Unit.

2 Linear Transformation
1. Define Linear transformation from a vector space V to another vector space W .
Let V and W be vector spaces over a field F . A function T : V → W is said to be
a linear transformation from V to W if for all x, y  V and c  F , we have
a) T ( x + y ) = T ( x ) + T ( y )
b) T ( cx ) = cT ( x ) .

Note: The above two properties are combined into a single property namely
Linear property T ( x +  y ) = T ( x ) +  T ( y ) or T ( x + y ) = T ( x ) + T ( y ) ,
x, y V and  ,   F .
2. Show that T : R 2 → R 2 defined by T ( a1 , a2 ) = ( 2a1 + a2 , a1 ) is linear.
we know that T is linear if T ( x +  y ) = T ( x ) +  T ( y )
Let x, y  R 2 and  ,   R , where x = ( b1 , b2 ) , y = ( d1 , d 2 )
 x +  y =  ( b1 , b2 ) +  ( d1 , d2 ) = ( b1 +  d1 ,  b2 +  d 2 )
T ( x +  y ) = T ( b1 +  d1 ,  b2 +  d 2 )
= ( 2 ( b1 +  d1 ) +  b2 +  d 2 ,  b1 +  d1 )
= ( 2 b1 + 2 d1 +  b2 +  d 2 ,  b1 +  d1 )
 T ( x ) +  T ( y ) = T ( x ) +  T ( y ) = T ( b1 , b2 ) +  T ( d1 , d 2 )
=  ( 2b1 + b2 , b1 ) +  ( 2d1 + d 2 , d1 )
= ( 2 b1 + 2 d1 +  b2 +  d 2 ,  b1 +  d1 )
T ( x +  y ) = T ( x ) +  T ( y )
 T is linear.
3. Show that T : R3 → R 2 defined by T ( x, y, z ) = ( x , y + z ) is not linear.
we know that T is linear if T ( x +  y ) = T ( x ) +  T ( y )
Let x, y  R3 and  ,   R , where x = ( a1 , a2 , a3 ) , y = ( b1 , b2 , b3 )
 x +  y =  ( a1 , a2 , a3 ) +  ( b1 , b2 , b3 ) = ( a1 +  b1 ,  a2 +  b2 ,  a3 +  b3 )
T ( x +  y ) = T ( a1 +  b1 ,  a2 +  b2 ,  a3 +  b3 )
= (  a1 +  b1 ,  a2 +  b2 +  a3 +  b3 )
T ( x ) +  T ( y ) = T ( a1 , a2 , a3 ) +  T ( b1 , b2 , b3 )
=  ( a1 , a2 + a3 ) +  ( b1 , b2 + b3 )

= ( a1 +  b1 ,  a2 +  a3 +  b2 +  b3 )
T ( x +  y )  T ( x ) +  T ( y ) is not linear.
4. Verify that T : R3 → R and T ( u ) = u is a linear transformation or not.
 T ( u + v ) = u + v  u + v By Triangle inequality
But T ( u + v )  T ( u ) + T ( v )
So, it is not a linear transformation.
5. Obtain the matrix representing the linear transformation T : V3 ( R ) → V3 ( R )
given by T ( a, b, c ) = ( 3a, a − b, 2a + b + c ) with respect to the standard basis
e1 , e2 , e3  .
The standard basis of V3 ( R ) is e1 , e2 , e3  = (1, 0, 0 ) , ( 0,1, 0 ) , ( 0, 0,1) .
T ( e1 ) = T (1, 0, 0 ) = ( 3,1, 2 ) = 3e1 + e2 + 2e3
T ( e2 ) = T ( 0,1, 0 ) = ( 0, −1,1) = 0e1 − e2 + e3
T ( e3 ) = T ( 0, 0,1) = ( 0, 0,1) = 0e1 + 0e2 + e3
 3 0 0
 
The matrix representing T is  1 −1 0  .
2 1 1
 
6. Let T : P3 ( R ) → P2 ( R ) be a linear transformation defined by T ( f ( x ) ) = f  ( x ) .
Let B1 and B2 be the standard bases for P3 ( R ) and P2 ( R ) respectively. Then
find T  .
The usual basis of P3 ( R ) is 1, x, x 2 , x 3 
T ( f ( x )) = f  ( x )
T (1) = 0 = 0.1 + 0.x + 0.x 2
T ( x ) = 1 = 1.1 + 0.x + 0.x 2
T ( x 2 ) = 2 x = 0.1 + 2.x + 0.x 2
T ( x3 ) = 3x 2 = 0.1 + 0.x + 3.x 2
0 1 0 0
The matrix representing T B = 0 0 2 0 
B2
1 .
0 0 0 3
7. For the following linear operator T on a vector space V and ordered basis 
compute T  , V = R 2 , T ( a, b ) = (10a − 6b,17a − 10b ) and  = (1, 2 ) , ( 2,3) .

T (1, 2 ) = ( −2, −3)

Let ( −2, −3) = a (1, 2 ) + b ( 2,3)
( −2, −3) = ( a + 2b, 2a + 3b )
−2 = a + 2b, −3 = 2a + 3b
By solving a = 0, b = −1
Thus T (1, 2 ) = ( −2, −3) = 0 (1, 2 ) − 1( 2,3)
T ( 2,3) = ( 2, 4 )
 ( 2, 4 ) = c (1, 2 ) + d ( 2,3)
( 2, 4 ) = ( c + 2d , 2c + 3d )
2 = c + 2d , 4 = 2c + 3d
By solving d = 0, c = 2
T ( 2,3) = ( 2, 4 ) = 2 (1, 2 ) + 0 ( 2,3)
0 2
T  =  −1 0 
.

8. Find the linear transformation T : V3 ( R ) → V3 ( R ) determined by the matrix
 1 2 1
 0 1 1  with respect to the standard basis e , e , e .
   1 2 3
 −1 3 4 
The standard basis is e1 , e2 , e3 
T ( e1 ) = e1 + 2e2 + e3 = (1, 2,1)
T ( e2 ) = 0e1 + e2 + e3 = ( 0,1,1)
T ( e3 ) = −e1 + 3e2 + 4e3 = ( −1,3, 4 )
Now ( a, b, c ) = a (1,0,0 ) + b ( 0,1,0 ) + c ( 0,0,1)
( a, b, c ) = ae1 + be2 + ce3
T ( a, b, c ) = T ( ae1 + be2 + ce3 )
= aT ( e1 ) + bT ( e2 ) + cT ( e3 )
= a (1, 2,1) + b ( 0,1,1) + c ( −1,3, 4 )
T ( a, b, c ) = ( a − c, 2a + b + 3c, a + b + 4c )
This is the required linear transformation.
9. Consider the basis S = v1 , v2 , v3  for R 3 , where v1 = (1,1,1) , v2 = (1,1,0 ) , v3 = (1,0,0 )
. Let T : R3 → R 2 be the linear transformation such that T ( v1 ) = (1,0 ) ,
T ( v2 ) = ( 2, −1) and T ( v3 ) = ( 4,3) . Find the formulae for T ( x1 , x2 , x3 ) , then use this
formulae to compute T ( 2, −3,5) .
We first express x = ( x1 , x2 , x3 ) as a linear combination of v1 , v2 and v3 .
If we write ( x1 , x2 , x3 ) = av1 + bv2 + cv3
( x1 , x2 , x3 ) = a (1,1,1) + b (1,1, 0 ) + c (1, 0, 0 )
( x1 , x2 , x3 ) = ( a + b + c, a + b, a )
Then on equating corresponding components
a + b + c = x1 , a + b = x2 , a = x3
By solving a = x3 , b = x2 − x3 , c = x1 − x2 .
( x1 , x2 , x3 ) = x3 (1,1,1) + ( x2 − x3 )(1,1,0 ) + ( x1 − x2 )(1,0,0 )
 = x3 v2 + ( x2 − x3 ) v2 + ( x1 − x2 ) v3
Thus T ( x1 , x2 , x3 ) = x3T (1,1,1) + ( x2 − x3 ) T (1,1,0 ) + ( x1 − x2 ) T (1,0,0 )
= x3 (1, 0 ) + ( x2 − x3 )( 2, −1) + ( x1 − x2 )( 4,3)
T ( x1 , x2 , x3 ) = ( 4 x1 − 2 x2 − x3 ,3x1 − 4 x2 + x3 )
From this we obtain T ( 2, −3,5) = ( 9, 2,3) .
10. Define Kernel (Null space) of a linear transformation T .
Let V ( F ) and W ( F ) be two vector spaces and let T : V → W be a linear. We
define the null space or kernel N (T ) of T as N (T ) =  x  V : T ( x ) = 0 .
11. Define Image space (Range space) of a linear transformation T
Let V ( F ) and W ( F ) be two vector spaces and let T : V → W be a linear. We
define the range or image R (T ) of T as Range (T ) = T ( x ) : x  V  .
12. Define Rank.
Let V ( F ) and W ( F ) be two vector spaces and let T : V → W be a linear
transformation then the rank of T is a dimension of the range space of T .
i.e., R (T ) =dimension [Range(T)].
13. Define Nullity.
Let V ( F ) and W ( F ) be two vector spaces and let T : V → W be a linear
transformation then the nullity of T is the dimension of null space of T .
i.e., N (T ) =dimension [Null space of (T)].
14. State and prove Dimension theorem.
Statement: Let V and W be vector spaces, and let T : V → W be linear. If V is
finite-dimensional, then nullity(T) + rank(T) = dim(V).
Proof:
Suppose that dim(V) = n, dim(N(T)) = k, and v1 , v 2 vk  is a basis for N(T).
Then v1 , v 2 vk  is a linearly independent set in V .
Therefore v1 , v 2 vk  can be extended to a basis  = v1 , v2 vn  for V .
We claim that S = T (vk +1 ), T (vk + 2 ), T (vn ) is a basis for R(T).
By known theorem  = v1 , v2 vn  is a basis for V
R(T ) = span(T (v1 ), T (v2 ), T (vn ))
= span T (vk +1 ), T (vk + 2 ), T (vn ) since T ( v1 ) = T ( v2 ) = ... = T ( vk ) = 0
= span( S ) -----(1)
Now we prove that S is linearly independent set.
To prove this let
bk +1T ( vk +1 ) + bk + 2T ( vk + 2 ) + ... + bnT ( vn ) = 0
 T ( bk +1vk +1 + bk + 2vk + 2 + ... + bn vn ) = 0 since T is linear.
 bk +1vk +1 + bk + 2vk + 2 + ... + bn vn  N (T )
  bk +1vk +1 + bk +2vk +2 + ... + bnvn = c1v1 + c2v2 + ... + ck vk
 ( −c1 ) v1 + ( −c2 ) v2 + ... + ( −ck ) vk + bk +1vk +1 + ... + bnvn = 0 ----(2)
Since v1 , v 2 vn  is a linearly independent set
(2)  bi = 0 for all i.
Therefore, S is a linearly independent set ----(3)
(1) and (3) imply that S is a basis for R (T )
Note that T ( vk +1 ) , T ( vk + 2 ) ,..., T ( vn ) are all distinct
Hence dim ( R ( T ) ) = n − k
dim ( R ( T ) ) = dim (V ) − dim ( N ( T ) )
dim ( N ( T ) ) + dim ( R ( T ) ) = dim (V )
Nullity (T ) + Rank (T ) = dim (V )
Hence the proof
15. Let V and W be vector spaces and let T : V → W be linear. Then T is one to one
if and only if N (T ) = 0 .
Assume that T is one-to-one and x  N (T ) .
Since N (T ) is a subspace, 0  N (T ) ----(i)
Let x  N (T ) .
Then T ( x ) = 0 = T ( 0 ) .
Since T is one-to-one, x = 0 .
Therefore, x  0
Hence N (T )  0 ---(ii)
From (i) and (ii)
Hence N (T ) = 0 .
Now assume that N (T ) = 0 , and suppose that T ( x ) = T ( y ) .
Then T ( x − y ) = 0
Therefore x − y  N (T ) = 0
Hence x − y = 0 . i.e., x = y
This means that T is one-to-one.
16. Let V and W be vector spaces of equal (finite) dimension and let T : V → W be
linear. Then the following are equivalent
(i) T is one to one
(ii) T is onto
(iii) Rank(T) = dim(V)
( i )  ( iii )
T is one to one
 N (T ) = 0
 dim ( N (T ) ) = 0
  0 + rank (T ) = dim (V )
 rank (T ) = dim (V )
( iii )  ( ii )
rank (T ) = dim (V )
 rank (T ) = dim (W )
 dim ( R (T ) )
 R (T ) = W
 T is onto
17. Determine whether T : R3 → R 2 defined by T ( a1 , a2 , a3 ) = ( a1 − a2 , 2a3 ) is one to
one or onto.
T is one to one or onto only if N (T ) = 0
N (T ) =  x : T ( x ) = 0
T ( a1 , a2 , a3 ) = 0
( a1 − a2 , 2a3 ) = 0
a1 − a2 = 0, 2a3 = 0
a1 = a2 and a3 = 0
One free variable is there
N (T )  0
Therefore T is not one to one.
18. Let T : R 4 → R3 be the linear mapping defined by
T ( x, y, z, t ) = ( x − y + z + t , x + 2 z − t , x + y + 3z − 3t ) . Find a basis and dimension of
image of T and the kernel of T .
The usual basis of R 4 is (1, 0, 0, 0 ) , ( 0,1, 0, 0 ) , ( 0, 0,1, 0 ) , ( 0, 0, 0,1)
The image of usual basis of R 4 is
T (1, 0, 0, 0 ) = (1,1,1) , T ( 0,1, 0, 0 ) = ( −1, 0,1) ,
T ( 0, 0,1, 0 ) = (1, 2,3) , T ( 0, 0, 0,1) = (1, −1, −3)
The image vectors span Image T .
Hence form the matrix whose rows are these image vectors and row reduce to
echelon form
1 1 1
 −1 0 1 
A= 
1 2 3
 
 1 −1 −3
 1 1 1 
0 1 2  R  R + R
  2 2 1

0 1 2  R3  R3 − R1
 
0 −2 −4 R4  R4 − R1
1 1 1 
0 1 2 
 
0 0 0  R3  R3 − R2
 
0 0 0  R4  R4 + 2 R2
Which is echelon form of matrix
Thus the set of nonzero vectors (1,1,1) , ( 0,1, 2 ) is the basis for Image T .
Hence dim (Image T )=dim ( R (T ) ) =2.
To find the basis and dimension for N (T )
Set T ( v ) = 0 , where v = ( x, y, z, t )
T ( x, y , z , t ) = 0
( x − y + z + t , x + 2 z − t , x + y + 3z − 3t ) = ( 0, 0, 0 )
x − y + z + t = 0

x + 2 z − t = 0  --------(1)
x + y + 3z − 3t = 0 
1 −1 1 1 
Coeffecient matrix = 1 0 2 −1
1 1 3 −3 
1 −1 1 1 
0 1 1 −2  R  R − R
  2 2 1

0 2 2 −4  R3  R3 − R1
1 −1 1 1 
0 1 1 −2 
 
0 0 0 0  R3  R3 − 2 R2
Which is echelon form of matrix
Thus (1) is equivalent to
x− y + z +t = 0
y + z − 2t = 0
The Free variables are z and t . Hence dim (Ker T )=no. of free variables=2
(i) choosing z = −1 , t = 0 we get y = 1, x = 2 .
(ii) choosing z = 0 , t = 1 we get y = 2, x = 1 .
Thus ( 2,1, −1, 0 ) , (1, 2, 0,1) is the basis for Ker T .
19. For the following transformation T : R 2 → R3 defined by
T ( a1 , a2 ) = ( a1 + a2 ,0, 2a1 − a2 ) show that
(i) T is a linear transformation
(ii) Find basis for both N(T) and R(T)
(iii) Compute the nullity and rank(T) and verify the dimension theorem
(iv) Finally use the appropriate theorems to determine whether T is one to
one or onto.
Consider the transformation T : R 2 → R3 defined by
T ( a1 , a2 ) = ( a1 + a2 ,0, 2a1 − a2 )
Let x = ( a1 , a2 ) and y = ( b1 , b2 )  R 2 and  ,   R .
 x +  y = ( a1 +  b1 ,  a2 +  b2 )
T ( x +  y ) = T ( a1 +  b1 ,  a2 +  b2 )
= ( a1 +  b1 +  a2 +  b2 , 0, 2( a1 +  b1 ) − ( a2 +  b2 ))
 T ( x) +  T ( y ) =  (a1 + a2 , 0, 2a1 − a2 ) +  (b1 + b2 , 0, 2b1 − b2 )
= ( a1 +  a2 +  b1 +  b2 , 0, 2 a1 −  a2 + 2 b1 −  b2 )
= ( a1 +  b1 +  a2 +  b2 , 0, 2( a1 +  b1 ) − ( a2 +  b2 )
T ( x +  y ) =  T ( x) +  T ( y )
Hence T is linear .
The usual basis of R 2 is (1, 0 ) , ( 0,1)
The image of usual basis of R 2 is
T (1, 0 ) = (1, 0, 2 ) , T ( 0,1) = (1, 0, −1)
The image vectors span Image T
Hence form the matrix whose rows are these image vectors and row reduce to
echelon form
1 0 2 
A= 
1 0 −1
1 0 2 
0 0 −3 R2 = R2 − R1
 
Which is echelon form of matrix
Thus the set of nonzero vectors (1, 0, 2 ) , ( 0, 0, −3) is the basis for Image T .
Hence dim (Image T )=dim ( R (T ) ) =2.
To find the basis and dimension for N (T )
Set T ( v ) = 0 , where v = ( x, y )
N (T ) =  x  R 2 : T ( x) = 0
T (a1 , a2 ) = 0
(a1 + a2 ,0, 2a1 − a2 ) = (0,0,0)
  a1 + a2 = 0 & 2a1 − a2 = 0
 a2 = −a1 & a2 = 2a1
This is possible only when a1 = a2 = 0
 N (T) = 0
Hence the basis for N (T ) = 
Nullity = dim( N (T )) = 0.
dim(V ) = dim( R2 ) = 2
nullity(T ) + rank(T) = dim(V).
N(T) = 0 if and only if T is one to one.
R(T )  R 3
 it is not onto
20. Prove that the transformation T : M 23 ( F ) → M 22 ( F ) defined by

a a a13   2a11 − a12 a13 + 2a12 

T  11 12 =  
 a21 a22 a23   0 0 
(i). T is linear.
(ii). Find bases for N(T) and R(T).
(iii). Compute the nullity and rank (T) and verify the dimension theorem.
(iv). Use appropriate theorems to determine whether T is one to one or onto.
 a11 a12 a13   2a11 − a12 a13 + 2a12 
Given : T  =  
 a21 a22 a23   0 0 
 a11 a12 a13   b11 b12 b13 
Let , x =   , y =  M 23 ( F ) and  ,   F
 a21 a22 a23   b21 b22 b23 
  a11 +  b11  a12 +  b12  a13 +  b13 
x+ y=  
  a21 +  b21  a22 +  b22  a23 +  b23 
 2 ( a11 +  b11 ) − ( a12 +  b12 ) ( a13 +  b13 ) + 2( a12 +  b12 ) 
T ( x +  y ) =  
 0 0 
 2a11 − a12 a13 + 2a12   2b11 − b12 b13 + 2b12 
Also ,  T ( x ) +  T ( y ) =   +   
 0 0   0 0 
 2 a11 + 2 b11 −  a12 −  b12  a13 +  b13 + 2 a12 + 2 b12 
= 
 0 0 
T ( x +  y ) = T ( x ) +  T ( y )
 T is linear.
(ii) N (T ) =  x  M 23 ( F ); T ( x) = 0
 a11 a12 a13   0 0 
given T  =
 a21 a22 a23   0 0 
 2a11 − a12 a13 + 2a12   0 0 
 = 
 0 0  0 0
2a11 − a12 = 0 ,
−a13
a13 + 2a12 = 0  a12 =
2
The free variables are a13 , a21 , a22 , a23
Therefore dim ( N (T ) ) = 4
Choosing a13 = 4, a21 = 0, a22 = 0, a23 = 0
 −1 −2 4   −2 + 2 4 − 2   0 0
T  =  = 
 0 0 0  0 0  0 0
0 0 0 0 0
a13 = 0, a21 = 1, a22 = 0, a23 = 0  T  =  
1 0 0 0 0
 0 0 0 0 0
a13 = 0, a21 = 0, a22 = 1, a23 = 0  T  =  
 0 1 0 0 0
 0 0 0 0 0
a13 = 0, a21 = 0, a22 = 0, a23 = 1  T  =  
 0 0 1 0 0
 −1 −2 4   0 0 0   0 0 0   0 0 0  
Basis for N(T) =  , , , 
 0 0 0   1 0 0   0 1 0   0 0 1   .
R(T ) = T ( x); x  M 23 ( F )
a a a13   2a11 − a12 a13 + 2a12 
T  11 12 =
 a21 a22 a23   0 0


The usual basis for M 23 ( F ) is e1 , e2 , e3 , e4 , e5 , e6 
 1 0 0   0 1 0   0 0 1   0 0 0   0 0 0   0 0 0  
ie.,  , , , , , 
 0 0 0   0 0 0   0 0 0   1 0 0   0 1 0   0 0 1  
1 0 0  2 0
T ( e1 ) = T  = 
0 0 0  0 0
0 1 0   −1 2
T ( e2 ) = T  = 
0 0 0  0 0
0 0 1 0 1
T ( e3 ) = T  = 
0 0 0 0 0
0 0 0 0 0
T ( e4 ) = T  = 
1 0 0 0 0
0 0 0 0 0
T ( e5 ) = T  = 
0 1 0 0 0
 0 0 0 0 0
T ( e6 ) = T  = 
0 0 1 0 0
2 0 0 0
 −1 2 0 0 

0 1 0 0
A= 
0 0 0 0
0 0 0 0
 
0 0 0 0
2 0 0 0
0 4 0 0 

0 1 0 0
  R3  4 R3 − R2
0 0 0 0
0 0 0 0
 
0 0 0 0
2 0 00
0 4 00 

0 0 00
  R3  4 R3 − R2
0 0 00
0 0 00
 
0 0 00
Which is in row echelon form
 2 0   0 4  
Thus the set of nonzero vectors  ,   is the basis for Image T .
  0 0   0 0 
 1 0   0 1  
ie., Thus the set of nonzero vectors  ,   is the basis for Image T .
 0 0   0 0 
Hence dim (Image T )=dim ( R (T ) ) =2.

(iii) Nullity of T = dim (N(T)) = 4

Rank of T = dim(R(T)) = 2
Dim (M2x3(F)) = 6
Nullity(T) + Rank (T) = dim (M2x3(F))
Dimension theorem is verified.
(iv) Since N(T)  {0} ,by theorem
Let V and W be vector spaces and let T : V → W be linear.
Then T is one to one if and only if N (T ) = 0

T is not one – one and not onto.

21. Definition: Inner Product space.
 Let V be a vector space over F . An inner product on V is a function that assigns, to
every ordered pair of vectors x and y in V a scalar in F denoted by x, y , such
that for all x, y and z in V and all c  F the following hold
i). x + z, y = x, y + z, y
ii). cx, y = c x, y
iii). x, y = y, x
iv). x, x  0 if x  0
22. In an Euclidean inner product find cosine of the angle between the vectors u = ( 2,3,5 )

and v = (1, −4,3)

u, v
cos  =
u v
u, v = 2 − 12 + 15 = 5 u = 4 + 9 + 25 = 38, v = 1 + 16 + 9 = 26
,
u, v 5
cos  = =
u v 38 26 .
23. Compute the angle between two vectors ( x, y ) and ( − y, x ) in an Euclidean inner
product space R 2 .
Let u = ( x, y ) and v = ( − y, x )
u, v
cos  =
u v

u, v = − xy + xy = 0 , u = x 2 + y 2 , v = x 2 + y 2
u, v 0
cos  = = 2 =0
u v x + y2

 = .
2
24. Let V be the set off all continuous real valued functions defined on the closed interval
1

0,1 , then prove that V is a real inner product f , g =  f ( t ) g ( t ) dt .

0

Proof:
Let f , g , h V and  R
1
(i) f + g , h =   f ( t ) + g ( t )  h ( t ) dt
0
1 1
=  f ( t ) h ( t ) dt +  g ( t ) h ( t ) dt = f , h + g , h
0 0
1
(ii)  f , g =  f ( t ) g ( t ) dt
0
 1
=   f ( t ) g ( t ) dt =  f , g
0
1
(iii) f , g =  f ( t ) g ( t ) dt
0
1
=  g ( t ) f ( t ) dt = g , f .
0
1
f , f =   f ( t )  dt  0 and f , f = 0 iff f = 0 .
2
(iv)
0

25. Let V = M nn ( F ) and define for A, B in V , A, B = tr B  A Then Verify that this is ( )
an inner product
Solution:
For A, B, C V and a  F
(
a). A + B, C = tr C  ( A + B ) )
= tr ( C A + C B )
 

= tr ( C A ) + tr ( C B )
 

= A, C + B, C

( )  ( B A)
n
b). A, B = tr B A = 
ii
i =1
n
 n 
=    ( B ) ik ( A )ki 


i =1  k =1 
n n n n
=  ( B )ki ( A)ki =  ( A)ki ( B )ki
i =1 k =1 i =1 k =1

( ) ( B) = ( A B)
n n n
=  A ki

ik ii
i =1 k =1 i =1

(
= tr A B = B, A )
(
c). aA, B = tr BaA = atr B  A = a A, B ) ( )
( )  ( A A)
n
 
d). A, A = tr A A =
ii
i =1

 n 
n n

( ) ( A)ki  =  Aki Aki

n
=    A
i =1  k =1  i =1 k =1
ik

n n
=  Aki
2

i =1 k =1

If A  0 , Aki 0 for some k and i

n n
Therfore, A  0 then A, A =  A 0
2
ki
i =1 k =1

Hence A, B is an inner product on V .

26. Let V be an inner product space. Then for x, y, z  V and c  F the following
statements are true a). x, y + z = x, y + x, z
b). x, cy = c x, y
c). x,0 = 0, x = 0

d). x, x = 0 if and only if x = 0

e). x, y = x, z for all x V , then y = z
Proof:
a). x, y + z = y + z , x
= y, x + z , x
= y, x + z , x
= x, y + x, z
b). x, cy = cy, x = c y, x = c x, y
c). x, 0 = x, a + ( −a ) = x, a + x, −a
= x, a − x, a = 0 where a V
Similarly 0, x = 0
d). Assume that x, x = 0 ----(1)
Then x = 0 . Otherwise (i.e. if x  0 ), x, x  0 contradicts (1)
Assume that x = 0 . Then by (c)
x, x = 0, 0 = 0
e). x, y = x, z . Then x, y − x, z = 0
 x, y − z = 0, x V
Taking x = y − z, y − z , y − z = 0
By (d), y − z = 0  y = z .
27. Let S consist of the following vectors in R 4 : u1 = (1,1, 0, −1) , u2 = (1, 2,1,3) ,
u3 = (1,1, −9, 2 ) , u4 = (16, −3,1,3) . Find the coordinates of any arbitrary vector
v = ( a, b, c, d ) in R 4 relative to basis S
Solution:
u1.u2 = 1 + 2 + 0 − 3 = 0

u1.u2 = 1 + 1 + 0 − 2 = 0

u1.u4 = 16 − 13 + 0 − 3 = 0

u2 .u3 = 1 + 2 − 9 + 6 = 0
u2 .u4 = 16 − 26 + 1 + 9 = 0

u3 .u4 = 16 − 13 − 9 + 6 = 0
 Thus S is orthogonal and s is linearly independent. Accordingly S is a basis for R 4 because
any four linearly independent vectors form a basis of R 4
v, u1 a+b−d
k1 = =
u1 , u1 3

v, u 2 a + 2b + c + 3d
k2 = =
u2 , u2 15

v, u3 a + b − 9c + 2d
k3 = =
u3 , u3 87

v, u 4 16a − 13b + c + 3d
k4 = =
u4 , u4 435
28. Define norm of a vector in an inner product space.
Let V in an inner product space. For x V , we define the norm or length of x by
x = x, x
29. Find norm and distance between the vectors u = (1, 0,1) and v = ( −1,1,0 ) .

u = 12 + 02 + 12 = 2, v = ( −1) + 12 + 02 = 2
2 2 2

The distance between two vectors u and v is defined by d ( u, v ) = u − v

u −v = u − v, u − v = u , u − u , v − v , u + v , v
2

u −v = 2 +1+1+ 2 = 6
2

u −v = 6
30. Consider f ( t ) = 3t − 5 and g ( t ) = t in the polynomial space P (t ) with inner product
2

1
f , g =  f ( t ) g ( t ) dt then find f and g .
0

f ( t ) f ( t ) = 9t 2 − 30t + 25 , g ( t ) g ( t ) = t 4
1
= f , f =  ( 9t 2 − 30t + 25 ) dt = 3t 3 − 15t 2 + 25t  = 13
2 1
f
0
0
1
t5  1
1
g = g , g =  t dt =   =
2 4

0  5 0 5
31. Let V be an inner product space over F . Then for all x, y  V and c  F , Then
the following statements are true
1. cX = c X
2. x = 0 if and only if x = 0 .
3. Cauchy Schwarz Inequality x, y  x y
4. Triangle Inequality x + y  x + y
 Solution:

= cx, cx = cc x, x = c
2 2 2
1. cx x

2. x = 0  x = 0  x, x = 0  x = 0
2

3. case(i). let y = 0 . Then x, y = x, 0 = 0

and x y = x 0 = 0 .

Therefore x, y = x y

Case(ii). Let y  0 . For any c  F , we have

0  x − cy = x − cy, x − cy
2

= x, x − x, cy − cy, x + cy, cy

= x − c x, y − c x, y + cc y , y
2

x, y
Take c = 2
.
y
2
x, y x, y x, y
Then 0  x − x, y − x, y +
2 2
2 2 4
y
y y y
2 2 2
0 x y − x, y − x, y + x, y
2 2

2
 x, y  x
2 2
y

 x, y  x y

4. we have x + y = x + y, x + y = x, x + x, y + y, x + y, y
2

x + y = x + x, y + x, y + y
2 2 2

x + y = x + 2 Re x, y + y
2 2 2

x + y  x + 2 x, y + y
2 2 2

x+ y ( x + y )
2 2

x+ y  x + y
32. Using Euclidean inner product on R 3 show that u = ( −3,1, 0 ) and v = ( 2, −1,3)
satisfy Cauchy Schwartz inequality.
Solution:
 x = x, x = 9 + 1 + 0 = 10
2

x = 10

y = y, y = 4 + 1 + 9 = 15
2

y = 15

x y = 10 15 = 150

x, y = −6 − 1 = −7

x, y = 7

 x, y  x y
33. Definition: Orthogonal
A vector u V is said to be orthogonal to v V if u, v = 0 .
Definition: Orthogonal Subspaces
Two subspaces V & W of a vector space are orthogonal if every vector v in
V is perpendicular to every vector w in W . ie. v, w = 0, v V , w W .
Definition: Orthonormal Set
A subset S of V is called an orthonormal set if
i). x = 1, x  X
ii). x, y = 0 x, y  S
Definition: Orthonormal Basis
A basis of an inner product space that consists of mutually orthogonal unit vectors is
called an Orthonormal basis.
34. Find the value of a if the vectors ( 2, a ) and ( 6, 4 ) are orthogonal vectors in R 2 .

Let u = ( 2, a ) and v = ( 6, 4 )
Since the vectors are orthogonal u, v = 0
12 + 4a = 0
4a = −12  a = −3
35. Find k so that u = (1, 2, k ,3) and v = ( 3, k , 7, −5 ) in R 4 are orthogonal.

u, v = (1, 2, k ,3) . ( 3, k ,7, −5 ) = 3 + 2k + 7k − 15 = 9k − 12

Then set u, v = 9k − 12 = 0
4
k = .
3
36.  
Let S = (1,1,0 ) , (1, −1,1) , ( −1,1, 2 ) be an orthogonal set then orthonormal set is

 1 1 1 
 (1,1, 0 ) , (1, −1,1) , ( −1,1, 2 ) both are basis of R 3 . Let X = ( 2,1,3)  R3 .
 2 3 6 
Express X as a linear combination of orthogonal set S and orthonormal set.
 Let v1 , v2 , v3  be an orthonormal basis of V . Then for any x V ,
3
x, vi
x= 2
vi
i =1 vi
x, v1 x, v2 x, v3
x= 2
v1 + 2
v2 + 2
v3
v1 v2 v3
i.e. x as a linear combination at the basis vectors are
x, v1 1 3
2
= ( 2 + 1) =
v1 2 2
x, v2 1 4
2
= ( 2 − 1 + 3) =
v2 3 3
x, v3 1 5
2
= ( −2 + 1 + 6 ) =
v3 6 6
3 4 5
i.e ( 2,1,3) = (1,1, 0 ) + (1, −1,1) + ( −1,1, 2 ) is a linear combination of orthogonal set.
2 3 6
37. If v = (1, 2,1) and u = ( 2,1, 2 ) find proj ( v, u ) .
v, u
Proj ( v, u ) = u
u, u

Proj ( v, u ) = ( 2,1, 2 ) =  , , 
6 4 2 4
9 3 3 3
38. Prove that in an inner product space V , for any u, v V ,
u +v + u−v =2 u +2 v .
2 2 2 2

u +v + u −v = u + v, u + v + u − v , u − v
2 2

= u, u + u, v + v, u + v, v + u, u − u, v − v, u + v, v
= u + v + u + v =2 u +2 v
2 2 2 2 2 2

39. Suppose u, v & w are vectors in an inner product space such that u, v = 2 , u, w = −3
, v, w = 5 , u = 1 , v = 2 , w = 7 .

Evaluate (i). u + v, v + w

(ii). 2u − w,3u + 2w

(i). u + v, v + w = u, v + u, w + v, v + v, w
= 2 −3+ 4 +5 = 8
(ii). 2u − w,3u + 2w = 6 u, u + 4 u, w − 3 w, u − 2 w, w
= 6 u + 4 ( −3) − 3 ( −3) − 2 w = 6 − 12 + 9 − 2(49) = −95
2 2

40. If u and v are orthonormal vectors in an inner product space V then find u + v .
 u + v = u + v, u + v = u , u + u , v + v, u + v, v
2

= u + 2 u, v + v since vectors are orthonormal u, v = 0

2 2

= u + v
2 2

41. In Euclidean inner product space R 2 verify u, v + w = u, v + u, w for the vectors

u = ( 3, −2 ) , v = ( 4,5 ) and w = ( −1,6 ) .

u, v + w = u, v + u, w
u, v = 12 − 10 = 2, u, w = −3 − 12 = −15
RHS = u, v + u, w = 2 − 15 = −13
v + w = ( 3,11)
LHS = u, v + w = 9 − 22 = −13
Hence verified
42. Find the norm of the vector u = (1,1, −1) and v = ( −1,1,0 ) in R 3 with respect to the

inner product defined by u, v = u1v1 + 2u2v2 + 3u3v3 where u = ( u1 , u2 , u3 ) and

v = ( v1 , v2 , v3 )
= 1 + 2 + 3 = 6, v = 1+ 2 + 0 = 3
2 2
u
 u = 6, v = 3
43. Let V be an inner product space and S = v1 , v2 , v3 ,...vk  be an orthogonal subset of V
k y, vi
consisting of non-zero vectors. If y  span( s) then y = i =1 vi
2
vi

Proof:
Let y  span( s) and S = v1 , v2 , v3 ,...vk 
 there exists scalars a1 , a2 , a3 ,...ak such that
k
y = a1v1 + a2v2 + a3v3 + ... + ak vk =  ai vi − − − − − (1)
i =1
For 1  j  k ,
k
y, v j = a v ,v
i =1
i i j = a1 v1 , v j + ... + a j v j , v j + ... + an vn , v j
2
y, vi = a j v j since S is an orthogonal set

y, v j
 aj = 2
− − − − − (2)
vj
k y, vi
Using (2) in (1) y = 
i =1 vi
2
vi .

44. (Gram – Schmidt orthogonalization process)

 Let V be an inner product space and S = {w1 , w2 ,..., wn } be a linearly independent subset
k −1  wk , v j 
of V . Define S ' = {v1 , v2 ,..., vn } where v1 = w1 and vk = wk − j =1 || v j ||2
v j , for

2 k  n. Then S  is an orthogonal set of non-zero vectors such that

span ( S  ) = span ( S ) .
Proof:
The proof is by induction on n .
Let Sk = {w1 , w2 ,..., wk }, for k = 1, 2,3,..., n
Since {w1} is linearly independent, v1 = w1  0 .
Clearly, {v1} is orthogonal and span ( v1 ) = span ( w1 ) .
Therefore the theorem is valid for n = 1 .
Assume that the theorem is valid for n = k − 1 .
i.e., {v1 , v2 ,..., vk −1} is an orthogonal set of non-zero vectors and
span ( v1 , v2 ,..., vk −1 ) = span ( w1 , w2 ,..., wk −1 ) -----------(1)
Now, we prove that the theorem is valid for n = k .
k −1  wk , v j 
vk = wk −  v j  vk  0 − − − − − − − −(2)
j =1 || v j ||2
Now for m  k − 1 ,
k −1 wk , v j
vk , vm = wk −  2
v j , vm
j =1 vj

k −1 wk , v j
vk , vm = wk , vm −  2
v j , vm
j =1 vj
wk , vm
vk , vm = wk , vm − 2
vm , vm since v1 , v2 ,..., vk −1 is orthogonal.
vm
wk , vm
vk , vm = wk , vm − =0
2
2
vm
vm
 {v1 , v2 ,..., vk } is an orthogonal set of non-zero vectors.
Further from (1)
span ( v1 , v2 ,..., vk −1 , vk ) = span ( w1 , w2 ,..., wk −1 , vk )
= span ( w1 , w2 ,..., wk −1 , wk ) by (2)
Therefore the theorem is true for n = k .
Hence by induction the theorem is valid for all n  1.
45. Let R 3 have the Euclidean inner product. Use the Gram-Schmidt process to
convert basis B = u1 , u2 , u3  where u1 = (1,0,1) , u2 = ( −1,1,0 ) , u3 = ( −3, 2,0 ) into an
orthogonal basis.
Solution:
Let w1 = (1,0,1) , w2 = ( −1,1,0 ) , w3 = ( −3, 2,0 )
v1 = w1 = (1,0,1)
 1 w2 , v j
v2 = w2 −  2
vj
j =1 vj
w2 , v1
= w2 − 2
v1
v1
( −1,1, 0 )(1, 0,1)
T

= ( −1,1, 0 ) − (1, 0,1)

2
1
= ( −1,1, 0 ) + (1, 0,1)
2
 1 1
v2 =  − ,1, 
 2 2
2 w3 , v j
v3 = w3 −  2
vj
j =1 vj

 w ,v w ,v 

= ( −3, 2, 0 ) −  3 21 v1 + 3 22 v2 

 v1 v2 

  1 1
T

 −3, 2, 0 1, 0,1 ( − 3, 2, 0 )  − ,1,  
( )( ) 1, 0,1 +
T

= ( −3, 2, 0 ) −   2 2   − 1 ,1, 1  
( )  
 (1 + 1) 1 1  2 2 
  +1+  
4 4
3 7  1 1
= ( −3, 2, 0 ) + (1, 0,1) −  − ,1, 
2 6 2 2
2 
4
 3 3 7 1 1
=  − , 2,  −  − ,1, 
 2 2 3 2 2
 1 1 1
v3 =  − , − ,  .
 3 3 3
  1 1   1 1 1 
Therefore v1 , v2 , v3  = (1, 0,1) ,  − ,1,  ,  − , − ,   is an orthogonal basis.
  2 2   3 3 3 
 v v v   1 1 1   1 2 1   1 1 1  
  1 , 2 , 3  =  , ,  ,  − , ,  ,  − ,− ,   is an
 v1 v2 v3   2 2 2   6 3 6  3 3 3  
orthonormal basis.
46. 3
Let R have the Euclidean inner product. Use the Gram-Schmidt process to transform the
basis u1 , u2 , u3  into an orthonormal basis, where u1 = (1,1,1) , u2 = ( 0,1,1) , u3 = ( 0,0,1) .
Solution:
Let w1 = (1,1,1) , w2 = ( 0,1,1) , w3 = ( 0, 0,1)
Let v1 = w1 = (1,1,1)
 1 w2 , v j
v2 = w2 −  2
vj
j =1 vj
w2 , v1
= w2 − 2
v1
v1
( 0,1,1)(1,1,1)
T

= ( 0,1,1) − (1,1,1)
3
2
= ( 0,1,1) − (1,1,1)
3
 2 1 1
= − , , 
 3 3 3
2 w3 , v j
v3 = w3 −  2
vj
j =1 vj

 w ,v w ,v 

= ( 0, 0,1) −  3 21 v1 + 3 22 v2 

 v1 v2 

  2 1 1
T

 0, 0,1 1,1,1 T ( 0, 0,1)  − , ,  
( )( )  3 3 3   2 1 1 
= ( 0, 0,1) −  (1,1,1) +  − , , 
 (1 + 1 + 1)  4 1
+ + 
1   3 3 3 
  
9 9 9
1 1  2 1 1
= ( 0, 0,1) − (1,1,1) − − , , 
3  6 3 3 3
3 
9
 1 1 2 1 2 1 1
= − ,− , − − , , 
 3 3 3 2 3 3 3
 1 1 2 1 1 1
= − ,− , + ,− ,− 
 3 3 3 3 6 6
 1 1
v3 =  0, − ,  .
 2 2
  2 1 1  1 1 
Therefore v1 , v2 , v3  = (1,1,1) ,  − , ,  ,  0, − ,   is an orthogonal basis.
  3 3 3  2 2 
 v v v   1 1 1   2 1 1   1 1  
  1 , 2 , 3  =  , ,  ,  − , ,  ,  0, − ,   is an
 v1 v2 v3   3 3 3   3 6 6   2 2  
orthonormal basis.
1
47. Let the vector space P2 have the inner product p, q =  p ( x ) q ( x ) dx . Apply the Gram-
0

 
schmidt process to transform the basis S = u1 , u2 , u3  = 1, x, x 2 into an orthonormal
basis.
Solution:
Let w1 = 1, w2 = x, w3 = x 2

v1 = w1 = 1
w2 , v1 x, v1
v2 = w2 − 2
v1 = x − 2
v1
v1 v1
1
 x2 
1
x,1 =  xdx =   = 0
−1  2 −1
1
v1 = v1 , v1 =  dx = 2
2

−1

v2 = x − 0 = x
w3 , v1 w3 , v2
v3 = w3 − 2
v1 − 2
v2
v1 v2

x 2 , v1 x 2 , v2
v3 = x − 2
2
v1 − 2
v2
v1 v2
1
 x3  2
1
w3 , v1 = x ,1 =  x dx =   =
2 2

−1  3 −1 3
1
w3 , v2 = x , x =  x 3dx = 0
2

−1

2
1
v3 = x − 3 − 0 = x 2 −
2

2 3
2
 1
1
= v3 , v3 =   x 2 −  dx
2
v3
−1 
3
1
 1
1 2 x2   x5 x 2 x3 
=   x4 + −  dx =  + − 
−1    5 9 9  −1
9 3

1 1 2 1 1 2 
= + − + + − 
5 9 9 5 9 9 
8
=
2
v3 .
45
 1
Therfor 1, x, x 2 −  is an orthogonal basis for P2 ( R )
 3
 v v v   1 3 3 5  2 1  
 1 , 2 , 3  =  , x,  x −   is an orthonormal basis for P2 ( R ) .
 v1 v2 v3   2 2 2 2  3  
48. Let V be the vector space of polynomials f ( t ) with inner product
1
f ,g =
−1
 f ( t ) g ( t ) dt . Apply the Gram-schmidt orthogonalization process to
1, t , t 2
, t  to find an orthonormal basis  f , f , f , f  with integer coefficients for
3
0 1 3 4

P3 ( t ) .
Solution:
Let w1 = 1, w2 = t , w3 = t 2

v1 = w1 = 1

w2 , v1
v2 = w2 − 2
v1 ,
v1
1
 t2 
1
w2 , v1 =  tdt =   = 0
−1  2 −1
1
v1 = v1 , v1 =  dt = 2
2

−1

v2 = t − 0 = t

w3 , v1 w3 , v2
v3 = w3 − 2
v1 − 2
v2
v1 v2
1
 t3 
1
2
w3 , v1 =  t dt =   = 2

−1  3 −1 3
1
t4 
1
w3 , v2 =  t dt =   = 0 3

−1  4  −1
1
1
 t3  1 1 2
1
= v2 , v2 =  [Link] =  t dt =   = + =
2 2
v2
−1 −1  3  −1 3 3 3
2
1
v3 = t − 3 − 0 = t 2 −
2

2 3
w4 , v1 w4 , v2 w4 , v3
v4 = w4 − 2
v1 − 2
v2 − 2
v3
v1 v2 v3
1
t4 
1
1 1
w4 , v1 =  t dt =   = − = 0
3

−1  4  −1 4 4
1
 t5 
1
1 1 2
w4 , v2 =  t dt =   = + =
4

−1  5  −1 5 5 5
 1
 1
1
 t3 
w4 , v3 =  t 3  t 2 −  dt =   t 5 −  dt
−1  3 −1 
3
1
t6 t4  1 1 1 1 
w4 , v3 =  +  =  + − −  = 0
 6 12  −1  6 12 6 12 
2
 1
1
v3 = v3 , v3 =   t 2 −  dt
2

−1 
3
1
 4 1 2t 2 
1
 t 5 t 2t 3 
=  t + −  dt =  + − 
−1 
9 3   5 9 9  −1
1 1 2 1 1 2 
= + − + + − 
5 9 9 5 9 9 
8
=
2
v3
45
2
 3 
1
= v4 , v4 =   t 3 − t  dt
2
v4
−1 
5 
2
 3   6 
1 1
9
=   t 3 − t  dt =   t 6 + t 2 − t 4  dt
2
v4
−1 
5  −1 
25 5 
1
 t7 9 t3 6 t5  24
= + −  =
 7 25 3 5 5  −1 525
2 3 3
v4 = t 3 −  t = t 3 − t .
5 2 5
 1 3 
Therefore v1 , v2 , v3 , v4  = 1, t , t 2 − , t 3 − t  is an orthogonal basis.
 3 5 

 v1 v2 v3 v4   1 3 3 5  2 1  525  3 3  
Now  , , , =  , t, t − ,  t − t   is an
 v1 v2 v3 v4   2 2 2 2  3  24  5  
orthonormal basis for P3 ( R ) .