THE OPEN UNIVERSITY OF TANZANIA

LINEAR ALGEBRA I:OMT 112

Mr. Gekonga W Chacha VECTOR SPACES AND SUBSPACES LECTURE 2.

VECTOR SPACES

Definition
A vector space is a set V of vectors that is closed with respect to two operations, addition and scalar
multiplication which satisfies the following axioms.
1. Closure property of addition;
~u + ~v ∈ V ∀~u, ~v ∈ V .

2. Commutative property of addition;

~u + ~v = ~v + ~u ∀~u, ~v ∈ V .

3. Associative property of addition;

(~u + ~v ) + w ~ ∀~u, ~v , w
~ = ~u + (~v + w) ~ ∈V.

4. Identity property of addition;

∃0 ∈ V called a zero vector, such that ∀~u ∈ V ~u + 0 = 0 + ~u = ~u.

5. Inverse property of addition;

∀~u ∈ V, ∃ − ~u ∈ V called negative of ~u such that ~u − ~u = ~u + (−~u) = 0.

6. Closure property of scalar multiplication;

α~u ∈ V ∀~u ∈ V where α is a scalar.

7. Distributive property with single scalar and two vectors;

α(~u + ~v ) = α~u + α~u ∀~u, ~v ∈ V where α is a scalar.

8. Distributive property with two scalars and single vector;

(α + β)~u = α~u + β~u ∀~u ∈ V where αandβ are scalars.

9. Associative property of scalar multiplication;

α(β~u) = (αβ)~u ∀~u ∈ V where α and β are scalars.

10. Identity property of scalar multiplication;

1~u = ~u ∀~u ∈ V.

Example 1
If n is any positive integer then show that the set V = Rn with the standard addition and scalar multiplication
is a vector space.
Solution
Given V = Rn
To prove that V is a vector space, we need to show that V satisfies all 10 axioms of a for a set of vectors to
be a vector space.

1
1. Let ~u = (u1 , u2 , ..., un ), ~v = (v1 , v2 , ..., vn ) ∈ Rn
We need to prove that. ~u + ~v ∈ Rn
From
~u + ~v = (u1 , u2 , ..., un ) + (v1 , v2 , ..., vn )
= (u1 + v1 , u2 + v2 , ..., un + vn )
since ui + vi ∈ R for i = 1, 2, ..n implies that (u1 + v1 , u2 + v2 , u3 + v3 ) ∈ Rn .
Thus,
~u + ~v ∈ Rn ∀~u, ~v ∈ Rn
therefore axiom 1 holds.

2. Let ~u = (u1 , u2 , ..., un ), ~v = (v1 , v2 , ..., vn ) ∈ Rn

We need to prove that ~u + ~v = ~v + ~u ∀~u, ~v ∈ Rn
From LHS
~u + ~v = (u1 , u2 , ..., un ) + (v1 , v2 , ..., vn )
= (u1 + v1 , u2 + v2 , ..., un + vn )
= (v1 + u1 , v2 + u2 , ..., vn + un ), R is commutative under addition
= (v1 , v2 , ..., vn ) + (u1 , u2 , ..., un )
= ~v + ~u.
Therefore the axiom 2 holds.

~ = (w1 , w2 , ..., wn ) ∈ Rn
3. Let ~u = (u1 , u2 , ..., un ), ~v = (v1 , v2 , ..., vn ), w
We need to prove that (~u + ~v ) + w ~ = ~u + (~v + w) ~ ∀~u, ~v , w ~ ∈ Rn .
From LHS
(~u + ~v ) + w
~ = ((u1 , u2 , ..., un ) + (v1 , v2 , ..., vn )) + (w1 , w2 , ..., wn )
= (u1 + v1 , u2 + v2 , ..., un + vn ) + (w1 , w2 , ..., wn )
= (u1 + v1 + w1 , u2 + v2 + w2 , ..., un + vn + wn )
= (u1 , u2 , ..., un ) + (v1 + w1 , v2 + w2 , ..., vn + wn )
= (u1 , u2 , ..., un ) + ((v1 , v2 , ..., vn ) + (w1 , w2 , ..., wn ))
= ~u + (~v + w)
~
Therefore, because the LHS=RHS the axiom holds.

4. Since ∃(0, 0, ..., 0) ∈ Rn such that ∀~u = (u1 , u2 , ..., un ) ∈ Rn

(0, 0, ..., 0) + (u1 , u2 , ..., un ) = (u1 , u2 , ..., un )

implies that the axiom holds.

5. Since ∀(u1 , u2 , ..., un ) ∈ Rn ∃(−u1 , −u2 , ..., −un ) ∈ Rn such that

(u1 , u2 , ..., un ) + (−u1 , −u2 , ..., −un ) = (0, 0, ..., 0).

Thus the axiom holds.

6. Let α ∈ R be any scalar and ~u = (u1 , u2 , ..., un ) ∈ Rn

We need to prove that α~u ∈ Rn
From
α~u = α(u1 , u2 , ..., un )
= (αu1 , αu2 , ..., αun )
since αui ∈ R for i = 1, 2, ..., n implies (αu1 , αu2 , ..., αun ) ∈ Rn .
Thus, α~u ∈ Rn ∀~u ∈ Rn therefore the axiom holds.

2
 7. Let ~u = (u1 , u2 , ..., un ), ~v = (v1 , v2 , ..., vn ) ∈ Rn and α ∈ R be any scalar.
We need to show that α(~u + ~v ) = α~u + α~v
From LHS
α(~u + ~v ) = α((u1 , u2 , ..., un ) + (v1 , v2 , ..., vn ))
= α(u1 + v1 , u2 + v2 , ..., un + vn )
= (αu1 + αv1 , αu2 + αv2 , ..., αun + αvn )
= (αu1 , αu2 , ..., αun ) + (αv1 , αv2 , ..., αvn )
= α(u1 , u2 , ..., un ) + α(v1 , v2 , ..., vn )
= α~u + α~v .
Therefore the axiom holds, since LHS=RHS.

8. Let k, α ∈ R be scalars and ~u = (u1 , u2 , ..., un ) ∈ Rn is any vector.

We need to prove that (k + α)~u = k~u + α~u.
From LHS
(k + α)~u = (k + α)(u1 , u2 , ..., un )
= ((k + α)u1 , (k + α)u2 , ..., (k + α)un )
= (ku1 + αu1 , ku2 + αu2 , ..., kun + αun )
= (ku1 , ku2 , ..., kun ) + (αu1 , αu2 , ..., αun )
= k(u1 , u2 , ..., un ) + α(u1 , u2 , ..., un )
= k~u + α~u.
Thus the axiom holds, since LHS=RHS

9. Let k, α ∈ R be scalars and ~u = (u1 , u2 , ..., un ) ∈ Rn is any vector.

We need to prove that k(α~u) = (kα)~u.
From LHS
k(α~u) = k(α(u1 , u2 , ..., un ))
= k((αu1 , αu2 , ..., αun ))
= (kαu1 , kαu2 , ..., kαun )
= kα(u1 , u2 , ..., un )
= kα(~u).
Thus, the axiom holds since the LHS = RHS.

10. Sine ∃1 ∈ R a scalar such that 1u = 1(u1 , u2 , ..., un ) = (u1 , u2 , ..., un ) = ~u ∀~u ∈ R.
Thus, the axiom holds.
Therefore the set V = Rn is a vector space since it satisfies all the axioms.

Example 2

Show that V = R2 with the standard vector addition and scalar multiplication defined as

c(u1 , u2 ) = (u1 , cu2 )

is either a vector space or not.

Solution
We need to prove or disprove.
Let c, k be any scalars in R and ~u = (u1 , u2 ) ∈ R2 .
V = R2 is not a vector space because it does not satisfy the axioms that

(c + k)(~u) = c~u + k~u ∀~u ∈ R2

3
 i.e (c + k)(~u) = (c + k)(u1 , u2 )
= (u1 , (c + k)u2 )
= (u1 , cu2 + ku2 )
= (u1 , cu2 ) + (0, ku2 )
6= k~u + c~u.
Since the axiom above does not hold, thus R2 is not a vector space.

Example 3

Show that the set V = Rn with the standard scalar multiplication and addition defined as

(u1 , u2 ) + (v1 + v2 ) = (u1 + 2v1 , u2 + v2 )

is not the vector space.

Solution
Let ~u = (u1 , u2 ) and ~v = (v1 , v2 ) be any vectors in R2 .

R2 is not a vector space since does not satisfy the commutative of addition that

~u + ~v = (u1 , u2 ) + (v1 , v2 )

~u + ~v = (u1 + 2v1 , u2 + v2 )
and
~v + ~u = (v1 , v2 ) + (u1 , u2 )
= (v1 + 2u1 , v2 + u2 ).
~u + ~v 6= ~v + ~u ∀~u, ~v ∈ R2

SUBSPACES

Defintion
Suppose V is a vector space and W is a subset of V . If, under addition and scalar multiplication that is
defined on V . W is also a vector space then we call W a subspace of V .

TEST FOR SUBSPACES

A subspace of a vector space V is a subset W of V that satisfies three axioms.

1. A zero vector 0 is in W .

2. ∀~u, ~v ∈ W , ~u + ~v ∈ W .

3. ∀~u ∈ W and for each scalar α ∈ R α~u ∈ W.

Example 1
Let   
 a 
W = 0 : a, b ∈ R .
b
 

Show that W is a subspace of R3

4
Solution
  
 a 
Given W = 0 : a, b ∈ R
b
 
To prove that W is a subspace of R3 we need to show that it satisfies all three axioms.

 
0
I. Since ∃ 0 ∈ W when a = b = 0 ∈ R. Therefore, the axiom holds.
0
   
a1 a2
II. Let 0 , 0  ∈ W ∀a1 , b1 , a2 , b2 ∈ R
  
b1 b2
   
a1 a2
We need to prove that  0  +  0  ∈ W
b1 b2
From      
a1 a2 a1 + a2
 0  +  0  =  0 .
b1 b2 b1 + b2
     
a1 + a2 a1 a2
Thus, since a1 + a2 , b1 + b2 ∈ R =⇒  0  ∈ W =⇒ 0 + 0  ∈ W ∀a1 , a2 , b1 , b2 ∈ R.
  
b1 + b2 b1 b2
Therefore the axiom holds.
 
a
III. Let α ∈ R be any scalar and ~u = 0 ∈ W be any vector.

b
We need to prove that α~u ∈ W
From  
a
α~u = α 0

b
 
αa
= 0 .

αb
 
αa
Since αa, αb are in R implies that 0  is in W and therefore α~u is in W ∀α ∈ R and ∀~u ∈ W.

αb
Therefore W is a subspace of R3 because it satisfies all three axioms of a subspace.

Example 2
Let A be 3 × 3 matrix such that the matrix equation

A~x = (1, 2, 4)T

has infinitely many solutions. Does the set of solutions form a subspace of R3 .

Solution
The set of solutions does not form a subspace of R3 since are not closed under addition.
Let ~x1 , ~x2 be solutions of the given equation.

i.e A(~x1 + ~x2 ) = A~x1 + A~x2 = (1, 2, 4)T + (1, 2, 4)T = (2, 4, 8)T 6= (1, 2, 4)T