T ( w2 )]B ' , !, [T ( wn )]B ' ] (matrix of T relative to B' )
P = [[w1 ]B , [w2 ]B , !, [wn ]B ] ( transition matrix from B' to B )
P -1 = [[v1 ]B ' , [v2 ]B ' , !, [vn ]B ' ] ( transition matrix from B to B' )

[v]B = P[v]B' [v]B' = P-1[v]B

[T(v)]B = A[v]B [T(v)]B' = A'[v]B'

66

n Two ways to get from [v ]B ' to [T ( v ) ]B ' :

indirect
(1)(direct )
A'[ v ]B ' = [T ( v )]B '

(2)(indirect)
P -1 AP[ v ]B ' = [T ( v )]B '
direct
Þ A' = P -1 AP

67
 n Ex 1: (Finding a matrix for a linear transformation)
Find the matrix A' for T：R 2 ® R 2
T ( x1 , x2 ) = (2 x1 - 2 x2 , - x1 + 3 x2 )
reletive to the basis B' = {(1, 0), (1, 1)}
Sol:
(I) A' = [[T (1, 0)]B ' [T (1, 1)]B ' ]
é3ù
T (1, 0) = (2, - 1) = 3(1, 0) - 1(1, 1) Þ [T (1, 0)]B ' = ê ú
ë- 1û
é - 2ù
T (1, 1) = (0, 2) = -2(1, 0) + 2(1, 1) Þ [T (1, 1)]B ' = ê ú
ë2û
é 3 - 2ù
Þ A' = [[T (1, 0)]B ' [T (1, 1)]B ' ] = ê
ë- 1 2 úû

68

(II) standard matrix for T ( matrix of T relative to B = {(1, 0), (0, 1)})
é 2 - 2ù
A = [T (1, 0) T (0, 1)] = ê
ë- 1 3 úû
transition matrix from B' to B

[(1, 1)]B ] = éê
1 1ù
P = [[(1, 0)]B ú
ë0 1û
transition matrix from B to B'
é1 - 1ù
P -1 = ê ú
ë0 1 û
matrix of T relative B'
é1 - 1ù é 2 - 2ù é1 1ù é 3 - 2ù
A' = P -1 AP = ê úê úê ú=ê ú
ë0 1 û ë- 1 3 û ë0 1û ë- 1 2 û

69
 n Ex 2: (Finding a matrix for a linear transformation)
Let B = {( -3, 2), (4, - 2)} and B' = {( -1, 2), (2, - 2)} be basis for R 2 ,
é- 2 7 ù
and let A = ê ú be the matrix for T : R 2 ® R 2 relative to B.
ë - 3 7û
Find the matrix of T relative to B'.
Sol:
3 - 2ù
transition matrix from B' to B : P = [[(-1, 2)]B [(2, - 2)]B ] = éê ú
ë2 - 1 û
- 1 2ù
transition matrix from B to B': P -1 = [[(-3, 2)]B ' [(4, - 2)]B ' ] = éê ú
ë - 2 3û
matrix of T relative to B ':
é - 1 2 ù é - 2 7 ù é 3 - 2 ù é 2 1ù
A' = P -1 AP = ê úê úê ú=ê ú
ë- 2 3û ë - 3 7 û ë2 - 1 û ë- 1 3û

70

n Ex 3: (Finding a matrix for a linear transformation)

For the linear transformation T : R 2 ® R 2 given in Ex.2, find [v ]B、
[T ( v)]B and [T ( v)]B ' , for the vector v whose coordinate matrix is
- 3ù
[v ]B ' = éê ú
Sol: ë - 1û
- 2ù é- 3ù é- 7 ù
[v ]B = P[v ]B ' = éê32 =
- 1 úû êë - 1úû êë - 5úû
ë
[T ( v)]B = A[v ]B = éê-- 23 7 ù é- 7 ù é- 21ù
=
7 úû êë - 5úû êë - 14 úû
ë
[T ( v)]B ' = P -1 [T ( v)]B = éê-- 12 2ù é- 21ù é- 7 ù
=
3úû êë - 14 úû êë 0 úû
ë
é 2 1ù é- 3ù é- 7 ù
or [T ( v )]B ' = A' [v ]B ' =ê úê ú = ê ú
ë- 1 3û ë - 1û ë 0 û

71
 n Similar matrix:
For square matrices A and A' of order n, A' is said to be
similar to A if there exist an invertible matrix P s.t. A' = P -1 AP
n Thm 6.13: (Properties of similar matrices)
Let A, B, and C be square matrices of order n.
Then the following properties are true.
(1) A is similar to A.
(2) If A is similar to B, then B is similar to A.
(3) If A is similar to B and B is similar to C, then A is similar to C.

Pf: (1) A = I n AI n
(2) A = P -1 BP Þ PAP -1 = P( P -1 BP) P -1
PAP -1 = B Þ Q -1 AQ = B (Q = P -1 )

72

n Ex 4: (Similar matrices)

é 2 - 2ù é 3 - 2ù
(a) A = ê ú and A' = ê ú are similar
ë- 1 3 û ë- 1 2 û
é1 1ù
because A' = P -1 AP, where P = ê ú
ë0 1û
é- 2 7 ù é 2 1ù
(b) A = ê ú and A' = ê ú are similar
ë - 3 7û ë- 1 3û
é 3 - 2ù
because A' = P -1 AP, where P = ê ú
ë2 - 1 û

73
 n Ex 5: (A comparison of two matrices for a linear transformation)
é1 3 0 ù
Suppose A = êê3 1 0 úú is the matrix for T : R 3 ® R 3 relative
êë0 0 - 2úû
to the standard basis. Find the matrix for T relative to the basis
B' = {(1, 1, 0), (1, - 1, 0), (0, 0, 1)}
Sol:
The transition matrix from B' to the standard matrix
é1 1 0 ù
P = [[(1, 1, 0)]B [(1, - 1, 0)]B [(0, 0, 1)]B ] = êê1 - 1 0úú
êë0 0 1úû
é 12 12 0ù
Þ P -1 = ê 12 - 12 0ú
ê ú
ë0 0 1û

74

matrix of T relative to B':

é 12 12 0ù é1 3 0 ù é1 1 0ù
A' = P -1 AP = êê 12 - 12 0úú êê3 1 0 úú êê1 - 1 0úú
êë0 0 1úû êë0 0 - 2úû êë0 0 1úû
é4 0 0ù
= ê0 - 2 0 úú
ê
êë0 0 - 2úû

75
 n Notes: Computational advantages of diagonal matrices:

éd1k 0 ! 0ù éd1 0 ! 0ù
ê0 d 2k ! 0ú ê0 d2 ! 0ú
(1) D k = ê ú D=ê ú
ê" " # " ú ê" " # "ú
êë 0 0 ! d nk úû ëê 0 0 ! d n úû

(2) D T = D
é d11 0 ! 0ù
ê0 1
! 0ú
(3) D -1 = ê d2
ú, d i ¹ 0
ê" " # "ú
ê0 0 ! 1 ú
ë dn û

76