Linear Algebra - Linear Transformation

Tejas N S
Assistant Professor in Mathematics, NIE First Grade College, Mysuru
Web: http://nstejas.poeticmath.com
Email: nstejas@gmail.com Mobile: +91 9845410469

9 Linear Transformation
• Definition: Let U and V be two vector spaces over a same field F , the mapping
T : U → V is said to be linear transformation if ∀ α, β ∈ U and c ∈ F we have

1. T (α + β) = T (α) + T (β)
2. T (cα) = c T (α)

• Remarks:

(1.) T : U → V is a linear tranformation if T (c1 α + c2 β) = c1 T (α) + c2 T (β) ∀ α, β ∈ U

and c1 , c2 ∈ F .
(2.) T is called the linear operator. Linear transformation preserves vector addition
and scalar multiplication.
(3.) If a linear transformation T : U → V is 1 − 1 then T is called monomorphism.
(4.) If a linear transformation T : U → V is onto then T is called epimorphism.
(5.) If a linear transformation T : U → V is both 1 − 1 and onto then T is called
isomorphism.
(6.) Let V be a vector space and if T : V → V is a linear transformation and is both
1 − 1 and onto, then T is called automorphism.
(7.) If T : U → V , then the elements of U are denoted by α, β, γ, ... and the elements
of V are denoted by α0 , β 0 , γ 0 .... i.e., T (α) = α0 , T (β) = β 0 ,...

Theorems on Linear transformation

• Theorem: If T : U → V is a linear transformation then T (0) = 00 where 0 is the zero

vector of U and 00 is the zero vector of V .
Proof: Since T : U → V is a linear transformation, for α ∈ U ∃ α0 ∈ V such that
T (α) = α0
Consider α + 0 = α ⇒ T (α + 0) = T (α)
⇒ T (α) + T (0) = T (α) [∵ T is linear]
0 0 0
⇒ α + T (0) = α + 0
⇒ T (0) = 00 [By left cancellation law] 

43
1
• Theorem: If T : U → V is a linear transformation then T (−α) = −T (α) ∀ α ∈ U .
Proof: Since T : U → V is a linear transformation, for α ∈ U ∃ α0 ∈ V such that
T (α) = α0
Consider α + (−α) = 0 ⇒ T (α + (−α)) = T (0)
⇒ T (α) + T (−α) = 00 [∵ T is linear]
0
⇒ T (−α) = 0 − T (α)
⇒ T (−α) = −T (α) 
• Theorem: If T : U → V is a linear transformation then T (c1 α1 + c2 α2 + .... + cn αn )
= c1 T (α1 ) + c2 T α2 + .... + cn T (αn ) where α1 , α2 , ..., αn ∈ U and c1 , c2 , ..., cn ∈ F
Proof: Let T (c1 α1 + c2 α2 + .... + cn αn ) = c1 T (α1 ) + c2 T α2 + .... + cn T (αn ) ...(1)
We prove (1) by mathematical induction on n.
For n = 1, T (c1 α1 ) = c1 T (α1 ) [∵ T is linear]
∴ (1) is true for n = 1.
Assume that (1) is true for n = k i.e.,
T (c1 α1 + c2 α2 + .... + ck αk ) = c1 T (α1 ) + c2 T α2 + .... + ck T (αk )
Consider T (c1 α1 + c2 α2 + .... + ck αk + ck+1 αk+1 )
= T (c1 α1 + c2 α2 + .... + ck αk ) + T (ck+1 αk+1 ) [∵ T is linear]
= T (c1 α1 + c2 α2 + .... + ck αk ) + ck+1 T (αk+1 ) [∵ T is linear]
= c1 T (α1 ) + c2 T α2 + .... + ck T (αk ) + ck+1 T (αk+1 ) [By assumption]
∴ T (c1 α1 +c2 α2 +....+ck αk +ck+1 αk+1 ) = c1 T (α1 )+c2 T α2 +....+ck T (αk )+ck+1 T (αk+1 )
Thus (1) is true for n = k + 1 and so (1) is true for all positive integers n 
• Theorem: If α1 , α2 , ..., αn is any basis of a vector space U and β1 , β2 , ..., βn is any basis
of a vector space V then there exist unique linear transformation T from U to V such
that, T (αi ) = βi ∀ 0 ≤ i ≤ n.
Proof: Since {α1 , α2 , ..., αn } is a basis of U for any α ∈ U , we have α = a1 α1 + a2 α2 +
....+an αn where a1 , a2 , ..., an ∈ F . Also for any β ∈ U , we have β = b1 α1 +b3 α+....+bn αn
where b1 , b2 , ..., bn ∈ F
Define T : U → V by T (α) = T (a1 α1 + a2 α2 + ... + an αn ) = a1 β1 + a2 β2 + ... + an βn .
So T (β) = T (b1 α1 + b2 α2 + ... + bn αn ) = b1 β1 + b2 β2 + ... + bn βn
To prove: T is a unique linear transformation.
Consider T (α + β) = T (a1 α1 + a2 α2 + ... + an αn + b1 α1 + b2 α2 + ... + bn αn )
= T [(a1 + b1 )α1 + (a2 + b2 )α2 + ... + (an + bn )αn ]
= (a1 + b1 )β1 + (a2 + b2 )β2 + ... + (an + bn )βn [By definition of T ]
= a1 β1 + b1 β1 + a2 β2 + b2 β2 + ...an αn + bn βn
= a1 β1 + a2 β2 + ... + an βn + b1 β1 + b2 β2 + ... + bn βn
= T (α) + T (β)
∴ T (α + β) = T (α) + T (β)
Now consider T (cα) = T [ca1 α1 + ca2 α2 + ... + can αn ]
= ca1 β1 + ca2 β2 + ... + can βn [By definition of T ]
= c(a1 β1 + a2 β2 + ... + an βn ) = cT (α)
∴ T (cα) = cT (α)
Hence T is linear.
Now consider T (αi ) = T (0α1 + 0α2 + ... + αi + ... + 0αn ) = 0β1 + 0β2 + ... + βi + ... + 0βn
= βi ∴ T (αi ) = βi for 0 ≤ i ≤ n
Uniqueness: Suppose if T 0 is also a linear transformation such that T 0 (αi ) = βi

44

2
 Then T 0 (α) = T 0 (a1 α1 + a2 α2 + ... + an αn )
= T 0 (a1 α1 ) + T 0 (a2 α2 ) + ... + T 0 (an αn ) [∵ T 0 is linear]
= a1 T 0 (α1 ) + a2 T 0 (α2 ) + ... + an T 0 (αn ) [∵ T 0 is linear]
= a1 β1 + a2 β2 + ... + an βn [∵ T 0 (αi ) = βi ]
= T (α) and so T 0 (α) = T (α)
∴ T is unique 

• Theorem: Every n dimensional vector space V over a field F is isomorphic to Vn (F ).

Proof: Let {α1 , α2 , ..., αn } be a basis of V . Then for any α ∈ V we have α = a1 α1 +
a2 α2 + ... + an αn where a1 , a2 , ..., an ∈ F . Also for β ∈ V we have β = b1 α1 + b2 α2 +
... + bn αn where b1 , b2 , ..., bn ∈ F
Define T : V → Vn (F ) by T (α) = (a1 , a2 , ..., an ) for all α ∈ V
T is well defined: For (a1 , a2 , ..., an ) = (b1 , b2 , ..., bn )
⇒ a1 = b1 , a2 = b2 ,..., an = bn
⇒ a1 α1 + a2 α2 + ... + an αn = b1 α1 + b2 α2 + ... + bn αn
⇒ T (a1 α1 + a2 α2 + ... + an αn ) = T (b1 α1 + b2 α2 + ... + bn αn )
⇒ T (α) = T (β)
T is linear: For T (α + β) = T [(a1 α1 + a2 α2 + ... + an αn )+ (b1 α1 + b2 α2 + ... + bn αn )]
= T [(a1 + b1 )α1 + (a2 + b2 )α2 + ... + (an + bn )αn ]
= (a1 + b1 , a2 + b2 , ..., an + bn )
= (a1 , a2 , ..., an ) + (b1 , b2 , ..., bn )
= T (α) + T (β) ∴ T (α + β) = T (α) + T (β)
Now consider T (cα) = T [ca1 α1 + ca2 α2 + ...can αn ] = (ca1 , ca2 , ..., can ) = c(a1 , a2 , ..., an )
= cT (α) ∴ T (cα) = cT (α)
T is 1-1: For T (α) = T (β)
⇒ T (a1 α1 + a2 α2 + ... + an αn ) = T (b1 α1 + b2 α2 + ... + bn αn )
⇒ (a1 , a2 , ..., an ) = (b1 , b2 , ..., bn )
⇒ a1 = b1 , a2 = b2 ,...., an = bn
⇒ α = (a1 α1 + a2 α2 + ... + an αn ) = (b1 α1 + b2 α2 + ... + bn αn ) = β
⇒ α = β ∴ T (α) = T (β) ⇒ α = β
T is onto: For each (a1 , a2 , ..., an ) ∈ Vn (F ) ∃ α ∈ V such that (a1 , a2 , ..., an ) = T (α).
Thus V is isomorphic to Vn (F ) 

• Theorem: Sum of any two linear transformation is a linear transformation.

Proof: Let T : U → V and T 0 : U → V be two linear transformation from a vector
space U into a vector space V over the field F . To prove: T + T 0 : U → V is linear.
Then (T + T 0 )(α) = T (α) + T 0 (α) ∀ α ∈ U
For α, β ∈ V Consider (T + T 0 )(α + β) = T (α + β) + T 0 (α + β)
= T (α) + T (β) + T 0 (α) + T 0 (β) [∵ T and T 0 are linear]
0 0
= (T + T )(α) + (T + T )(β)
∴ (T + T 0 )(α + β) = (T + T 0 )(α) + (T + T 0 )(β)
For α ∈ U and c ∈ F Now consider (T + T 0 )(cα) = T (cα) + T 0 (cα)
= cT (α) + cT 0 (α) [∵ T and T 0 are linear]
= c[T (α) + T 0 (α)] = c(T + T 0 )(α)
∴ (T + T 0 )(cα) = c(T + T 0 )(α)
Thus T + T 0 is linear 

45
3
• Theorem: Product of any two linear transformation is a linear transformation.
Proof: Let T : U → V and T 0 : U → V be two linear transformation from a vector
space U into a vector space V over the field F . To prove: T T 0 : U → V is linear.
Define (T T 0 )(α) = T [T 0 (α)] ∀ α ∈ U
For α, β ∈ V Consider (T T 0 )(α + β) = T [T 0 (α + β)]
= T [T 0 (α) + T 0 (β)] [∵ T 0 is linear]
0 0
= (T T )(α) + (T T )(β) [∵ T is linear]
0 0 0
∴ (T T )(α + β) = (T T )(α) + (T T )(β)
For α ∈ U and c ∈ F Now consider (T T 0 )(cα) = T [T 0 (cα)]
= T [cT 0 (α)] [∵ T 0 is linear]
= c[(T T 0 )(α)] [∵ T is linear]
∴ (T T 0 )(cα) = c(T T 0 )(α)
Thus T T 0 is linear 

• Theorem: Composition of two linear transformation is a linear transformation.

Proof: Let T : U → V and T 0 : U → V be two linear transformation from a vector
space U into a vector space V over the field F . To prove: T ◦ T 0 : U → V is linear.
Define (T ◦ T 0 )(α) = T [T 0 (α)] ∀ α ∈ U
For α, β ∈ V Consider (T ◦ T 0 )(α + β) = T [T 0 (α + β)]
= T [T 0 (α) + T 0 (β)] [∵ T 0 is linear]
= (T T 0 )(α) + (T T 0 )(β) [∵ T is linear]
∴ (T ◦ T 0 )(α + β) = (T ◦ T 0 )(α) + (T ◦ T 0 )(β)
For α ∈ U and c ∈ F Now consider (T ◦ T 0 )(cα) = T [T 0 (cα)]
= T [cT 0 (α)] [∵ T 0 is linear]
0
= c[(T T )(α)] [∵ T is linear]
0 0
∴ (T ◦ T )(cα) = c(T ◦ T )(α)
Thus T ◦ T 0 is linear 

Problems

1. If T : V3 (R) → V2 (R) defined by T (x, y, z) = (x + y, y + z) then show that T is linear.

Also find T (1, 1, 1) and T (0, −1, 2).
Solution : Given, T (x, y, z) = (x + y, y + z)
Let α = (x1 , y1 , z1 ) ⇒ T (α) = T (x1 , y1 , z1 ) = (x1 + y1 , y1 + z1 )
and β = (x2 , y2 , z2 ) ⇒ T (β) = T (x2 , y2 , z2 ) = (x2 + y2 , y2 + z2 )
Consider T (α + β) = T (x1 + x2 , y1 + y2 , z1 + z2 )
= (x1 + x2 + y1 + y2 , y1 + y2 + z1 + z2 )
= (x1 + y1 , y1 + z1 ) + (x2 + y2 , y2 + z2 ) = T (α) + T (β)
∴ T (α + β) = T (α) + T (β)
Consider, T (cα) = T (cx1 , cy1 , cz1 )
= (cx1 + cy1 , cy1 + cz1 )
= c(x1 + y1 , y1 + z1 ) = cT (α)
∴ T (cα) = cT (α)
Hence, T is a linear transformation.
Since T (x, y, z) = (x + y, y + z) we have
T (1, 1, 1) = (2, 2) and T (0, −1, 2) = (−1, 1) ♠

46

4
2. Examine whether the transformation T : R2 → R3 defined by T (x, y) = (x − y, y, x + y)
is a linear transformation or not.
Solution : Given, T (x, y) = (x − y, y, x + y)
Let α = (x1 , y1 ) ⇒ T (α) = T (x1 , y1 ) = (x1 − y1 , y1 , x1 + y1 )
and β = (x2 , y2 ) ⇒ T (β) = T (x2 , y2 ) = (x2 − y2 , y2 , x2 + y2 )
Consider T (α + β) = T (x1 + x2 , y1 + y2 )
= (x1 + x2 − (y1 + y2 ), y1 + y2 , x1 + x2 + y1 + y2 )
= (x1 + x2 − y1 − y2 ), y1 + y2 , x1 + x2 + y1 + y2 )
= (x1 − y1 , y1 , x1 + y1 ) + (x2 − y2 , y2 , x2 + y2 )
= T (α) + T (β)
∴ T (α + β) = T (α) + T (β)
Consider, T (cα) = T (cx1 , cy1 )
= (cx1 − cy1 , cy1 , cx1 + cy1 )
= c(x1 − y1 , y1 , x1 + y1 ) = cT (α)
∴ T (cα) = cT (α)
Hence, T is a linear transformation. ♠
3. If T is a mapping from V (R) to V3 (R) defined by x → x, 2x2 , x3 . Verify whether T is
linear or not.
Solution: Given, x → x, x2 , x3 . Let T (x) = (x, x2 , x3 )
Let α = x1 ⇒ T (α) = T (x1 ) = (x1 , x21 , x31 )
and β = x2 ⇒ T (β) = T (x2 ) = (x2 , x22 , x32 )
Consider, T (α + β) = T (x1 + x2 ) = (x1 + x2 , (x1 + x2 )2 , (x1 + x2 )3 )
= (x1 + x2 , x21 + x22 + 2x1 x2 , x31 + x32 + 3x21 x2 + 3x1 x22 )......(1)
Consider, T (α) + T (β) = (x1 , x21 , x31 ) + (x2 , x22 , x32 )
= (x1 + x2 , x21 + x22 , x31 + x32 )......(2)
From (1) and (2), we have T (α + β) 6= T (α) + T (β)
Hence, T is not linear. ♠
4. If T : V3 (R) → V2 (R) defined by T (a1 , a2 , a3 ) = (a2 , a3 ). Verify whether T is linear or
not.
Solution: Given, T (a1 , a2 , a3 ) = (a2 , a3 ).
Let T (x, y, z) = (y, z)
Let α = (x1 , y1 , z1 ) ⇒ T (α) = T (x1 , y1 , z1 ) = (y1 , z1 )
and β = (x2 , y2 , z2 ) ⇒ T (β) = T (x2 , y2 , z2 ) = (y2 , z2 )
Consider, T (α + β) = T (x1 + x2 , y1 + y2 , z1 + z2 )
= (y1 + y2 , z1 + z2 )
= (y1 , z1 ) + (y2 , z2 )
= T (α) + T (β)
∴ T (α + β) = T (α) + T (β)
Consider, T (cα) = T (cx1 , cy1 , cz1 ) = (cy1 , cz1 ) = c(y1 , z1 ) = cT (α)
∴ T (cα) = cT (α)
Hence, T is linear. ♠
5. Is the projection from V3 (R) to V2 (R) given by T (x1 , x2 , x3 ) = (x1 , x2 ) a linear trans-
formation?
Solution: Given, T (x1 , x2 , x3 ) = (x1 , x2 )

47

5
 Let T (x, y, z) = (x, y)
Let α = (x1 , y1 , z1 ) ⇒ T (α) = T (x1 , y1 , z1 ) = (x1 , y1 )
and β = (x2 , y2 , z2 ) ⇒ T (β) = T (x2 , y2 , z2 ) = (x2 , y2 )
Consider, T (α + β) = T (x1 + x2 , y1 + y2 , z1 + z2 )
= (x1 + x2 , y1 + y2 )
= (x1 , y1 ) + (x2 , y2 )
= T (α) + T (β)
∴ T (α + β) = T (α) + T (β)
Consider, T (cα) = T (cx1 , cy1 , cz1 ) = (cx1 , cy1 ) = c(x1 , y1 ) = cT (α)
∴ T (cα) = cT (α)
Hence, T is linear. ♠

6. i) Is the transformation T (x, y) = (x + 4, y − 2) linear?

ii) Is the transformation T (x, y) = (x + 1, y + 2) linear?
Solution: i) Given, T (x, y) = (x + 4, y − 2)
Let α = (x1 , y1 ) ⇒ T (α) = T (x1 , y1 ) = (x1 + 4, y1 − 2)
and β = (x2 , y2 ) ⇒ T (β) = T (x2 , y2 ) = (x2 + 4, y2 − 2)
Consider, T (α + β) = T (x1 + x2 , y1 + y2 )
= (x1 + x2 + 4, y1 + y2 − 2)......(1)
Consider, T (α) + T (β) = (x1 + 4, y1 − 2) + (x2 + 4, y2 − 2)
= (x1 + x2 + 8, y1 + y2 − 4)......(2)
From (1) and (2), we have T (α + β) 6= T (α) + T (β)
Hence, T is not linear.
ii) Given, T (x, y) = (x + 1, y + 2)
Let α = (x1 , y1 ) ⇒ T (α) = T (x1 , y1 ) = (x1 + 1, y1 + 2)
and β = (x2 , y2 ) ⇒ T (β) = T (x2 , y2 ) = (x2 + 1, y2 + 2)
Consider, T (α + β) = T (x1 + x2 , y1 + y2 )
= (x1 + x2 + 1, y1 + y2 + 2)......(1)
Consider, T (α) + T (β) = (x1 + 1, y1 + 2) + (x2 + 1, y2 + 2)
= (x1 + x2 + 2, y1 + y2 + 4)......(2)
From (1) and (2), we have T (α + β) 6= T (α) + T (β)
Hence, T is not linear. ♠

7. Show that the correspondence (x1 , x2 , x3 ) → (−x2 , x1 , x3 ) is a linear transformation in

V3 (R).
Solution: Given, T (x1 , x2 , x3 ) = (−x2 , x1 , x3 )
Let T (x, y, z) = (−y, x, z)
Let α = (x1 , y1 , z1 ) ⇒ T (α) = T (x1 , y1 , z1 ) = (−y1 , x1 , z1 )
Let β = (x2 , y2 , z2 ) ⇒ T (β) = T (x2 , y2 , z2 ) = (−y2 , x2 , z2 )
Consider, T (α + β) = (x1 + x2 , y1 + y2 , z1 + z2 )
= (−(y1 + y2 ), x1 + x2 , z1 + z2 )
= (−y1 − y2 , x1 + x2 , z1 + z2 )
= (−y1 , x1 , z1 ) + (−y2 , x2 , z2 ) = T (α) + T (β)
∴ T (α + β) = T (α) + T (β)
Consider, T (cα) = T (cx1 , cy1 , cz1 ) = (−cy1 , cx1 , cz1 ) = c(−y1 , x1 , z1 ) = cT (α)
∴ T (cα) = cT (α)

48

6
 Hence, T is linear. ♠

8. If T : R2 → R2 defined by T (x, y) = (xcosθ − ysinθ, xsinθ + ycosθ) then prove that T

is linear.
Solution: Given, T (x, y) = (x cos θ − y sin θ, x sin θ + y cos θ)
Let α = (x1 , y1 ) ⇒ T (α) = T (x1 , y1 ) = (x1 cos θ − y1 sin θ, x1 sin θ + y1 cos θ)
Let α = (x1 , y1 ) ⇒ T (β) = T (x2 , y2 ) = (x2 cos θ − y2 sin θ, x2 sin θ + y2 cos θ)
Consider, T (α + β) = T (x1 + x2 , y1 + y2 )
= ((x1 + x2 ) cos θ − (y1 + y2 ) sin θ, (x1 + x2 ) sin θ + (y1 + y2 ) cos θ)
= (x1 cos θ + x2 cos θ − y1 sin θ − y2 sin θ, x1 sin θ + x2 sin θ + y1 cos θ + y2 cos θ)
= (x1 cos θ − y1 sin θ, x1 sin θ + y1 cos θ) + (x2 cos θ − y2 sin θ, x2 sin θ + y2 cos θ)
= T (α + β)
∴ T (α + β) = T (α) + T (β)
Consider, T (cα) = T (cx1 , cy1 )
= (cx1 cos θ − cy1 sin θ, cx1 sin θ + cy1 cos θ)
= c(x1 cos θ − y1 sin θ, x1 sin θ + y1 cos θ) = cT (α)
∴ T (cα) = cT (α)
Hence, T is linear. ♠

9. Prove that the transformation (x, y, −z) is a linear.

Solution: Let T (x, y, z) = (x, y, −z)
Let α = (x1 , y1 , z1 ) ⇒ T (α) = T (x1 , y1 , z1 ) = (x1 , y1 , −z1 )
and β = (x2 , y2 , z2 ) ⇒ T (β) = T (x2 , y2 , z2 ) = (x2 , y2 , −z2 )
Consider, T (α + β) = (x1 + x2 , y1 + y2 , z1 + z2 )
= (x1 + x2 , y1 + y2 , −(z1 + z2 ))
= (x1 + x2 , y1 + y2 , −z1 − z2 )
= (x1 , y1 , −z1 ) + (x2 , y2 , −z2 ) = T (α) + T (β)
∴ T (α + β) = T (α) + T (β)
Consider, T (cα) = T (cx1 , cy1 , cz1 ) = (cx1 , cy1 , −cz1 ) = c(x1 , y1 , −z1 ) = cT (α)
∴ T (cα) = cT (α)
Hence, T is linear. ♠

10. Show that T : R3 → R2 defined by T (x, y, z) = (x + z, x + y + z) is linear.

Solution: Given, T (x, y, z) = (x + z, x + y + z)
Let α = (x1 , y1 , z1 ) ⇒ T (α) = T (x1 , y1 , z1 ) = (x1 + z1 , x1 + y1 + z1 )
and β = (x2 , y2 , z2 ) ⇒ T (β) = T (x2 , y2 , z2 ) = (x2 + z2 , x2 + y2 + z2 )
Consider, T (α + β) = T (x1 + x2 , y1 + y2 , z1 + z2 )
= (x1 + x2 + z1 + z2 , x1 + x2 + y1 + y2 + z1 + z2 )
= (x1 + z1 , x1 + y1 + z1 ) + (x2 + z2 , x2 + y2 + z2 ) = T (α) + T (β)
∴ T (α + β) = T (α) + T (β)
Consider, T (cα) = T (cx1 , cy1 , cz1 )
= (cx1 + cz1 , cx1 + cy1 + cz1 )
= c(x1 + z1 , x1 + y1 + z1 ) = cT (α)
∴ T (cα) = cT (α)
Hence, T is linear. ♠

11. Let T1 : V3 (R) → V2 (R) defined by T1 (x, y, z) = (y, x + z) and T2 : V3 (R) → V2 (R)

49

7
 defined by T2 (x, y, z) = (2z, x − y). Find T1 + T2 and 3T1 − T2
Solution: Given, T1 (x, y, z) = (y, x + z) and T2 (x, y, z) = (2z, x − y)
Consider, (T1 + T2 )(x, y, z) = T1 (x, y, z) + T2 (x, y, z)
= (y, x + z) + (2z, x − y)
= (y + 2z, x + z + x − y)
= (y + 2z, 2x − y + z)
∴ (T1 + T2 )(x, y, z) = (y + 2z, 2x − y + z)
Consider, (3T1 − T2 )(x, y, z) = 3T1 (x, y, z) − T2 (x, y, z)
= 3(y, x + z) − (2z, x − y)
= (3y, 3x + 3z) − (2z, x − y)
= (3y − 2z, 3x + 3z − x + y)
= (3y − 2z, 2x + y + 3z)
∴ (3T1 − T2 )(x, y, z) = (3y − 2z, 2x + y + 3z) ♠

12. Show that the transformation (α, β, γ) 7→ (−β, α, γ) is an automorphism in V3 (R).

Solution: Given, (α, β, γ) 7→ (−β, α, γ)
Let T (x, y, z) = (−y, x, z)
Let α = (x1 , y1 , z1 ) ⇒ T (α) = T (x1 , y1 , z1 ) = (−y1 , x1 , z1 )
Let β = (x2 , y2 , z2 ) ⇒ T (β) = T (x2 , y2 , z2 ) = (−y2 , x2 , z2 )
Consider, T (α + β) = (x1 + x2 , y1 + y2 , z1 + z2 )
= (−(y1 + y2 ), x1 + x2 , z1 + z2 )
= (−y1 − y2 ), x1 + x2 , z1 + z2 )
= (−y1 , x1 , z1 ) + (−y2 , x2 , z2 ) = T (α) + T (β)
∴ T (α + β) = T (α) + T (β)
Consider, T (cα) = T (cx1 , cy1 , cz1 ) = (−cy1 , cx1 , cz1 ) = c(−y1 , x1 , z1 ) = cT (α)
∴ T (cα) = cT (α)
Hence, T is linear.
Let T (α) = T (β)
⇒ (−y1 , x1 , z1 ) = (−y2 , x2 , z2 )
⇒ −y1 = −y2 , x1 = x2 , z1 = z2
⇒ y1 = y2 , x1 = x2 , z1 = z2
∴ α = (x1 , y1 , z1 ) = (x2 , y2 , z2 ) = β
Hence, T (α) = T (β) ⇒ α = β and so T is one-one.
For each (−y1 , x1 , z1 ) ∃ (x1 , y1 , z1 ) such that (−y1 , x1 , z1 ) = T (x1 , y1 , z1 )
Hence, T is onto.
Thus, T is an automorphism. ♠

13. Show that the correspondence T (x, y, z) = (x, y, −z) is an automorphism.

Solution: Let T (x, y, z) = (x, y, −z)
Let α = (x1 , y1 , z1 ) ⇒ T (α) = T (x1 , y1 , z1 ) = (x1 , y1 , −z1 )
Let β = (x2 , y2 , z2 ) ⇒ T (β) = T (x2 , y2 , z2 ) = (x2 , y2 , −z2 )
Consider, T (α + β) = (x1 + x2 , y1 + y2 , z1 + z2 )
= (x1 + x2 , y1 + y2 , −(z1 + z2 ))
= (x1 + x2 , y1 + y2 , −z1 − z2 )
= (x1 , y1 , −z1 ) + (x2 , y2 , −z2 ) = T (α) + T (β)
∴ T (α + β) = T (α) + T (β)

50
8
 Consider, T (cα) = T (cx1 , cy1 , cz1 ) = (cx1 , cy1 , −cz1 ) = c(x1 , y1 , −z1 ) = cT (α)
∴ T (cα) = cT (α)
Hence, T is linear.
Let T (α) = T (β)
⇒ (x1 , y1 , −z1 ) = (x2 , y2 , −z2 )
⇒ x1 = x2 , y1 = y2 , −z1 = −z2
⇒ x1 = x2 , y1 = y2 , z1 = z2
∴ α = (x1 , y1 , z1 ) = (x2 , y2 , z2 ) = β
Hence, T (α) = T (β) ⇒ α = β and so T is one-one.
For each (x1 , y1 , −z1 ) ∃ (x1 , y1 , z1 ) such that (x1 , y1 , −z1 ) = T (x1 , y1 , z1 )
Hence, T is onto.
Thus, T is an automorphism. ♠

14. Show that the correspondence T (x, y, z) = (y, −x, z) is an automorphism.

Solution: Let T (x, y, z) = (y, −x, z)
Let α = (x1 , y1 , z1 ) ⇒ T (α) = T (x1 , y1 , z1 ) = (y1 , −x1 , z1 )
Let β = (x2 , y2 , z2 ) ⇒ T (β) = T (x2 , y2 , z2 ) = (y2 , −x2 , z2 )
Consider, T (α + β) = (x1 + x2 , y1 + y2 , z1 + z2 )
= (y1 + y2 , −(x1 + x2 ), z1 + z2 )
= (y1 + y2 , −x1 − x2 , z1 + z2 )
= (y1 , −x1 , z1 ) + (y2 , −x2 , z2 ) = T (α) + T (β)
∴ T (α + β) = T (α) + T (β)
Consider, T (cα) = T (cx1 , cy1 , cz1 ) = (cy1 , −cx1 , cz1 ) = c(y1 , −x1 , z1 ) = cT (α)
∴ T (cα) = cT (α)
Hence, T is linear.
Let T (α) = T (β)
⇒ (y1 , −x1 , −z1 ) = (y2 , −x2 , −z2 )
⇒ y1 = y2 , −x1 = −x2 , z1 = z2
⇒ y1 = y2 , x1 = x2 , z1 = z2
∴ α = (x1 , y1 , z1 ) = (x2 , y2 , z2 ) = β
Hence, T (α) = T (β) ⇒ α = β and so T is one-one.
For each (y1 , −x1 , z1 ) ∃ (x1 , y1 , z1 ) such that (y1 , −x1 , z1 ) = T (x1 , y1 , z1 )
Hence, T is onto.
Thus, T is an automorphism. ♠

Exercise

1. Show that (x, y, z) → (x, y) is linear.

2. Show that T : R2 → R2 defined by T (x, y) = (x + 2y, 2x + y) is linear.

3. Show that the correspondence (x1 , x2 , x3 ) → (x2 , −x1 , −x3 ) is an automorphism from
V3 (R) into V3 (R)

4. Is the projection from V3 (R) to V2 (R) defined by T (x1 , x2 , x3 ) = (x1 , x2 ) linear? Is T

one-one and onto?

5. Show that T : V3 (R) → V3 (R) defined by T (x, y, z) = (2x + y, y − z, 2y + 4z) is linear.

51

9
6. If T1 (x, y, z) = (y, −x, z) and T2 (x, y, z) = (x + z, x + y + z, y − 2x), find T1 − 3T2 , 4T2
and −2T1 + T2

Problems

1. Find the linear transformation T : V2 (R) → V2 (R) such that T (1, 2) = (3, 0) and
T (2, 1) = (1, 2).
Solution: Let T (x, y) be the required linear transformation.
Consider (x, y) = c1 (1, 2) + c2 (2, 1)
c1 + 2c2 = x ...(1)
2c1 + c2 = y ...(2)
Solving (1) and (2) we get
2c1 + 4c2 = 2x
−2c1 − c2 = −y
3c2 = 2x − y
2x − y
∴ c2 =
3
 2x − y  3x − 4x + 2y
(1) ⇒ c1 = x − 2c2 = x − 2 =
3 3
−x + 2y
∴ c1 =
3
2x − y

Hence (x, y) = −x+2y


3
(1, 2) + (2, 1)
3
Applying T on both sides, we get
−x + 2y
2x − y

T (x, y) = T (1, 2) + T (2, 1)
3 3
−x + 2y
2x − y

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

T (x, y) = − x + 2y + ,0 +
3 3
 −x + 5y 4x − 2y 
∴ T (x, y) = , ♠
3 3
2. Find the linear transformation if T (1, 1, 1) = 3, T (0, 1, −2) = 1 and T (0, 0, 1) = 2.
Solution: Let T (x, y, z) be the required linear transformation.
Consider (x, y, z) = c1 (1, 1, 1) + c2 (0, 1, −2) + c3 (0, 0, 1)
c1 = x
c1 + c2 = y ...(1)
c1 − 2c2 + c3 = z ...(2)
∴ c1 = x
(1) ⇒ c2 = y − c1 = y − x ∴ c2 = y − x
(2) ⇒ c3 = z − c1 + 2c2 = z − x + 2y − 2x ∴ c3 = −3x + 2y + z
Hence (x, y, z) = x(1, 1, 1) + (y − x)(0, 1, −2) + (−3x + 2y + z)(0, 0, 1)
Applying T on both sides, we get
T (x, y, z) = xT (1, 1, 1) + (y − x)T (0, 1, −2) + (−3x + 2y + z)T (0, 0, 1)
T (x, y, z) = 3x + (y − x) + 2(−3x + 2y + z)

52

10
 T (x, y, z) = 3x + y − x − 6x + 4y + 2z
∴ T (x, y, z) = −4x + 5y + 2z ♠

3. Find the linear transformation if T (−1, 1) = (−1, 0, 2) and T (2, 1) = (1, 2, 1).
Proof: Let T (x, y) be the required linear transformation.
Consider (x, y) = c1 (−1, 1) + c2 (2, 1)
−c1 + 2c2 = x ...(1)
c1 + c2 = y ...(2)
Solving (1) and (2) we get
−c1 + 2c2 = x
c1 + c2 = y
3c2 = x + y
x+y
∴ c2 =
3
x + y
3y − x − y
(2) ⇒ c1 = y − c2 = y − =
3 3
−x + 2y
∴ c1 =
3
−x + 2y
x + y

Hence (x, y) = (−1, 1) + (2, 1)
3 3
Applying T on both sides, we get
−x + 2y
x + y

T (x, y) = T (−1, 1) + T (2, 1)
3 3
−x + 2y
x + y

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

T (x, y) = + ,0 + , +
3 3 3 3 3
2x − y 2x + 2y −x + 5y

∴ T (x, y) = , , ♠
3 3 3
4. Find the linear transformation if T (1, 1, 1) = (1, 1, 1), T (1, 2, 3) = (−1, −2, −3) and
T (1, 1, 2) = (2, 2, 4).
Solution: Let T (x, y, z) be the required linear transformation.
Consider (x, y, z) = c1 (1, 1, 1) + c2 (1, 2, 3) + c3 (1, 1, 2)
c1 + c2 + c3 = x ...(1)
c1 + 2c2 + c3 = y ...(2)
c1 + 3c2 + 2c3 = z ...(3)
Solving (1) and (2) we get
c1 + c2 + c3 = x
−c1 − 2c2 − c3 = −y
−c2 = x − y
∴ c2 = −x + y
Solving (2) and (3) we get
c1 + 2c2 + c3 = y
−c1 − 3c2 − 2c3 = −z
−c2 − c3 = y − z

53

11
 ⇒ c2 + c3 = −y + z ⇒ c3 = −y + z − c2 = −y + z + x − y = x − 2y + z
∴ c3 = x − 2y + z
(1) ⇒ c1 = x − c2 − c3 = x + x − y − x + 2y − z = x + y − z
∴ c1 = x + y − z
Hence (x, y, z) = (x + y − z)(1, 1, 1) + (−x + y)(1, 2, 3) + (x − 2y + z)(1, 1, 2)
Applying T on both sides, we get
T (x, y, z) = (x + y − z)T (1, 1, 1) + (−x + y)T (1, 2, 3) + (x − 2y + z)T (1, 1, 2)
T (x, y, z) = (x + y − z)(1, 1, 1) + (−x + y)(−1, −2, −3) + (x − 2y + z)(2, 2, 4)
T (x, y, z) = (x + y − z + x − y + 2x − 4y + 2z, x + y − z + 2x − 2y + 2x − 4y + 2z, x +
y − z + 3x − 3y + 4x − 8y + 4z)
∴ T (x, y, z) = (4x − 4y + z, 5x − 5y + z, 8x − 10y + 3z) ♠

5. Find T (x, y) if T (1, 1) = (0, 1, 2) and T (−1, 1) = (2, 1, 0).

Solution: Let T (x, y) be the required linear transformation.
Consider (x, y) = c1 (1, 1) + c2 (−1, 1)
c1 − c2 = x ...(1)
c1 + c2 = y ...(2)
Solving (1) and (2) we get
c1 − c2 = x
c1 + c2 = y
2c1 = x + y
x+y
∴ c1 = 2
x + y
2y − x − y
(2) ⇒ c2 = y − c1 = y − =
2 2
−x + y
∴ c1 =
2
x + y
−x + y

Hence (x, y) = (1, 1) + (−1, 1)
2 2
Applying T on both sides we get
x + y
−x + y

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

T (x, y) = (0, 1, 2) + (2, 1, 0)
2 2
x + y −x + y

T (x, y) = 0 − x + y, + ,x + y + 0
2 2
∴ T (x, y) = (−x + y, y, x + y) ♠

6. Find the linear transformation T : V3 (R) → V2 (R) such that T (e1 ) = (−1, 0), T (e2 ) =
(1, 1) and T (e3 ) = (0, −1)
Solution: Let T (x, y, z) be the required linear transformation.
Given T (1, 0, 0) = (−1, 0), T (0, 1, 0) = (1, 1) and T (0, 0, 1) = (0, −1)
Consider (x, y, z) = c1 (1, 0, 0) + c2 (0, 1, 0) + c3 (0, 0, 1)
c1 = x , c2 = y , c3 = z
Hence (x, y, z) = x(1, 0, 0) + y(0, 1, 0) + z(0, 0, 1)
Applying T on both sides we get
T (x, y, z) = xT (1, 0, 0) + yT (0, 1, 0) + zT (0, 0, 1)

54
12
 T (x, y, z) = x(−1, 0) + y(1, 1) + z(0, −1)
∴ T (x, y, z) = (−x + y, y − z) ♠

7. If T : R2 → R is linear such that T (1, 0) = 2 and T (1, 1) = 0. Find T (2, 3) and T (0, 1)
Solution: Let T (x, y) be the required linear transformation.
Consider (x, y) = c1 (1, 0) + c2 (1, 1)
c1 + c2 = x ...(1)
c2 = y
∴ c2 = y
(1) ⇒ c1 = x − c2 = x − y ∴ c1 = x − y
Hence (x, y) = (x − y)(1, 0) + y(1, 1)
Applying T on both sides we get
T (x, y) = (x − y)T (1, 0) + yT (1, 1)
T (x, y) = (x − y)2 + y0 = 2x − 2y
∴ T (x, y) = 2x − 2y
So T (2, 3) = 4 − 6 ∴ T (2, 3) = −2 and T (0, 1) = 0 − 2 ∴ T (0, 1) = −2 ♠

Exercise

1. Find T (x, y) for T (1, 1) = (1, −1, 1, −1) and T (1, −1) = (1, 1, 1, 1)
2. Find a linear transformation T : R2 → R3 given by T (1, 2) = (1, 4, 9) and T (2, −1) =
(2, −3, 4)
3. Find a linear transformation T : R2 → R2 such that T (1, 1) = (0, 1) and T (−1, 1) =
(3, 2)
4. Find a linear mapping if T (1, 1, 1) = (1, 1, 1), T (2, 2, 3) = (3, 3, 5) and T (1, 1, 2) =
(2, 2, 4)

• Definition: Let T be a linear transformation. The order of linear transformation

T is a positive integer n such that T n (x, y, z) = (x, y, z) or T n (x, y) = (x, y)

Problems

1. Find T 2 (x, y) if T (x, y) = (x − y, y)

Solution: Given T (x, y) = (x − y, y)
⇒ T T (x, y) = T (x − y, y)
⇒ T 2 (x, y) = (x − y − y, y) = (x − 2y, y) ♠
2. Find T 2 (x, y) for the transformation defined by T (x, y) = (−y, x + y)
Solution: Given T (x, y) = (−y, x + y)
⇒ T T (x, y) = T (−y, x + y)
⇒ T 2 (x, y) = T (−(x + y), −y + x + y) = (−x − y, x) ♠
3. Find T 2 (x, y, z) if T (x, y, z) = (z, −x, y)
Solution: Given T (x, y, z) = (z, −x, y)
⇒ T T (x, y, z) = T (z, −x, y)
⇒ T 2 (x, y, z) = T (y, −z, −x) ♠

55

13
 Linear Algebra - Linear Transformation
Tejas N S
Assistant Professor in Mathematics, NIE First Grade College, Mysuru
Web: http://nstejas.poeticmath.com
Email: nstejas@gmail.com Mobile: +91 9845410469

10 Matrix of a linear transformation

• Definition: Let T : U → V be a linear transformation from a vector space U into a
vector space V over the field F and let B1 = {α1 , α2 , ..., αm } and B2 = {β1 , β2 , ..., βn }
be the basis of U and V respectively. Then T (α1 ), T (α2 ),..., T (αm ) ∈ V and every
element of V can be written as a linear combination of elements of B2 . Say
T (α1 ) = a11 β1 + a12 β2 + ... + a1n βn
T (α2 ) = a21 β1 + a22 β2 + ... + a2n βn
..
.
T (αm ) = am1 β1 + am2 β2 + ... + amn βn
where aij ∈ F for 1 ≤ i ≤ m, 1 ≤ j ≤ n
a11 a12 · · · · · · a1n
 
 a21 a22 · · · · · · a2n 
 . .. .. 
 .
The coefficient matrix is given by  . . . 

 . .. .. 
 .. . . 
am1 am2 · · · · · · amn
The above matrix is called the matrix of the linear transformation relative to the
basis B1 and B2 .

Problems

1. Find the matrix of linear transformation T : V2 (R) → V2 (R) defined by T (x, y) =

(x, −y) w.r.t basis {(1, 1), (1, 0)} and {(2, 3), (4, 5)}.
Solution: Given, T : V2 (R) → V2 (R) defined by T (x, y) = (x, −y)
Choose B1 = {(1, 1), (1, 0)} and B2 = {(2, 3), (4, 5)}
Consider T (1, 1) = (1, −1)
Let (1, −1) = c1 (2, 3) + c2 (4, 5)
2c1 + 4c2 = 1 ...(1)
3c1 + 5c2 = −1 ...(2)
Solving (1) and (2) we get
6c1 + 12c2 = 3
−6c1 − 10c2 = 2
2c2 = 5
∴ c2 = 5/2
5
2 − 20
(1) ⇒ 2c1 = 1 − 4c2 = 1 − 4 = = −9
2 2

57
14
 ∴ c1 = −9/2
Consider T (1, 0) = (1, 0)
Let (1, 0) = c1 (2, 3) + c2 (4, 5)
2c1 + 4c2 = 1 ...(1)
3c1 + 5c2 = 0 ...(2)
Solving (1) and (2) we get
6c1 + 12c2 = 3
−6c1 − 10c2 = 0
2c2 = 3
∴ c2 = 3/2
3
2 − 12
(1) ⇒ 2c1 = 1 − 4c2 = 1 − 4 = = −5
2 2
∴ c1 = −5/2
 
−9/2 5/2
∴ The matrix is ♠
−5/2 3/2

2. Find the matrix of linear transformation T : V2 (R) → V2 (R) defined by T (x, y) =

(x, y − x) w.r.t basis {(1, 0), (1, 1)}.
Solution: Given T : V2 (R) → V2 (R) defined by T (x, y) = (x, y − x)
Choose B1 = B2 = {(1, 0), (1, 1)}
Consider T (1, 0) = (1, −1)
Let (1, −1) = c1 (1, 0) + c2 (1, 1)
c1 + c2 = 1 ...(1)
c2 = −1
∴ c2 = −1
(1) ⇒ c1 = 1 − c2 = 1 − (−1) = 2 ∴ c1 = 2
Consider T (1, 0) = (1, 0)
Let (1, 0) = c1 (1, 0) + c2 (1, 1)
c1 + c2 = 1 ...(1)
c2 = 0
∴ c2 = 0
(1) ⇒ c1 = 1 − c2 = 1 − 0 = 1 ∴ c1 = 1
 
2 −1
∴ The matrix is ♠
1 0

3. Find the matrix of linear transformation defined by T (x, y, z) = (x + y, y + z) w.r.t

basis {(1, 1, 0), (1, 0, 1), (1, 1, −1)} and {(2, −3), (1, 4)}.
Solution: Given T (x, y, z) = (x + y, y + z)
Choose B1 = {(1, 1, 0), (1, 0, 1), (1, 1, −1)} and B2 = {(2, −3), (1, 4)}
Consider T (1, 1, 0) = (2, 1)
Let (2, 1) = c1 (2, −3) + c2 (1, 4)
2c1 + c2 = 2 ...(1)
−3c1 + 4c2 = 1 ...(2)
Solving (1) and (2) we get

58
15
 8c1 + 4c2 = 8
3c1 − 4c2 = −1
11c1 = 7
∴ c1 = 7/11
7
11 + 21
(2)⇒ 4c2 = 1 + 3c1 = 1 + 3 = = 32/11
11 11
∴ c2 = 8/11
Consider T (1, 0, 1) = (1, 1)
Let (1, 1) = c1 (2, −3) + c2 (1, 4)
2c1 + c2 = 1 ...(1)
−3c1 + 4c2 = 1 ...(2)
Solving (1) and (2) we get
8c1 + 4c2 = 4
3c1 − 4c2 = −1
11c1 = 3
∴ c1 = 3/11
3
11 + 9
(2)⇒ 4c2 = 1 + 3c1 = 1 + 3 = = 20/11
11 11
∴ c2 = 5/11
Consider T (1, 1, −1) = (2, 0)
Let (2, 0) = c1 (2, −3) + c2 (1, 4)
2c1 + c2 = 2 ...(1)
−3c1 + 4c2 = 0 ...(2)
Solving (1) and (2) we get
8c1 + 4c2 = 8
3c1 − 4c2 = 0
11c1 = 8
∴ c1 = 8/11
8

(2)⇒ 4c2 = 3c1 = 3 = 24/11
11
∴ c2 = 6/11
 
7/11 8/11
∴ The matrix is 3/11 5/11 ♠
8/11 6/11

4. Find the matrix of linear transformation T : V2 (R) → V2 (R) defined by T (x, y) =

(x, −y) w.r.t standard basis.
Solution: Given T : V2 (R) → V2 (R) defined by T (x, y) = (x, −y)
Choose B1 = B2 = {(1, 0), (0, 1)}
Then T (1, 0) = (1, 0)
and T (0, 1) = (0,−1) 
1 0
∴ The matrix is ♠
0 −1

59
16
5. Find the matrix of T (x, y, z) = (z − 2y, x + 2y − z).
Solution: Given T (x, y, z) = (z − 2y, x + 2y − z)
Choose B1 = {(1, 0, 0), (0, 1, 0), (0, 0, 1)} and B2 = {(1, 0), (0, 1)}
Then T (1, 0, 0) = (0, 1)
T (0, 1, 0) = (−2, 2)
T (0, 0, 1) = (1, −1)
 
0 1
∴ The matrix is −2 2  ♠
1 −1

6. Find the matrix of linear transformation T : R2 → R3 defined by T (−1, 1) = (−1, 0, 2)

and T (2, 1) = (1, 2, 1).
Solution: Given T (−1, 1) = (−1, 0, 2) and T (2, 1) = (1, 2, 1).
Let T (x, y) be the required linear transformation.
Consider (x, y) = c1 (−1, 1) + c2 (2, 1)
−c1 + 2c2 = x ...(1)
c1 + c2 = y ...(2)
Solving (1) and (2) we get
−c1 + 2c2 = x
c1 + c2 = y
3c2 = x + y
x+y
∴ c2 = 3

(2) ⇒ c1 = y − c1 = y − x+y ∴ c1 = −x+2y

3y−x−y
3
= 3 3
−x+2y x+y



Hence, (x, y) = 3
(−1, 1) + 3 (2, 1)
Applying T on both
sides, we get
T (x, y) = −x+2y x+y


3
T (−1, 1) + 3
T (2, 1)
−x+2y x+y


T (x, y) = 3
(−1, 0, 2) + 3 (1, 2, 1)

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


3
, 3
, 3
Choose B1 = {(1, 0), (0, 1)} and B2 = {(1, 0, 0), (0, 1, 0), (0, 0, 1)}
 2/3, −1/3) and T (0, 1) = (−1/3, 2/3, 5/3)
Now T (1, 0) = (2/3,
2/3 2/3 −1/3
∴ The matrix is ♠
−1/3 2/3 5/3

7. Find the matrix of linear transformation if T (1, 1) = (0, 1) and T (−1, 1) = (3, 2).
Solution: Given T (1, 1) = (0, 1) and T (−1, 1) = (3, 2).
Let T (x, y) be the required linear transformation.
Consider (x, y) = c1 (1, 1) + c2 (−1, 1)
c1 − c2 = x ...(1)
c1 + c2 = y ...(2)
Solving (1) and (2) we get
c1 − c2 = x
c1 + c2 = y
2c1 = x + y

60
17
 x+y
∴ c1 = 2

(2) ⇒ c2 = y − c1 = y − x+y ∴ c2 = −x+y

2y−x−y
2
= 2 2
−x+y
Hence, (x, y) = x+y



2
(1, 1) + 2
(−1, 1)
Applying T on
both sides, we get
T (x, y) = x+y T (1, 1) + −x+y


2
2

T (−1, 1)
x+y −x+y
T (x, y) = 2 (0, 1) + 2
(3, 2)
−3x+3y x+y


T (x, y) = 2
, 2 + (−x + y)
∴ T (x, y) = −3x+3y , −x+3y


2 2
Choose B1 = B2 = {(1, 0), (0, 1)}
Then T (1, 0) = (−3/2,
 −1/2) and  T (0, 1) = (3/2, 3/2)
−3/2 −1/2
∴ The matrix is ♠
3/2 3/2

8. Find the linear transformation and matrix for T (1, 0, 0) = (1, 0), T (0, 1, 0) = (1, 1) and
T (0, 0, 1) = (0, 1).
Solution: Given T (1, 0, 0) = (1, 0), T (0, 1, 0) = (1, 1) and T (0, 0, 1) = (0, 1)
Then B1 = {(1, 0, 0), (0, 1, 0), (0, 0, 1)}
Choose B2 = {(1,0), (0,1)}
1 0
∴ The matrix is 1 1
0 1
To find the linear transformation: Let T (x, y, z) be the required linear transformation.
Consider (x, y, z) = c1 (1, 0, 0) + c2 (0, 1, 0) + c3 (0, 0, 1)
Then, c1 = x , c2 = y and c3 = z
Hence, (x, y, z) = x(1, 0, 0) + y(0, 1, 0) + z(0, 0, 1)
Applying T on both sides, we get
T (x, y, z) = xT (1, 0, 0) + yT (0, 1, 0) + zT (0, 0, 1)
T (x, y, z) = x(1, 0) + y(1, 1) + z(0, 1)
T (x, y, z) = (x + y, y + z) ♠

9. Find the matrix of linear transformation T : R2 → R3 defined by T (x, y) = (−x+2y, y−

3x, −3x + y) relative to basis {(1, 2), (−2, 1)} and {(−1, 0, 2), (1, 2, 3), (1, −1, −1)}.
Solution: Given T : R2 → R3 defined by T (x, y) = (−x + 2y, y − 3x, −3x + y)
Choose B1 = {(1, 2), (−2, 1)} and B2 = {(−1, 0, 2), (1, 2, 3), (1, −1, −1)}.
Consider T (1, 2) = (3, −1, −1)
Let (3, −1, −1) = c1 (−1, 0, 2) + c2 (1, 2, 3) + c3 (1, −1, −1)
−c1 + c2 + c3 = 3 ...(1)
2c2 − c3 = −1 ...(2)
2c1 + 3c2 − c3 = −1 ...(3)
Solving (1) and (3) we get
−2c1 + 2c2 + 2c3 = 6
2c1 + 3c2 − c3 = −1
c2 + c3 = 5 ...(4)
Solving (2) and (4) we get

61
18
 2c2 − c3 = −1
5c2 + c3 = 5
7c2 = 4
∴ c2 = 4/7
4 8+7


(2)⇒ c3 = 2c2 + 1 = 2 7
+1= 7
∴ c3 = 15/7
(1)⇒ c1 = c2 + c3 − 3 = + − 3 = 4+15−21
4
7
15
7 7
∴ c1 = −2/7
Consider T (−2, 1) = (4, 7, 7)
Let (4, 7, 7) = c1 (−1, 0, 2) + c2 (1, 2, 3) + c3 (1, −1, −1)
−c1 + c2 + c3 = 4 ...(1)
2c2 − c3 = 7 ...(2)
2c1 + 3c2 − c3 = 7 ...(3)
Solving (1) and (3) we get
−2c1 + 2c2 + 2c3 = 8
2c1 + 3c2 − c3 = −1
c2 + c3 = 15 ...(4)
Solving (2) and (4) we get
2c2 − c3 = 7
5c2 + c3 = 15
7c2 = 22
∴ c2 = 22/7
(2)⇒ c3 = 2c2 − 7 = 2 22 − 7 = 44−49


7 7
∴ c3 = −5/7
(1)⇒ c1 = c2 + c3 − 4 = 22
7
+ −5
7
− 4 = 22−5−28 ∴ c1 = −11/7
  7
−2/7 4/7 15/7
∴ The matrix is ♠
−11/7 22/7 −5/7

10. Find the matrix of linear transformation T : V2 (R) → V3 (R) defined by T (x, y) =
(x + y, x, 3x − y) w.r.t bases {(1, 1), (3, 1)} and {(1, 1, 1), (1, 1, 0), (1, 0, 0)}.
Solution: Given T : V2 (R) → V3 (R) defined by T (x, y) = (x + y, x, 3x − y)
B1 = {(1, 1), (3, 1)} and B2 = {(1, 1, 1), (1, 1, 0), (1, 0, 0)}.
Consider T (1, 1) = (2, 1, 2)
Let (2, 1, 2) = c1 (1, 1, 1) + c2 (1, 1, 0) + c3 (1, 0, 0)
c1 + c2 + c3 = 2 ...(1)
c1 + c2 = 1 ...(2)
c1 = 2
∴ c1 = 2
(2) ⇒ c2 = 1 − c1 = 1 − 2 ∴ c2 = −1
(1) ⇒ c3 = 2 − c1 − c2 = 2 − 2 + 1 ∴ c3 = 1
Consider T (3, 1) = (4, 3, 8)
Let (4, 3, 8) = c1 (1, 1, 1) + c2 (1, 1, 0) + c3 (1, 0, 0)
c1 + c2 + c3 = 4 ...(1)
c1 + c2 = 3 ...(2)
c1 = 8

62
19
∴ c1 = 8
(2) ⇒ c2 = 3 − c1 = 3 − 8 ∴ c2 = −5
(1) ⇒ c3 = 4 − c1 − c2 = 4 − 8 + 5 ∴ c3 = 1
 
2 −1 1
∴ The matrix is ♠
8 −5 1

63
20
 Exercise

1. Find the matrix of linear transformation defined by T (x, y, z) = (x + y, y + z) w.r.t

standard basis.

2. Find the matrix of linear transformation defined by T (x, y) = (2x + 3y, 4x − 5y) w.r.t
standard basis.

3. Find the matrix of linear transformation defined by T (x, y, z) = (x − y, 2x + z, 2y)

4. Find the matrix of linear transformation defined by T (x, y) = (−x + 2y, y, −3x + 3y)
relative to the basis B1 = {(1, 1), (−1, 1)} and B2 = {(1, 1, 1), (1, −1, 1), (0, 0, 1)}.

Problems
 
1 −1 2
1. For a matrix find the linear transformation w.r.t standard bases.
3 1 0  
1 −1 2
Solution: Given matrix of linear transformation
3 1 0
So there exist a linear transformation T : V2 (R) → V3 (R)
Choose B2 = {(1, 0), (0, 1)} and B1 = {(1, 0, 0), (0, 1, 0), (0, 0, 1)}
Let T (x, y) be the required linear transformation.
Consider T (1, 0) = c1 (1, 0, 0) + c2 (0, 1, 0) + c3 (0, 0, 1)
⇒ T (1, 0) = 1(1, 0, 0) − 1(0, 1, 0) + 2(0, 0, 1) = (1, −1, 2)
Now consider T (0, 1) = c1 (1, 0, 0) + c2 (0, 1, 0) + c3 (0, 0, 1)
⇒ T (0, 1) = 3(1, 0, 0) + 1(0, 1, 0) + 0(0, 0, 1) = (3, 1, 0)
So, we have
T (1, 0) = (1, −1, 2)
T (0, 1) = (3, 1, 0)
∴ T (x, y) = (x + 3y, −x + y, 2x) ♠
 
1 3
2. For a matrix −2 1 find the linear transformation w.r.t standard bases.
2 0  
1 3
Solution: Given, the matrix of linear transformation −2 1
2 0
So there exist a linear transformation T : V3 (R) → V2 (R)
Choose B3 = {(1, 0, 0), (0, 1, 0), (0, 0, 1)} and B2 = {(1, 0), (0, 1)}.
Let T (x, y, z) be the required linear transformation.
Consider T (1, 0, 0) = c1 (1, 0) + c2 (0, 1)
⇒ T (1, 0, 0) = 1(1, 0) + 3(0, 1) = (1, 3)
Now consider T (0, 1, 0) = c1 (1, 0) + c2 (0, 1)
⇒ T (0, 1, 0) = −2(1, 0) + 1(0, 1) = (−2, 1)
Now consider T (0, 0, 1) = c1 (1, 0) + c2 (0, 1)
⇒ T (0, 0, 1) = 2(1, 0) + 0(0, 1) = (2, 0)
So, we have

64
21
 T (1, 0, 0) = (1, 3)
T (0, 1, 0) = (−2, 1)
T (0, 0, 1) = (2, 0)
∴ T (x, y, z) = (x − 2y + 2z, 3x + y) ♠
 
1 2
3. For a matrix find the linear transformation w.r.t {(0, 1), (1, 1)}
3 4  
1 2
Solution: Given, the matrix of linear transformation
3 4
So there exist a linear transformation T : V2 (R) → V2 (R).
Choose B1 = B2 = {(0, 1), (1, 1)}
Let T (x, y) be the required linear transformation.
Consider T (0, 1) = c1 (0, 1) + c2 (1, 1)
⇒ T (0, 1) = 1(0, 1) + 2(1, 1) = (2, 3)
Now consider T (1, 1) = c1 (0, 1) + c2 (1, 1)
⇒ T (1, 1) = 3(0, 1) + 4(1, 1) = (4, 7)
So, we have
T (0, 1) = (2, 3)
T (1, 1) = (4, 7)
Let (x, y) = c1 (0, 1) + c2 (1, 1)
c2 = x
c1 + c2 = y ...(1)
∴ c2 = x
(1) ⇒ c1 = y − c2 ∴ c1 = y − x
Hence, (x, y) = (y − x)(0, 1) + x(1, 1)
Applying T on both sides, we get
T (x, y) = (y − x)T (0, 1) + xT (1, 1)
T (x, y) = (y − x)(2, 3) + x(4, 7)
T (x, y) = (2y − 2x + 4x, 3y − 3x + 7x)
∴ T (x, y) = (2x + 2y, 4x + 3y) ♠
 
−1 0
4. For a matrix  2 0 find the linear transformation w.r.t basis {(1, 0), (2, −1)} and
1 3
{(1, 2, 0), (0, −1, 0), (1, −1, 1)}.  
−1 0
Solution: Given, the matrix of linear transformation  2 0
1 3
So there exist a linear transformation T : V3 (R) → V2 (R)
Choose B1 = {(1, 2, 0), (0, −1, 0), (1, −1, 1)} and B2 = {(1, 0), (2, −1)}
Let T (x, y, z) be the required linear transformation.
Consider T (1, 2, 0) = c1 (1, 0) + c2 (2, −1)
⇒ T (1, 2, 0) = −1(1, 0) + 0(2, −1) = (−1, 0)
Now consider T (0, −1, 0) = c1 (1, 0) + c2 (2, −1)
⇒ T (1, 2, 0) = 2(1, 0) + 0(2, −1) = (2, 0)

65
22
 Now consider T (1, −1, 1) = c1 (1, 0) + c2 (2, −1)
⇒ T (1, −1, 1) = 1(1, 0) + 3(2, −1) = (7, −3)
So, we have
T (1, 2, 0) = (−1, 0)
T (0, −1, 0) = (2, 0)
T (1, −1, 1) = (7, −3)
Consider (x, y, z) = c1 (1, 2, 0) + c2 (0, −1, 0) + c3 (1, −1, 1)
c1 + 2c3 = x ...(1)
2c1 − c2 − c3 − c2 = y ...(2)
c3 = z
∴ c3 = z
(1) ⇒ c1 = x − c3 ∴ c1 = x − z
(2) ⇒ c2 = 2c1 − c3 − y = 2(x − z) − z − y c2 = 2x − y − 3z
Hence, (x, y, z) = (x − z)(1, 2, 0) + (2x − y − 3z)(0, −1, 0) + z(1, −1, 1)
Applying T on both sides, we get
T (x, y, z) = (x − z)T (1, 2, 0) + (2x − y − 3z)T (0, −1, 0) + zT (1, −1, 1)
T (x, y, z) = (x − z)(−1, 0) + (2x − y − 3z)(2, 0) + z(7, −3)
T (x, y, z) = (−x + z + 4x − 2y − 6z + 7z, 0 + 0 − 3z)
∴ T (x, y, z) = (3x − 2y + 2z, −3z) ♠

Exercise
 
1 2
1. For a matrix find the linear transformation w.r.t {(0, 1), (1, 0)}
3 4

Problems 10.3

Find inverse of the following matrices using linear transformation:

 
3 1 2
1. 2 1 2
6 2 5    
3 1 2 A1
Solution: Let A = 2 1 2 = A2 
6 2 5 A3
Then,
3e1 + e2 + 2e3 = A1 ...(1)
2e1 + e2 + 2e3 = A2 ...(2)
6e1 + 2e2 + 5e3 = A3 ...(3)
Solving (1) and (2) we get
3e1 + e2 + 2e3 = A1
−2e1 − e2 − 2e3 = −A2
e1 = A1 − A2
∴ e1 = A1 − A2
Solving (1) and (3) we get

66
23
 Linear Algebra - Linear Transformation
Tejas N S
Assistant Professor in Mathematics, NIE First Grade College, Mysuru
Web: http://nstejas.poeticmath.com
Email: nstejas@gmail.com Mobile: +91 9845410469

14 Rank, nullity and range of a linear transformation

• Definition: Let T : U → V be a linear transformation. The range space of T is the
set {T (α)/α ∈ U } denoted by R(T ). The dimension of R(T ) is called rank.

• Definition: Let T : U → V be a linear transformation. The null space or kernel

of T is the set {α ∈ U/T (α) = 0} denoted by N (T ). The dimension of N (T ) is called
nullity.

Theorems on rank, nullity and range space

• Theorem: If T : U → V is a linear transformation, then the range space of T is a

subspace of V .
Proof: T : U → V is a linear transformation from a vector space U to a vector space
V over a field F . Then range space of T is defined by R(T ) = {T (α)/α ∈ U }.
Clearly, R(T ) is a non empty subset of V .
Let β1 , β2 ∈ R(T ), then ∃ α1 , α2 ∈ U such that T (α1 ) = β1 and T (α2 ) = β2 .
Consider β1 + β2 = T (α1 ) + T (α2 ) = T (α1 + α2 ) ∈ R(T ) ⇒ β1 + β2 ∈ R(T )
Now for any c ∈ F and β1 ∈ R(T ),
Consider cβ = cT (α1 ) = T (cα1 ) ∈ R(T ) ⇒ cβ1 ∈ R(T )
∴ R(T ) is a subspace of V . 

• Theorem: If T : U → V is a linear transformation, then the null space of T is a

subspace of U .
Proof: T : U → V is a linear transformation from a vector space U to a vector space
V over a field F . Then null space of T is defined by N (T ) = {α ∈ U/T (α) = 0}.
Clearly, N (T ) is a non empty subset of U .
Let α1 , α2 ∈ N (T ) ⇒ T (α1 ) = 0 and T (α2 ) = 0
Consider T (α1 + α2 ) = T (α1 ) + T (α2 ) = 0 ⇒ α1 + α2 ∈ N (T )
Now for any c ∈ F and α1 ∈ N (T ),
Consider T (cα1 ) = cT (α1 ) = 0 ⇒ cα1 ∈ N (T )
∴ N (T ) is a subspace of U . 

• Theorem: Let T : U → V be a linear transformation then T is 1-1 if and only if

N (T ) = 0 where 0 is the zero vector of U .
Proof: T : U → V is a linear transformation. Let T be 1-1.
Let α ∈ N (T ) ⇒ T (α) = 0 ⇒ T (α) = T (0) ⇒ α = 0
∴ N (T ) = {0}.

2485
 Conversely let N (T ) = {0}.
Let T (α) = T (β) ⇒ T (α) − T (β) = 0 ⇒ T (α) + T (−β) = 0 ⇒ T (α − β) = 0 ⇒
α − β ∈ N (T ).
Since N (T ) = {0} we have α − β = 0 ⇒ α = β
∴ T (α) = T (β) ⇒ α = β. Hence T is 1-1 

• Theorem: Rank nullity theorem Let T : U → V be a linear transformation then

rank+nullity=dimU .
Proof: T : U → V is a linear transformation.
Let P = {α1 , α2 , ..., αm } be the basis of N (T ). Then dimN (T )=nullity=m
The above basis can be extended to the basis of U .
Let Q = {α1 , α2 , ..., αm , β1 , β2 , ..., βn } be the basis of U . Then dimU = m + n
⇒ dimU =nullity+n
Now we shall prove that {T (β1 ), T (β2 ), ..., T (βn )} is a basis for R(T )
First we shall prove that {T (β1 ), T (β2 ), ..., T (βn )} spans R(T )
Let β ∈ R(T ) ⇒ ∃ α ∈ U such that T (α) = β
Since α ∈ U and Q is a basis of U we have α = a1 α1 + a2 α2 + ... + am αm + b1 β1 + b2 β2 +
... + bn βn where a’s and b’s ∈ F .
Now β = T (α)
⇒ β = T (a1 α1 + a2 α2 + ... + am αm + b1 β1 + b2 β2 + ... + bn βn )
⇒ β = T (a1 α1 ) + T (a2 α2 ) + ... + T (am αm ) + T (b1 β1 ) + T (b2 β2 ) + ... + T (bn βn )
⇒ β = a1 T (α1 ) + a2 T (α2 ) + ... + am T (αm ) + b1 T (β1 ) + b2 T (β2 ) + ... + bn T (βn )
Since α1 , α2 , ..., αm ∈ N (T ) ⇒ T (αi ) = 0 for 1 ≤ i ≤ m
⇒ β = 0 + 0 + ... + 0 + b1 T (β1 ) + b2 T (β2 ) + ... + bn T (βn )
⇒ β = b1 T (β1 ) + b2 T (β2 ) + ... + bn T (βn )
⇒ {T (β1 ), T (β2 ), ..., T (βn )} spans R(T ).
Finally we shall prove that {T (β1 ), T (β2 ), ..., T (βn )} is linearly independent.
Consider c1 T (β1 ) + c2 T (β2 ) + ... + cn T (βn ) = 0
⇒ T (c1 β1 ) + T (c2 β2 ) + ... + T (cn βn ) = 0
⇒ T (c1 β1 + c2 β2 + ... + cn βn ) = 0
⇒ c1 β1 + c2 β2 + ... + cn βn ∈ N (T )
Since P is a basis of N (T )
we have c1 β1 + c2 β2 + ... + cn βn = d1 α1 + d2 α2 + ... + dm αm
⇒ c1 β1 + c2 β2 + ... + cn βn − d1 α1 − d2 α2 − ... − dm αm = 0 Since Q is a basis of N (T )
⇒ c1 = c2 = ... = cn = d1 = d2 = ... = dm = 0
∴ {T (β1 ), T (β2 ), ..., T (βn )} is a basis of R(T ) ⇒ dimR(T )=rank=n
Hence, dimU =nullity+n ⇒ dimU =nullity+rank 

Problems
 
1 3 2
1. Verify rank nullity theorem for the matrix A = −1 7 2.
  1 0 1
1 3 2
Solution: Given A = −1 7
 2
1 0 1
R2 → R2 + R1 ; R3 → R3 − R1

86
25
  
1 3 2
u 0 10 4 
0 −3 −1
∴ Rank=3
Null space, N (T ) = T (x, y, z) = 0
⇒ T (x, y, z) = (x − y + z, 3x + 7y, 2x + 2y + z) = 0
x − y + z = 0 ...(1)
3x + 7y = 0 ...(2)
2x + 2y + z = 0 ...(3)
Solving (1) and (3) we get,
x−y+z =0
−2x − 2y − z = 0
−x − 3y = 0 ...(4)
Solving (2) and (4) we get
3x + 7y = 0
−3x − 9y = 0
−2y = 0
∴y=0
(2) ⇒ 3x = −7y ⇒ x = 0
(1) ⇒ z = y − x ⇒ y = 0
∴ N (T ) = {(0, 0, 0)}
Hence, Nullity=0
∴ Rank+Nullity=3=dim(domain) ♠

2. Verify rank nullity theorem for T (x, y) = (x + y, x).

Solution: Given T (x, y) = (x + y, x)
T (1, 0) = (1, 1)
T (0, 1) = (1, 0)  
1 1
∴ The matrix is
1 0
R2→ R2 −  R1
1 1
u
0 −1
∴ Rank=2
Null space, N (T ) = T (x, y, z) = 0}
⇒ T (x, y) = (x + y, x) = 0
x+y =0
x=0
⇒ x = 0 and y = 0.
∴ N (T ) = {(0, 0)}
Hence, Nullity=0
∴ Rank+Nullity=2=dim(domain) ♠

3. Find range, kernel, rank and nullity of the linear transformation T : R3 → R3 defined
by T (x, y, z) = (x + y, x − y, 2x + z).
Solution: Given, T (x, y, z) = (x + y, x − y, 2x + z)

87
26
 T (1, 0, 0) = (1, 1, 2)
T (0, 1, 0) = (1, −1, 0)
T (0, 0, 1) = (0, 0,1) 
1 1 2
∴ The matrix is 1 −1 0
0 0 1
R2→ R2 − R1 
1 1 2
u 0 −2 −2

0 0 1
∴ Rank=3
Range space, R(T ) = {x(1, 1, 2) + y(0, −2, −2) + z(0, 0, 1)/x, y, z ∈ R}
⇒ R(T ) = {(x, x − 2y, 2x − 2y + z)/x, y, z ∈ R}
Null space, N (T ) = T (x, y, z) = 0
⇒ (x + y, x − y, 2x + z) = 0
x + y = 0 ...(1)
x − y = 0 ...(2)
2x + z = 0 ...(3)
Solving (1) and (2) we get
x+y =0
x−y =0
2x = 0
∴ x = 0.
(2) ⇒ x = y ⇒ y = 0
(3) ⇒ z = −2x ⇒ z = 0
∴ N (T ) = {(0, 0, 0)}
Hence, Nullity=0

4. Find range, kernel, rank and nullity of the linear transformation T (e1 ) = (0, 1, 0, 2),
T (e2 ) = (0, 1, 1, 0) and T (e3 ) = (0, 1, −1, 4)
Solution: Given,
T (e1 ) = (0, 1, 0, 2)
T (e2 ) = (0, 1, 1, 0)
T (e3 ) = (0, 1, −1,4) 
0 1 0 2
∴ The matrix is 0 1 1 0
0 1 −1 4
R2→ R2 − R1 ; R3→ R3 − R1
0 1 0 2
u 0 0 1 −2

0 0 −1 2
R3→ R3 + R2 
0 1 0 2
u 0 0 1 −2

0 0 0 0
Hence, rank=2

88
27
 Range space, R(T ) = {x(0, 1, 0, 2) + y(0, 0, 1, −2) + z(0, 0, 0, 0)/x, y, z ∈ R}
⇒ R(T ) = {(0, x, y, 2x − 2y)/x, y, z ∈ R}
Null space, N (T ) = T (x, y, z) = 0
⇒ (0, x + y + z, y − z, 2x + 4z) = 0
x + y + z = 0 ...(1)
y − z = 0 ...(2)
2x + 4z = 0 ...(3)
(2) ⇒ y = z
Choose z = 1 then y = 1
(1) ⇒ x = −y − z = −2
Equation (3) will satisfy the values of x, y and z.
∴ N (T ) = {(−2, 1, 1)}
Hence, Nullity=1 ♠
5. Find range, kernel, rank and nullity of the linear transformation T (x, y) = (x, x + y, y).
Solution: Given T (x, y) = (x, x + y, y)
T (1, 0) = (1, 1, 0)
T (0, 1) = (0, 1, 1) 
1 1 0
∴ The matrix is
0 1 1
Clearly, rank=2
Range space, R(T ) = {x(1, 1, 0) + y(0, 1, 1)}
⇒ {(x, x + y, y)/x, y ∈ Z}
Null space, N (T ) = T (x, y) = 0
⇒ (x, x + y, y) = 0
⇒ x = 0 and y = 0
∴ N (T ) = {(0, 0)}
Hence, nullity=0 ♠
6. Find range, kernel, rank and nullity of the linear map T : R3 → R2 defined by
T (x, y, z) = (y − x, y − z)
Solution: Given T (x, y, z) = (y − x, y − z)
T (1, 0, 0) = (−1, 0)
T (0, 1, 0) = (1, 1)
T (0, 0, 1) = (0, −1)
 
−1 0
∴ The matrix is  1 1
0 −1
R2↔ R1 
1 1
u −1 0 

0 −1
R2→ R2 +  R1
1 1
u 0 1 
0 −1
R3 → R3 + R2

28
  
1 1
u 0 1 
0 0
Hence, rank=2
Range space, R(T ) = {x(1, 1) + y(0, 1) + z(0, 0)/x, y, z ∈ R}
⇒ R(T ) = {(x, x + y, 0)/x, y ∈ Z}
Null space, N (T ) = T (x, y, z) = 0
⇒ (y − x, y − z) = 0
y − x = 0 ...(1)
y − z = 0 ...(2)
Solving (1) and (2) we get
y−x=0
−y + z = 0
−x + z = 0
∴ x = z. Choose x = 1 then z = 1
(1)⇒ y = x = 1
∴ N (T ) = {(1, 1, 1)}
Nullity=1 ♠

7. Find range, kernel, rank and nullity of T (e1 ) = e1 − e2 , T (e2 ) = 2e1 + e2 and T (e3 ) =
e1 + e2 + e3 .
Solution: Given
T (e1 ) = e1 − e2 = (1, 0, 0) − (0, 1, 0) = (1, −1, 0)
T (e2 ) = 2e1 + e2 = 2(1, 0, 0) + (0, 1, 0) = (2, 1, 0)
T (e3 ) = e1 + e2 +e3 = (1, 0,0) + (0, 1, 0) + (0, 0, 1) = (1, 1, 1)
1 −1 0
∴ The matrix is 2 1 0

1 1 1
R2→ R2 − 2R 1 ; R3 → R3 − R1
1 −1 0
u 0 3 0 
0 2 1
Hence, rank=3
Range space, R(T ) = {x(1, −1, 0) + y(0, 3, 0) + z(0, 2, 1)/x, y, z ∈ R}
⇒ R(T ) = {(x, −x + 3y + 2z, z)/x, y, z ∈ R}
Null space, N (T ) = T (x, y, z) = 0
⇒ (x + 2y + z, −x + y + z, z) = 0
x + 2y + z = 0 ...(1)
−x + y + z = 0 ...(2)
z=0
Solving (1) and (2) we get
x + 2y + z = 0
−x + y + z = 0
3y + 2z = 0
Since z = 0 we have y = 0
(1) ⇒ x = −2y − z = 0

90
29
 ∴ N (T ) = {(0, 0, 0)}
Hence, nullity=0 ♠
8. Find range, kernel, rank and nullity for T (1, 0, 0) = (2, 1), T (0, 1, 0) = (0, 1) and
T (0, 0, 1) = (1, 1).
Solution: Given,
T (1, 0, 0) = (2, 1)
T (0, 1, 0) = (0, 1)
T (0, 0, 1) = (1, 1) 
2 1
∴ The matrix is 0 1
1 1
R3→ 2R3 − R1
2 1
u 0 1
0 1
R3→ R3− R2
2 1
u 0 1

0 0
Hence, rank=2
Range space, R(T ) = {x(2, 1) + y(0, 1) + z(0, 0)/x, y, z ∈ R}
⇒ R(T ) = {(2x, x + y)/x, y ∈ R}
Null space, N (T ) = T (x, y, z) = 0
⇒ (2x + z, x + y + z) = 0
2x + z = 0 ...(1)
x + y + z = 0 ...(2)
Substituting (1) in (2) we get
x + y − 2x = 0 ⇒ −x + y = 0 ⇒ x = y
Choose x = 1 then y = 1
(1) ⇒ z = −2x ⇒ z = −2
∴ N (T ) = {(1, 1, −2)}
Hence, nullity=1 ♠
9. Find the linear transformation T whose range space is spanned by (1, 2, 3) and (4, 5, 6).
Solution: Consider R(T ) =span{(1,2,3) and (4,5,6)}
⇒ R(T ) = {x(1, 2, 3) + y(4, 5, 6)}
⇒ R(T ) = (x + 4y, 2x + 5y, 3x + 6y)
∴ T (x, y, z) = (x + 4y, 2x + 5y, 3x + 6y) ♠
10. Find the linear transformation T whose range space is spanned by (2, 3, −4) and
(3, 4, −2).
Solution: Consider R(T ) =span{(2,3,-4) and (3,4,-2)}
⇒ R(T ) = {x(2, 3, −4) + y(3, 4, −2)}
⇒ R(T ) = (2x + 3y, 3x + 4y, −4x − 2y)
∴ T (x, y, z) = (2x + 3y, 3x + 4y, −4x − 2y) ♠
Exercise

91
30
 Linear Algebra - Linear Transformation
Tejas N S
Assistant Professor in Mathematics, NIE First Grade College, Mysuru
Web: http://nstejas.poeticmath.com
Email: nstejas@gmail.com Mobile: +91 9845410469

12 Eigen values and eigen vectors

• Definition: Let V be a vector space over a field F . Let T : V → V be a linear
transformation and A be the matrix of linear transformation. The eigen equation of
T is defined as the eigen equation of matrix A i.e., |A − λI| = 0. The eigen roots of
T is defined as the roots of the eigen equation of A.
• Definition: Let V be a vector space over a field F . Let T : V → V be a linear
transformation and A be the matrix of linear transformation. Let λ be the eigen
root of T then the non-zero vector which satisfies AX = λX is called eigen vector
corresponding to eigen root λ.

Problems

1. Find the eigen values and eigen vectors of T (x, y) = (2x + 5y, 4x + 3y)
Solution: Given T (x, y) = (2x + 5y, 4x + 3y)
T (1, 0) = (2, 4)
T (0, 1) = (5, 3)  
2 4
Hence, the matrix is A =
5 3
2−λ 4
Consider, |A − λI| = 0 ⇒ =0
5 3−λ
⇒ (2 − λ)(3 − λ) − 20 = 0
⇒ 6 − 2λ − 3λ + λ2 − 20 = 0
The eigen equation of A is λ2 − 5λ − 14 = 0
⇒ λ2 + 2λ − 7λ − 14 = 0
⇒ λ(λ + 2) − 7(λ + 2) = 0
⇒ (λ + 2)(λ − 7) = 0
⇒ Eigen values of A are λ = −2, 7
 
x
Let be the eigen vector such that AX = λX ⇒ AX − λX = 0 ⇒ (A − λ)X = 0
 y  
2−λ 5 x
⇒ =0
4 3−λ y
⇒ (2 − λ)x + 5y = 0
For λ = −2 we have 4x + 5y = 0
Choose y = 4 then, x =  −5
−5
∴ The eigen vector is corresponding to λ = −2
4

31
 For λ = 7 we have −5x + 5y = 0
Choose y = 1 then, x = 1
1
∴ The eigen vector is corresponding to λ = −2 ♠
1

2. Find the eigen values of the transformation T (x, y) = (x + 4y, 2x + 3y)

Solution: Given T (x, y) = (x + 4y, 2x + 3y)
T (1, 0) = (1, 2)
T (0, 1) = (4, 3)  
1 2
∴ The matrix is A =
4 3
1−λ 2
Consider, |A − λI| = 0 ⇒ =0
4 3−λ
⇒ (1 − λ)(3 − λ) − 8 = 0
⇒ 3 − 3λ − λ + λ2 − 8 = 0
The eigen equation of A is λ2 − 4λ − 5 = 0
⇒ λ2 + λ − 5λ − 14 = 0
⇒ λ(λ + 1) − 5(λ + 1) = 0
⇒ (λ + 1)(λ − 5) = 0
⇒ Eigen values of A are λ = −1, 5
 
x
Let be the eigen vector such that AX = λX ⇒ AX − λX = 0 ⇒ (A − λ)X = 0
 y  
1−λ 4 x
⇒ =0
2 3−λ y
⇒ (1 − λ)x + 4y = 0
For λ = −1 we have 2x + 4y = 0
Choose y = 1 then, x =  −2
−2
∴ The eigen vector is corresponding to λ = −1
1
For λ = 5 we have −4x + 4y = 0
Choose y = 1 then, x =  1
1
∴ The eigen vector is corresponding to λ = 5 ♠
1

3. Find the eigen vector of T : V3 (R) → V3 (R) defined by T (1, 0, 0) = (1, 0, 0), T (0, 1, 0) =
(1, 2, 2) and T (0, 0, 1) = (1, 1, 3)
Solution: Given
T (1, 0, 0) = (1, 0, 0)
T (0, 1, 0) = (1, 2, 2)
T (0, 0, 1) = (1, 1, 3)  
1 0 0
∴ The matrix is A = 1 2 2
1 1 3
1−λ 0 0
Consider |A − λI| = 0 ⇒ 1 2−λ 2 =0
1 1 3−λ

77
32
 ⇒ (1 − λ)[(2 − λ)(3 − λ) − 2] + 0 + 0 = 0
⇒ (1 − λ)(6 − 3λ − 2λ + λ2 − 2) = 0
⇒ (1 − λ)(λ2 − 5λ + 4) = 0
The eigen equation of A is (1 − λ)(λ2 − 5λ + 4) = 0
(1 − λ)(λ2 − 5λ + 4) = 0
⇒ (1 − λ) = 0 or λ2 − 5λ + 4 = 0
Consider λ2 − 5λ + 4 = 0
λ2 − λ − 4λ + 4 = 0
λ(λ − 1) − 4(λ − 1) = 0
(λ − 4)(λ − 1) = 0
λ = 1, 4
∴ The eigen values of A are λ = 1, 1, 4
 
x
Let y  be the eigen vector such that AX = λX ⇒ AX − λX = 0 ⇒ (A − λ)X = 0
 z  
1−λ 1 1 x
⇒  0 2 − λ 1   y = 0
0 2 3−λ z
⇒ (1 − λ)x + y + z = 0
For λ = 1 we have 0x + 2y + z = 0
 y = 1 then z = −2. Also choose x = 2
Choose
2
∴ 1  is the eigen vector corresponding to λ = 1

−2
For λ = 1 we have 0x + 2y + z = 0
 y = 2 then z = −4. Also choose x = −2
Choose
−2
∴  2  is the eigen vector corresponding to λ = 1
−4
For λ = 4 we have −3x + y + z = 0 ⇒ 3x = y + z
Choose
  y = 2 and z = 1 then x = 1
1
∴ 2 is the eigen vector corresponding to λ = 4
 ♠
1
 
2 2 1
4. Find the eigen vectors corresponding to the matrix of linear transformation 1 3 1
  1 2 2
2 2 1
Solution: Let A = 1 3 1

1 2 2
2−λ 2 1
Consider |A − λI| = 0 ⇒ 1 3−λ 1 =0
1 2 2−λ
⇒ (2 − λ)[(3 − λ)(2 − λ) − 2] − 2[(2 − λ) − 1] + 1[2 − (3 − λ)] = 0
⇒ (2 − λ)(6 − 2λ − 3λ + λ2 − 2) − 2(2 − λ − 1) + (2 − 3 + λ) = 0

78
33
 ⇒ (2 − λ)(λ2 − 5λ + 4) − 2(1 − λ) + (λ − 1) = 0
⇒ 2λ2 − 10λ + 8 − λ3 + 5λ2 − 4λ − 2 + 2λ + λ − 1 = 0
⇒ −λ3 + 7λ2 − 11λ + 5 = 0
The eigen equation of A is λ3 − 7λ2 + 11λ − 5 = 0
Put λ = 1 in above equation ⇒ 1 − 7 + 11 − 5 = 0
∴ λ = 1 is a root.
By synthetic division,
1 −7 11 − 5
1 1 −6 5
1 −6 5 0
∴ Resultant equation is λ2 − 6λ + 5 = 0
⇒ λ2 − 5λ − λ + 5 = 0
⇒ λ(λ − 5) − (λ − 5) = 0
⇒ (λ − 5)(λ − 1) = 0
⇒ λ = 1, 5
∴ The eigen values of A are λ = 1, 1, 5
 
x
Let y  be the eigen vector such that AX = λX ⇒ AX − λX = 0 ⇒ (A − λ)X = 0
 z  
2−λ 2 1 x
⇒ 1 3−λ 1  y  = 0
1 2 2−λ z
⇒ (2 − λ)x + 2y + z = 0
For λ = 1 we have x + 2y + z = 0 ⇒ x = −2y − z
 y = 1 and z = 2 then x = −4
Choose
−4
∴ 1  is the eigen vector corresponding to λ = 1

2
For λ = 1 we have x + 2y + z = 0 ⇒ x = −2y − z
 y = 1 and z = 1 then x = −3
Choose
−3
∴  1  is the eigen vector corresponding to λ = 1
1
For λ = 5 we have −3x + 2y + z = 0 ⇒ 3x = 2y + z
Choose
  y = 1 and z = 4 then x = 2
2
∴ 1 is the eigen vector corresponding to λ = 5
 ♠
4
 
1 1 2
5. Find the eigen vectors corresponding to the matrix of linear transformation −1 2 1
  0 1 3
1 1 2
Solution: Let A = −1 2 1

0 1 3

34
 1−λ 1 2
Consider |A − λI| = 0 ⇒ −1 2 − λ 1 =0
0 1 3−λ
⇒ (1 − λ)[(2 − λ)(3 − λ) − 1] − 1[−(3 − λ) − 0] + 2[−1 − 0] = 0
⇒ (1 − λ)(λ2 − 5λ + 5) − (−3 + λ) − 2 = 0
⇒ λ2 − 5λ + 5 − λ3 + 5λ2 − 5λ + 3 − λ − 2 = 0
⇒ −λ3 + 6λ2 − 11λ + 6 = 0
The eigen equation of A is λ3 − 6λ2 + 11λ − 6 = 0
Put λ = 1 in above equation ⇒ 1 − 6 + 11 − 6 = 0
∴ λ = 1 is a root.
By synthetic division,
1 −6 11 − 6
1 1 −5 6
1 −5 6 0
∴ Resultant equation is λ2 − 5λ + 6 = 0
⇒ λ2 − 3λ − 2λ + 6 = 0
⇒ λ(λ − 3) − 2(λ − 3) = 0
⇒ (λ − 3)(λ − 2) = 0
⇒ λ = 2, 3
∴ The eigen values of A are λ = 1, 2, 3
 
x
Let y  be the eigen vector such that AX = λX ⇒ AX − λX = 0 ⇒ (A − λ)X = 0
 z  
1−λ 1 2 x
⇒  −1 2 − λ 1  y  = 0
0 1 3−λ z
⇒ (1 − λ)x + y + 2z = 0
For λ = 1 we have x + y + 2z = 0 ⇒ x = y − 2z
 y = 1 and z = 1 then x = −1
Choose
−1
∴ 1  is the eigen vector corresponding to λ = 1

1
For λ = 2 we have −x + y + 2z = 0 ⇒ x = y + 2z
Choose
  y = 1 and z = 2 then x = 5
5
∴ 1 is the eigen vector corresponding to λ = 2
2
For λ = 3 we have −2x + y + 2z = 0 ⇒ 2x = y + 2z
Choose
  y = 2 and z = 1 then x = 2
2
∴ 2 is the eigen vector corresponding to λ = 3
 ♠
1

Exercise

1. Find the eigen values of the transformation T (x, y) = (x + y, y)

80
35
2. Find the eigen space of the transformation T (x, y) = (x + y, y)

3. Find the eigen values and eigen vectors of T (x, y) = (3x + 3y, x + 5y)

4. Find the eigen values and eigen vectors of T (x, y) = (3x + y, 6x + 2y)

5. Find the eigen vector of T : V2 (R) → V2 (R) defined by T (1, 0) = (1, 2) and T (0, 1) =
(4, 3)

6. Find the eigen values and eigen vectors of T : V2 (R) → V2 (R) given by T (e1 ) = (1, 2)
and T (e2 ) = (2, −2).

81
36
Canonical Form: Triangular Form
Every square matrix in 𝑀𝑛 (𝐹) over a field F having all its Characteristic roots in F is similar to
a triangular form by an Unitary transformation 𝑃 ∗ 𝐴𝑃 = 𝑇 , where P is an unitary matrix and T
is a (upper) triangular matrix

37
38