Module 2: Vector Spaces

n n
A real coordinate space of dimension n (we write it as R or , where n is a natural
number) is a coordinate space over the real numbers. This means that it is the set of the n-
tuples of real numbers (sequences of n real numbers).
For example,
 R1 is the real line with points as a real numbers on it (R1: set of all real number).
 R2 is a plane with points as ordered pair of real numbers  x1 , x2  (R2: set of all ordered
pair of real numbers).
 R3 is of three-dimensional space with points as ordered triple of real numbers  x1 , x2 , x3 
(R3: set of all ordered triple of real numbers).
 R4 is a four-dimensional space with points as ordered quadruple of real numbers
 x1 , x2 , x3 , x4  (R4: set of all ordered quadruple of real numbers).
 Rn is an n-dimensional space with points as ordered n-tuple of real numbers
 x1 , x2 , x3 ...., xn  (Rn: set of all ordered n-tuple of real numbers).

What is point? A point is a location in a plane that has no size, i.e. no width, no length and
no depth.

What is Vector? A Vector is a point in n-dimensional space that has both magnitude and
direction.

What is Scalar? A Scalar is a quantity (number) that has only magnitude but not direction.
Dr. PAR VEZ ALAM
 Standard operations in Rn
The sum of two vectors and the scalar multiple of a vector in Rn are called the standard
operations in Rn.

 We say that two vectors are equal;

 Some important operations and vectors:

For example:
Dr. PAR VEZ ALAM

Let u=(2, 3, 1, 5) and v=(-2, -3, -1, -5); here v is additive inverse of u.

If we add zero vector(additive identity) 0=(0, 0, 0, 0) in u it will be same, i.e. 0+u=u+0=u

(0+2, 0+3, 0+1, 0+5)= (2+0, 3+0, 1+0, 5+0)= (2, 3, 1, 5)=u.
Dr. PAR VEZ ALAM
 Note:

The matrix operations of addition and scalar multiplication give the same results as the
corresponding vector operations.
Dr. PAR VEZ ALAM
 Binary operations/compositions:
Before giving the definition of vector space, we define what is an internal and external
composition?

Let G be a non-empty set, then a mapping : G  G  G is called an internal composition

in G. ( g1, g2 )  g3  G
Let G and F be two non-empty sets, then a mapping : G  F  F is called an external
composition in G. ( g1, f1 )  f 2  F

Group:
A group is a set G, together with a binary operation ∗, which satisfies the following axioms;

G1 closure: for all elements g and h of G, g ∗ h is an element of G;

G2 associativity: (g ∗ h) ∗ k = g ∗ (h ∗ k) for all g, h, k ∈ G;

G3 existence of identity: there exists an element e ∈ G, called the identity of G, such that
e ∗ g = g ∗ e = g for all g ∈ G;
G4 existence of inverse: for every g ∈G there exists an element h ∈ G, called the inverse of g,
such that g ∗ h = h ∗ g = e.

The group (G, ∗) is said to be commutative (or Abelian) if it satisfies one more additional
axiom:

G5 commutativity: g ∗ h = h ∗ g for all g, h ∈ G.

Example: Real numbers R with addition  ,  is an Abelian Group

Dr. PAR VEZ ALAM
 Vector Space
A non-empty set V is said to be a vector space over (field) if there exist two maps
 : V×V  V defined by  v1 , v2  v1  v2 called addition, and  : ×V  V defined by
 , v    v called scalar multiplication, satisfying following properties
(w.r.t. addition  : V×V  V )
V1 Closure: u  v V, for all elements u , v of V

V2 Commutativity: uvvu, for all elements u , v of V

V3 Associativity:  u  v   w  u   v  w , for all elements u , v, w of V

V4 Existence of There exists an 0 ∈ V, such that 0 v  v , for all elements v of V

additive identity:

V5 Existence of For every vV, there exists an u ∈ V, such that v  u  0  u  v .

additive inverse: This u is denoted by v .

*It means V is Abelian w.r.t. ‘+’ addition.

(w.r.t. scalar multiplication  : ×V  V )

V6 Closure w.r.t.   v V, for all elements v of V and scalar  
scalar multiplication

V7 Associative w.r.t.     v        v , for all elements v of V and scalar , 

scalar multiplication

V8 Distributive    u  v     u    v , for all elements u , v of V and scalar  

V9 Distributive      v    v    v , for all elements v of V and scalar , 

V10 Action of 1 1  v  v , for all elements v of V

Note:

 Usually we are writing a vector space as V , , ,   ,  and  are two defined

operations on which set V satisfy all above 10 axioms. These two operations may be
Dr. PAR VEZ ALAM

different for other vectors spaces.

 The elements (u,v,w,……. or x,y,z……., etc.) of a vector space V are called vectors.
 If any set that not satisfying any one condition then the set will not be a vector space.
 V  0; is a vector space and we say it zero vector space. (Check all conditions of VS
Exercise)
 V  is a vector space over itself . (Check all conditions of VS Exercise)
 Question 1: Show that V 
2
  x   x1, x2  ; x1, x2   is a vector space. V.S.  2
, , , 
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
 Question 2: Show that V 
n
 x   x1, x2 , x1......xn ; x1, x2 , x1......xn   is a vector space.
2 n
Similar to the , we can show that is also a vector space.

Hint: We take 3 vectors x   x1 , x2 , x1......xn  , y   y1, y2 , y1...... yn  & z   z1, z2 , z1......zn  and 2
scalars ,  .


V.S.  n
, , , 
Dr. PAR VEZ ALAM
 Question 3: Show that set of all 2x2 square real matrices
u u  
V  M 22    11 12  ; u11 , u12 , u21 , u22   is a vector space. V.S.  M 22 , , ,  
 u21 u22  
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
 Question 4: Show that set of all real matrices of order m x n
  u11 u12
... u1n  
 
 u ... u2 n 
u21 
V  M mn    21  ; u , u , u , u ....u   is a vector space.
  ... ...
... ...  11 12 21 22 mn

 um1 um1 
... umn  mn 
 
Similar to the M 22 , we can show that M mn is also a vector space.

Hint: We take 3 matrices

 a11 a12 ... a1n   b11 b12 ... b1n   c11 c12 ... c1n 
a a21 ... a2 n  b b21 ... b2 n  c c21 ... c2 n 
A   21  , B   21  & C   21  and 2 scalars
 ... ... ... ...   ... ... ... ...   ... ... ... ... 
     
 am1 am1 ... amn  bm1 bm1 ... bmn  cm1 cm1 ... cmn 
,  .

V.S.  M mn , , ,  

Dr. PAR VEZ ALAM
 Question 5: Show that n-th degree polynomial space: (It consist all degree polynomials upto
degree  n ) V.S.  Pn ( x), , ,  

V  Pn ( x)  an x n  an1x n1  ....  a2 x 2  a1x  a0 ; an , an1 ,....., a2 , a1, a0   is a vector space.
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
 Question 6: Show that Function space: (the set of all real-valued continuous functions
defined on the entire real line.) V  C  ,    { f :  ,    ; f is continuous} is a
vector space with addition and scalar multiplication defined by
 f  g  ( x)  f ( x)  g ( x); x 
;  is any scalar
 f  ( x)  f ( x); x 
V.S.  C  ,   , , ,  
Note:

Proof: Let f , g , h  C  ,   and a, b 

V1: We know the result that sum of two continues functions are continuous, so if
f , g , C  ,   then f  g  C  ,   .

V6: We know the result that if we multiply a constant (scalar) in a continues function then
that also will be continuous, so a f  C  ,   .
Dr. PAR VEZ ALAM
 
V.S.  2
, , , 
Dr. PAR VEZ ALAM
 Note: To show that a set is not a vector space, you need only
find one axiom that is not satisfied.

Ex1.The set of all integers (V  ) is not a vector space.

Proof: It is not closed under scalar multiplication (Failing V6 condition)

Ex2: The set of all second-degree polynomials is not a vector

space.
Dr. PAR VEZ ALAM

Proof: It is not closed under vector addition (Failing V1 condition)

 Ex3:
with defined operations

Proof: (Failing V10 condition; 1.v=v)

Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM

Exercise
 Vector Subspace
Definition: Let V be a vector space over , then W is said to be a subspace of V if and only if
W is a subset of V (i.e, W  V) and W is itself a vector space over with same operations
of V.

Note: Every vector space V has at least two subspaces known as trivial subspace of V.

(1) Zero vector space W={0} is a subspace of V.

(2) V is a subspace of V (itself).

Theorem: (Test for a subspace)

If W is a nonempty subset of a vector space V, then W is a subspace of V if and only if for

every elements w1 , w2 W and scalars   ; following two conditions should satisfy

1. (V1 Closure property w.r.t. vector addition): w1  w2  W

2. (V6 Closure property w.r.t. scalar multiplication): w1  W

Alternate: Some time we use combined form of V1 and V6 to show subspaces as; w1 , w2 W
and scalars ,  , w1  w2  W.

Remark: Also, for Non-emptiness, we use to check that the zero vector (additive identity) is
available in W or not.

Note: We can any of one to check subspaces, either test by showing V1 and V6 or together
(Alternate). But some time Alternate one is being complex see Ex. 1,2,6, however it’s time
saving. But go for normal test by showing V1 and V6 see Ex. 7,8,910.

Proof of Test: It’s only for understanding that how the subset W is also forming a vector
space w.r.t. the same operation of same field. Not important for exam
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
 Homework Questions:
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM

Exercise
Dr. PAR VEZ ALAM
 Intersection of two vector subspaces
Theorem1: The intersection W1  W2 of two vector subspaces W1 and W2 a vector space V
is also vector subspace of V.

Proof: Proof just shown to make understand you all, not in syllabus
(i). As W1 and W2 vector subspaces then 0 vector also be in both so, 0 W1  W2 .
(ii). Let u, v  W1  W2  u, v  W1 and u, v  W2 ,
then defiantly by definition of subspace, u  v  W1 and u  v  W2  u  v  W1  W2 .
Hence W1  W2 also forming subspace of V.

Union of two vector subspaces

Theorem2: The union W2  W1 of two vector subspaces W1 and W2 a vector space V is also
vector subspace of V if and only if, either W1  W2 or W2  W1 .

Linear sum of two vector subspaces

Let W1 and W2 are two vector subspaces of a vector space V. Then there linear sum is denote
by W1  W2 and defined by W1  W2  w1  w 2 ; w1  W1 and w 2  W2  .

Theorem3: The Linear sum W1  W2 of two vector subspaces W1 and W2 a vector space V is
also vector subspace.

Direct sum of two vector subspaces

Let W1 and W2 are two vector subspaces of a vector space V. Then there Direct sum is denote
by W1  W2 and every element of set W1  W2 is uniquely represented by the sum of
w1  W1 and w 2  W2
W1  W2  w1  w 2 ; w1  W1, w 2  W2 and W1  W2  {0} .
Dr. PAR VEZ ALAM

In particular, a vector space V is called direct sum of W1 and W2 , if it is linear sum of W1 ,

W2 and W1  W2  {0} , and we denote it by V=W1  W2 and can be written as
V=W1  W2  w1  w 2 ; w1  W1, w 2  W2 and W1  W2  {0} .
 Example 4: Let U and V are two vector subspaces of M 22 set of all 2x2 square real matrices
a b   a 0 
U   ; a, b   , V    ; a, c  
 0 0    c 0  
  2a b  
U  V    ; a, b, c 
 is also form a vector subspaces of M 22 . U+V is linear sum
 c 0 
a 0   0 0 
but not direct sum, because U  V     ; a      .
  0 0     0 0 

Example 6: Vector space of all n x n matrices is the direct sum of its subspaces “n x n
symmetric matrices” and “n x n skew-symmetric matrices”.
Dr. PAR VEZ ALAM
 What is ‘Linear combination’?
Let u1 , u2 , u3 , u4 , ...., un are n vectors in vector space V and
c1 , c2 , c3 , c4 , ...., cn are n
scalars in , then the combination c1u1  c2 u2  c3 u3  c4 u4 ...., cn un will give a new
vector x (say) and this x vector is called linear combination of vectors u1 , u2 , u3 , u4 , ...., un .

i.e., c1u1  c2 u2  c3u3  c4 u4 ...., cn un  x V.

and this x also belongs to the vector space V.
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
 Linearly Span
If we consider a set S  s1 ,s 2 ,s3 ,....s n   V , then the set of all linear combinations (for different
set of scalars we will have different linear combinations) of these vectors si's called linear span
of set S, we write it by symbol L(S) and given by

n 
L(S)   isi 1s1   2s 2 + 3s3 +...+ ns n ;  i  
 i1 
Note: 1. This L(S) will also form a vector subspace of V.
2. L(S) is smallest vector subspace of V containing S.

A spanning set of a vector space: If every vector in a given vector space V can be written as
a linear combination of vectors in a given set S, then S is called a spanning set of the vector
space and we can write as
L(S)=V
It means V is spanned (generated) by S.

Linearly Independent (L.I.)

&
Linearly Dependent (L.D.)
Let a set S  s1,s2 ,s3 ,....s n   V , then vectors s1,s2 ,s3 ,....sn are said to be linearly
independent, if the vector equation

1s1  2s2 + 3s3 +...+nsn  0 ………..(1)

has only the trivial solution 1  2 =3 =....=n  0 .

If the equation (1) has non-trivial solution, i.e., all scalars 1, 2 , 3 ,....n are not zero, then
we say that vectors s1,s2 ,s3 ,....s n are linearly dependent (L.D.).

Remarks:

1.  is L.I.
2. If the set S is L.I. then every subset of S is L.I.
Dr. PAR VEZ ALAM

3. If the set L.D. then every superset of S is L.D.

4. Every singleton set of non-zero vector is L.I.
5. The singleton set S  {v} is L.D. if and only if v is zero vector.
6. If a non-zero vector v is multiple of a non-zero vector u, then u, v are L.D. i.e., set {u, v}
is L.D.
For example; {u=(1, 2, 3), v=(2, 4, 6)} is L.D., because v is multiple of u, v=2u.
Ex.2 {u=(1, 2, 3), v=(1, 2, 3)} is also L.D. same element, v=1.u.
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
 Basis and dimension of a vector space
Basis: Let V be a vector space. A subset S of V is called basis if
1. S spans V, i.e., L(S)=V
2. Subset S is linearly independent
Dimension: The number of elements (cardinality) in a basis of V is called the
dimension of V, and is denoted by dim V.

Notes:
1.  is basis for V.S. V={0}, so dim 0  0.  
; B  e1  (1, 0), e2  (0, 1) .
2
2. The standard basis for
So, the dimension of
2
is 2, i.e., dim    2.
2

; B  e1  (1, 0, 0), e2  (0, 1, 0), e3  (0, 0, 1) .

3
3. The standard basis for
So, the dimension of
3
is 3, i.e., dim    3.
3

4
4. The standard basis for ;
B  e1  (1, 0, 0, 0), e2  (0, 1, 0, 0), e3  (0, 0, 1, 0), e4  (0, 0, 0, 1) .
So, the dimension of
4
is 4, i.e., dim    4.
4

n
5. The standard basis for ;
B  e1  (1, 0, 0, 0, ...,0), e2  (0, 1, 0, 0,...,0), e3  (0, 0, 1, 0,...,0), .......... en  (0, 0, 0, 0,...,1) .
So, the dimension of
3
is 2, i.e., dim    n.
n

6. The standard basis for 2 x 2 matrix space M 22 :

 1 0  0 1  0 0  0 0 
B  e1    , e2    , e3    , e4   
 0 0  0 0  1 0  0 1  
So, the dimension of M 22 is 4, i.e., dim  M 22   2  2  4.
7. The standard basis for 3 x 2 matrix space M 32 :

 1 0  0 1  0 0  0 0  0 0  0 0 
 
B  e1  0 0 , e2  0 0 , e3  1 0 , e4  0 1  , e5  0 0 , e6  0 0 
           
 0 0 0 0 0 0 0 0 1 0 0 1  

Dr. PAR VEZ ALAM

So, the dimension of M 32 is 2, i.e., dim  M 32   3  2  6.

8. So, the dimension of M mn is given by dim  M mn   m  n.

  . So, dim  P ( x)   2  1  3.
9. The standard basis for P2 ( x) ; B  e1  1, e2  x, e3  x
2
2

10. The standard basis for P ( x) ; B  e  1, e  x, e  x , e  x  . dim  P ( x)   3  1  4 .

2 3
3 1 2 3 4 3

11. The standard basis for P ( x) ; B  e  1, e  x, e  x ,..., e  x  . dim  P ( x)   n  1.

2 n
n 1 2 3 n n

12. The standard basis for P ( x ) ; B  e  1, e  x, e  x ,......... . So, dim  P ( x)    .

2
1 2 3 n
 12. Generally we are representing standard basis by B  e1 , e2 , e3 , .......... en  .

Theorem: (Number of vectors in a basis):

If a vector space V has one basis with n vectors, then every basis for V has n vectors.
(All bases for a finite-dimensional vector space have the same number of vectors.)

Finite dimensional: A vector space V is called finite dimensional, if it has a basis consisting of
a finite number of elements.
For example:  Pn ( x) set of polynomials of all degrees less or equals to n is dim  Pn ( x)   n .

Infinite dimensional: If a vector space V is not finite dimensional, then it is called infinite
dimensional.
For example:
Dr. PAR VEZ ALAM
 Question 1: Find the basis and dimension of subspaces U and W;
(a). W   d , c  d , c  ; c, d  , (b). U   2 x, x, 0 ; a, c  
Dr. PAR VEZ ALAM
 Que 2: Let W be the subspace of all symmetric matrices in M 22 . What is the dimension of W?
OR
a b  
Find the basis and dimension of subspace W     such that b  c; a , b, c , d  .
 c d  
OR
a b  
Find the basis and dimension of subspace W     ; a , b, c , d  .
  b d  
Dr. PAR VEZ ALAM
 3
Que 3: Find the basis and dimension of the plane x + 3y + z = 0 in .
OR
Find the basis and dimension of set W   x, y, z   3
; x  3 y  z  0 .

OR

Find the basis and dimension of subspace W

Dr. PAR VEZ ALAM

3
Que 4: (HW) Find the basis and dimension of the plane x + 2z = 0 in .
OR
Find the basis and dimension of set W   x, y, z   3
; x  2 z  0 .
Hint: W   2z, y, z ; y, z  .
 Note: In such kind of questions, try put direct values of scalars in the coordinate of the vector
in set (see last 2 questions). If directly not possible to put, then solve them and put
corresponding values (see the next question).

3
Que 5: Find the basis and dimension of subspace W of
W   x, y, z   3
; x  y  z  0, x  2y  2 z  0, 2 x  y  z  0 .
Solution: First we solve the equations x  y  z  0, x  2y  2 z  0, 2 x  y  z  0
given in set, then we replace in set.

1 1 1 0 1 1 1 0 
   
1 2 2 0  by Gauss Ele. method, we get 0 1 1 0 
 2 1 1 0  0 0 0 0 
 Let equations z  t , then we get x  0, y  t . Now the subspace W can be written as;
W   0,  t , t   3
; t .
Now, we find set of vectors that spans W;
 0,  t, t   t  0,  1, 1
S   0,  1, 1  L( S )  W
 Now, we check its L.I.; S is singleton set of a non-zero element, so it is L.I.
 Hence, S is the basis of W, and Dim W=1.
Dr. PAR VEZ ALAM
 Corollary 1. Let S be a set of n vectors in an n-dimensional vector space V. If S is
independent, then S is a basis for V.

Corollary 2. Let S be a set of n vectors in an n-dimensional vector space V. If S spans V, then

S is a basis for V.

Que 5: Is the set of vectors W  1, 1, 2 , 1, 2, 5 , 5, 3, 4   

in V3 ( ) i. e.,
3
 form a basis
3
for ? (Use of Corollary 1)
Dr. PAR VEZ ALAM
Dr. PAR VEZ ALAM
 Extension theorem and Deletion theorem:

Theorem: Let V be a finite-dimensional vector space.

A. Any linearly independent set in V can be extended to a basis by adding more vectors
if necessary.
B. Any set of vectors (L.D.) that spans V can be reduced to a basis by discarding vectors if
necessary.

Examples on Extension theorem

We add any one element (vector) of standard basis B into the given L.I. set S to make it
another basis of standard (given) set, but for this after adding any one vector in the set,
the new set should also be L.I. to become basis.

Question 1: Extend the Linearly Independent set S  1, 2, 1 ,  2, 1, 1  to obtain the basis of
3
the vector space .
Solution: As the set S  1, 2, 1 ,  2, 1, 1  is Linearly Independent and to make it basis of
3 3
L.I. set should have 3 vectors as Dim( )=3, so we add one more vector from its standard
basis B  e1  (1, 0, 0), e2  (0, 1, 0), e3  (0, 0, 1) , such that the new set also should be L.I.

Adding set e1  (1, 0, 0) in S, so the new set is 1, 2, 1 ,  2, 1, 1 , (1, 0, 0)

(After adding we check its L.I. or L.D., if L.D. then again tries by adding next vector from
standers basis, we do same until not getting L.I.)

Checking it Linearly independency; taking scalars 1, 2 , 3 such that

1 1, 2, 1   2  2, 1, 1   3 (1, 0, 0)  (0, 0, 0)
  1  2 2  3 , 21   2 , 1   2   (0, 0, 0)
1 2 1 0 
 
  2 1 0 0  on solving we get by G.J.E. mrthod, we are getting 1   2  3  0 .
1 1 0 0
So set 1, 2, 1 ,  2, 1, 1 , (1, 0, 0) is L.I. having 3 elements so by corollary 1, it is forming a
3
basis of .
Dr. PAR VEZ ALAM

Exercise 1: Find two basises for

3
by extend the L.I. set S  1, 1, 0 ,  2, 0, 1  .
Solution: (HW) Hint: add first e1 and the e3……

 1 0  1 0   0 0  
Exercise 2: Extend the set S    ,  ,    to obtain the basis of the vector
 0 0  0 1  0 1  
space M 22 . HW
 Examples on Deletion theorem

1. To find basis from this spanning set using (which may be L.D.), first we arrange
given vectors in rows of a matrix.
2. Then we find Row Echelon form of the matrix.
3. Then collect the corresponding of non-zero rows (vectors), means removing L.D.
vector(s), those for those row(s) coming zero.
4. These vectors will form basis for the set.
Important Note: No need to check L.D. of set. Just proceed the process; if set will be
L.D. automatically L.D. vector will become zero in Row Echelon form.

Question 1: Find a basis for the vector space V spanned by vectors

w1  1, 1, 0  , w2   0, 1, 1 , w3   2, 3, 1 , and w4  1, 1, 1.

Solution: The given vectors 1, 1, 0  ,  0, 1, 1 ,  2, 3, 1 , 1, 1, 1 are L.D. (Check H.W.; no
need to show in exam, just proceed the process…), however they spanning the space. To
be a basis they should be L.I. also. Now we try to remove L.D. vector by the following process.

Arranging vectors in row of a matrix as given below and finding Row echelon form of the
matrix
1
0 1 1 1 0
0
1  1 converting in Row Echelon form 0 1 0
   
2
1  (You have to show proper setps in exam) 0 0
3 1
   
1
1 1 0 0 0
Collecting the corresponding non-zero vectors; B  w1  1, 1, 0  , w2   0, 1, 0  , w3   0, 0, 1
forming a basis for V as it spanning V and L.I. also.

Question 2: If W is a subspace of Pn ( x) generated by the following polynomials;

Solution: Vectors can be written as

v1   1,  2, 4, 1 , v2    2,  3, 9,  1 , v3   1, 0, 6,  5 , v4    2,  5, 7, 5
Dr. PAR VEZ ALAM

 
The given vectors v1 , v2 , v3 , v4  are L.D. (Check H.W.; no need to show in exam, just
proceed the process…), however they spanning the space. To be a basis they should be L.I.
also. Now we remove L.D. vector by the following process.
 1  2 4 1 1  2 4 1
 2  3 9 1 converting in Row Echelon form 0 1 1 3
   
 1 0 6  5  (You have to show proper setps in exam)  0 0 0 0 
   
 2 5 7 5  0 0 0 0

Collecting the corresponding non-zero vectors;

B  v1   1,  2, 4, 1 , v2    0, 1, 1,  3 i.e., B  t 3  2 t 2  4t+1, t 2 +t  3 forming a
basis for W, as it spanning W and L.I. also.

Based on Deletion and extension theorem

Question 3: (The first part of the question is based on Deletion theorem and second part is on
extension theorem)

Solution:
Pat 1: (Use of Deletion theorem)
The given vectors are L.D. (Check H.W.; no need to show in exam, just proceed the
process….), however they spanning the subspace W. To be a basis of W they should be L.I.
also. Now we try to remove L.D. vector by the following process.

Arranging vectors in row of a matrix as

1 4 2 1  1 4 2 1 
1 3 1 2 converting in Row Echelon form
 0 1 1 1 
  (You have to show proper setps in exam)  
3 8 2 7  0 0 0 0 
Collecting the corresponding non-zero vectors; B  1,  4,  2, 1 ,  0, 1, 1, 1 forming a
basis for W, as they are spanning subspace W and L.I. also.

Pat 2: (Use of Extension theorem)

Adding vectors e3  (0, 0, 1, 0) and e4  (0, 0, 0, 1) into the set B, so the new set
B 4  1,  4,  2, 1 ,  0, 1, 1, 1 ,  0, 0, 1, 0  ,  0, 0, 0, 1 is also L.I. (check H.W., you have
Dr. PAR VEZ ALAM

4
to show in exam). Hence the set B 4 is forming a basis of by corollary 1.