6.

4 Basis and Dimension

DEF (→ p. 263)

A set S = {v 1 , v 2 , …v k } of vectors in a vector

space V is a basis for V if
(1) S spans V and
(2) S is linearly independent.

MATH 316U (003) - 6.4 (Basis and Dimension) / 1

EXAMPLE 1 (→ EXAMPLE 1 from the previous
lecture)
Let S = { i , j , k } = {(1, 0, 0), (0, 1, 0), (0, 0, 1)}
(1) we’ve shown that S spans R 3
(2) c 1 i + c 2 j + c 3 k = 0 corresponds to the
homogeneous system with the augmented
matrix
1 0 0 | 0
0 1 0 | 0
0 0 1 | 0
The solution is unique: c 1 = c 2 = c 3 = 0 (the
trivial solution).

Answer: S is a basis for R 3 .

→ Example 1 p. 263 ( n = 2, general n)

MATH 316U (003) - 6.4 (Basis and Dimension) / 2

EXAMPLE 2
Is the set S = {(1, 1), (1, −1)} a basis for R 2 ?

(1) Does S span R 2 ?

1 1 x
Solve c 1 + c2 =
1 −1 y

1c 1 + 1c 2 = x
1c 1 − 1c 2 = y

1 1 | x r 2 −r 1 →r 2 1 1 | x
Ò
1 −1 | y 0 −2 | y − x

This system is consistent for every x and y,

therefore S spans R 2 .

MATH 316U (003) - 6.4 (Basis and Dimension) / 3

(2) Is S linearly independent?
1 1 0
Solve c 1 + c2 =
1 −1 0

1c 1 + 1c 2 = 0
1c 1 − 1c 2 = 0

1 1 | 0 r 2 −r 1 →r 2 1 1 | 0
Ò
1 −1 | 0 0 −2 | 0

The system has a unique solution c 1 = c 2 = 0

(trivial solution).

Therefore S is linearly independent.

Consequently, S is a basis for R 2 .

MATH 316U (003) - 6.4 (Basis and Dimension) / 4

EXAMPLE 3 Is S = {(1, 2, 3), (0, 1, 2), (−1, 0, 1)}
a basis for R 3 ?

It was already shown (→ EXAMPLE 3 from the

previous lecture) that S does not span R 3 .

Therefore S is not a basis for R 3 .

EXAMPLE 4 Is S = {(1, 0), (0, 1), (−2, 5)} a basis

for R 2 ?

It was already shown (→ EXAMPLE 4 from the

previous lecture) that S is linearly dependent.

Therefore S is not a basis for R 2 .

→ Example 2 p.263.

MATH 316U (003) - 6.4 (Basis and Dimension) / 5

EXAMPLE 5
S={
1 0 0 1 0 0 0 0
, , , }
0 0 0 0 1 0 0 1

v1 v2 v3 v4
is a basis for the vector space M 22 .

a b
(1) c 1 v 1 + c 2 v 2 + c 3 v 3 + c 4 v 4 = v =
c d
is equivalent to:
c1 c2 a b
=
c3 c4 c d
which is consistent for every a, b, c, and d.
Therefore S spans M 22 .

MATH 316U (003) - 6.4 (Basis and Dimension) / 6

(2) c 1 v 1 + c 2 v 2 + c 3 v 3 + c 4 v 4 = 0
is equivalent to:
c1 c2 0 0
=
c3 c4 0 0
The system has only the trivial solution ⇒ S is
linearly independent.

Consequently, S is a basis for M 22 .

MATH 316U (003) - 6.4 (Basis and Dimension) / 7

EXAMPLE 6 Is S = {1, t, t 2 , t 3 } a basis for P 3 ?

(1) c 1 (1) + c 2 (t) + c 3 (t 2 ) + c 4 (t 3 )

= a + bt + ct 2 + dt 3
has a solution for every a, b, c, and d :
c 1 = a, c 2 = b, c 3 = c, c 4 = d.
Therefore S spans P 3 .

(2) c 1 (1) + c 2 (t) + c 3 (t 2 ) + c 4 (t 3 ) = 0

can only be solved by
c 1 = c 2 = c 3 = c 4 = 0.
Therefore S is linearly independent.

Consequently, S is a basis for P 3 .

→ Example 3 p.264.

MATH 316U (003) - 6.4 (Basis and Dimension) / 8

THEOREM (→ Th. 6.5 p. 265)
Let S = {v 1 , v 2 , …v k } be a set of nonzero vectors
in a vector space V. The following statements are
equivalent:
(A) S is a basis for V,
(B) every vector in V can be expressed as a linear
combination of the vectors in S in a unique
way.

MATH 316U (003) - 6.4 (Basis and Dimension) / 9

Proof (A)⇒(B)
• Every vector in V can be expressed as a linear
combination of vectors in S because S spans
V.
• Suppose v can be represented as a linear
combination of vectors in S in two ways:
v = c1 v1 + ` + ck vk
v = d1 v1 + ` + dk vk
Subtract:
0 = (c 1 − d 1 )v 1 + ` + (c k − d k )v k
Since S is linearly independent, then
c1 − d1 = ` = ck − dk = 0
so that
c1 = d1
_
ck = dk
The representation is unique.

MATH 316U (003) - 6.4 (Basis and Dimension) / 10

Proof (B)⇒(A)
• (B)⇒ Every vector in V is in span S.
• Zero vector in V can be represented in a
unique way as a linear combination of
vectors in S:
0 = c1 v1 + ` + ck vk
This unique way must be: c 1 = ` = c k = 0.
Therefore S is linearly independent.

Consequently, S is a basis for V.

MATH 316U (003) - 6.4 (Basis and Dimension) / 11

Back to EXAMPLE 2:
S = {(1, 1), (1, −1)}

Instead of showing that

• c 1 v 1 + c 2 v 2 = v has a solution, and

• c 1 v 1 + c 2 v 2 = 0 has a unique solution,

we can show

• c 1 v 1 + c 2 v 2 = v has a unique solution.

1 1 | x r 2 −r 1 →r 2 1 1 | x
Ò
1 −1 | y 0 −2 | y − x

Unique solution for every x and y ⇒ S is a basis

for R 2 .

MATH 316U (003) - 6.4 (Basis and Dimension) / 12

TH 6.6 (→ p. 266)
Let S = {v 1 , v 2 , …v k } be a set of nonzero vectors
in a vector space V. Some subset of S is a basis for
W = span S.

→ Procedure p. 268

MATH 316U (003) - 6.4 (Basis and Dimension) / 13

EXAMPLE 7 Find a basis for
span{(1, 2, 3), (−1, −2, −3), (0, 1, 1), (1, 1, 2)}.
v1 v2 v3 v4

Set c 1 v 1 + c 2 v 2 + c 3 v 3 + c 4 v 4 = 0 .
The corresponding system has augmented matrix:
1 −1 0 1 | 0
2 −2 1 1 | 0
3 −3 1 2 | 0
which is equivalent
(r 2 − 2r 1 → r 2 ; r 3 − 3r 1 → r 3 ; r 3 − r 2 → r 3 ) to

1 −1 0 1 | 0

0 0 1 −1 | 0
0 0 0 0 | 0

MATH 316U (003) - 6.4 (Basis and Dimension) / 14

Can set c 2 and c 4 arbitrary. For example
• If c 2 = 1, c 4 = 0 then v 2 can be expressed as
a linear combination of v 1 and v 3 .
• If c 2 = 0, c 4 = 1 then v 4 can be expressed as
a linear combination of v 1 and v 3 .
Therefore, every vector in span S can be
expressed as a linear combination of v 1 and v 3 .
Also note that v 1 and v 3 are linearly independent.
Consequently, they form a basis for span S.
Summarizing: The vectors corresponding to the
columns with leading entries form a basis for W.

Different initial ordering of vectors, e.g.,

{v 2 , v 1 , v 3 , v 4 } may change the basis obtained by
the procedure above (in this case: v 2 , v 3 ).

MATH 316U (003) - 6.4 (Basis and Dimension) / 15

TH 6.7 (→ p. 269)
Let S = {v 1 , v 2 , …v k } span V and let
T = {w 1 , w 2 , …w n } be a linearly independent set
of vectors in V. Then n ≤ k.

COROLLARY 6.1 (→ p. 270)

Let S = {v 1 , v 2 , …v k } and T = {w 1 , w 2 , …w n }
both be bases for V. Then n = k.

DEF (→ p. 270)
The dimension of a vector space V, denoted dim
V, is the number of vectors in a basis for V.
dim({ 0 }) = 0.

• dim(R n ) = n (→ Example 6 p. 270)

• dim(P n ) = n + 1 (→ Example 7 p. 270)
• dim(M mn ) = mn

MATH 316U (003) - 6.4 (Basis and Dimension) / 16

TH 6.8 (→ p. 271)
If S is a linearly independent set of vectors in a
finite-dimensional vector space V, then there
exists a basis T for V, which contains S.

EXAMPLE 8 (→ Example 9 p. 271)

Find a basis for R 4 that contains the vectors
v 1 = (1, 0, 1, 0) and v 2 = (−1, 1, −1, 0).
Solution:
The natural basis for R 4 :
{(1, 0, 0, 0), (0, 1, 0, 0), (0, 0, 1, 0), (0, 0, 0, 1)}
e1 e2 e3 e4

Follow the procedure of EXAMPLE 7 to

determine a basis of span{v 1 , v 2 , e 1 , e 2 , e 3 , e 4 }.
1 −1 1 0 0 0 | 0
0 1 0 1 0 0 | 0
1 −1 0 0 1 0 | 0
0 0 0 0 0 1 | 0

MATH 316U (003) - 6.4 (Basis and Dimension) / 17

has the reduced row echelon form:

1 0 0 1 1 0 | 0

0 1 0 1 0 0 | 0

0 0 1 0 −1 0 | 0

0 0 0 0 0 1 | 0

Answer: {v 1 , v 2 , e 1 , e 4 }.

TH 6.9 (→ p. 272)
Let V be an n-dimensional vector space, and let
S = {v 1 , v 2 , …v n } be a set of n vectors in V.
(a) If S is linearly independent then it is a basis
for V.
(b) If S spans V then it is a basis for V.

MATH 316U (003) - 6.4 (Basis and Dimension) / 18