VECTOR SPACES:

BASIS AND
DIMENSION
Lecturer: Askarbekkyzy Aknur
 Vector spaces
First, we state two equivalent definitions of a
basis of a vector space V.
Definition 1. A set of vectors 𝑆 = {𝑣! , . . . , 𝑣" }
is a basis of 𝑉 if
1. 𝑆 is linearly independent.
2. 𝑆 spans 𝑉.
 Vector spaces
Definition 2. A set of vectors 𝑆 = {𝑣! , . . . , 𝑣" }
is a basis of 𝑉 if every 𝑣 ∈ 𝑉 can be written
uniquely as a linear combination of the basis
vectors.
Proposition 1. The Definitions 1 and 2 are
equivalent.
Example 1. a) Let 𝑉 = ℝ# over ℝ. Consider the
vectors
1 0 0
𝑒! = 0 , 𝑒$ = 1 , 𝑒# = 0 .
0 0 1
We know from Lecture 3 that these vectors span
ℝ# . Furthermore, they are independent. Consider
𝜆! 𝑒! + 𝜆$ 𝑒$ + 𝜆# 𝑒# = 0.
We have
1 0 0 0
𝜆! 0 + 𝜆$ 1 + 𝜆# 0 = 0 .
0 0 1 0
One can easily check that 𝜆! = 𝜆$ = 𝜆# = 0.
#
So 𝑒! , 𝑒$ and 𝑒# are independent in ℝ . Hence
#
{𝑒! , 𝑒$ , 𝑒# } is a basis of ℝ . This basis is called a
#
standard basis of ℝ .
b) This example generalizes the preceding example.
"
Consider the following 𝑛 vectors in ℝ :

1 0 0
0 1 0
𝑒! = ⋮ , 𝑒$ = ⋮ , … , 𝑒" = ⋮ .
0 0 0
0 0 1
These vectors are linearly independent. Furthermore, any
vector
𝑎!
𝑎"
#
𝑣= ⋮ ∈ ℝ
𝑎#$!
𝑎#
can be written as a linear combination of 𝑒! , 𝑒" , . . . , 𝑒# .
Specifically, 𝑣 = 𝑎! 𝑒! + 𝑎" 𝑒" + ⋯ + 𝑎# 𝑒# .
Accordingly, the vectors form a basis of ℝ# which is
called standard basis of ℝ𝒏 .
c) Let 𝑉 = 𝑀$,$ be a vector space of all 2 × 2
matrices over ℝ. The following four matrices
from 𝑀$,$ form a basis of the vector space 𝑀$,$
over ℝ:

1 0 0 1 0 0 0 0
, , ,
0 0 0 0 1 0 0 1
Since
𝑎 𝑏 1 0 0 1
=𝑎 +𝑏 +
𝑐 𝑑 0 0 0 0
0 0 0 0
+𝑐 +𝑑
1 0 0 1
and their independence is evident.
d) Vector space 𝑃" (𝑡) of all polynomials of degree
less than or equal to 𝑛 . The set 𝑆 =
$ "
{1, 𝑡, 𝑡 , . . . , 𝑡 } is a basis of 𝑃" (𝑡) . Any
polynomial can be written as
" "
𝑎& + 𝑎! 𝑡+. . . +𝑎" 𝑡 = 𝑎& 1 + 𝑎! 𝑡+. . . +𝑎" 𝑡
and their independence is evident.
The following is a fundamental result in linear
algebra.
Theorem 1. Let 𝑉 be a vector space such that
one basis has 𝑚 vectors and another basis has 𝑛
vectors. Then 𝑚 = 𝑛.
Definition 3. A vector space 𝑉 is said to be of
finite dimension 𝒏 or 𝒏 −dimensional, written
dim𝑉 = 𝑛 if 𝑉 has a basis with 𝑛 vectors. If a
vector space 𝑉 does not have finite basis, then 𝑉 is
said to be of infinite dimension or to be infinite-
dimensional.
The vector space {0} is defined to have
dimension 0.
Example 4.
"
dim ℝ = 𝑛
dim 𝑀'," = 𝑚𝑛
dim 𝑃" 𝑡 = 𝑛 + 1
Theorem 2. Let 𝑉 be a vector space of finite
dimension 𝑛. Then:
◦Any 𝑛 + 1 or more vectors in 𝑉 are linearly
dependent.
◦Any linearly independent set 𝑆 = {𝑣! , . . . , 𝑣" }
with 𝑛 vectors is a basis of 𝑉.
◦Any spanning set 𝑇 = {𝑤! , . . . , 𝑤" } of 𝑉 with 𝑛
elements is a basis of 𝑉.
Example 2. a) Let 𝑉 = !
ℝ . Then𝑑𝑖𝑚ℝ! = 3.
We know the vectors
1 0 0
𝑒" = 0 , 𝑒# = 1 , 𝑒! = 0
0 0 1
forms a basis of ℝ! . Then for any nonzero 𝑣 ∈ ℝ!
by the first part of the Theorem above the four
vectors 𝑒" , 𝑒# , 𝑒! , 𝑣 are not linearly independent,
consequently {𝑒" , 𝑒# , 𝑒! , 𝑣} is not a basis of ℝ! .
 #
b) Let 𝑉 = ℝ . It is easy to show that
2 1 1
𝑣! = 0 , 𝑣$ = 3 , 𝑣# = 2
0 0 3
# #
are linearly independent in ℝ and 𝑑𝑖𝑚ℝ = 3.
Then by the second part of the Theorem above
#
{𝑣! , 𝑣$ , 𝑣# } is a basis of ℝ .
 #
c) Let 𝑉 = ℝ . We know from Lecture 3 that
the vectors
1 1 1
𝑓! = 1 , 𝑓$ = 1 , 𝑓# = 0
1 0 0
# #
span ℝ . Taking account that 𝑑𝑖𝑚ℝ = 3 and
the third part of the Theorem above we have
#
{𝑓! , 𝑓$ , 𝑓# } is a basis of ℝ .
In an echelon form matrix, no nonzero row is a
linear combination of the other nonzero rows.
The nonzero rows of an echelon form matrix
make up a linearly independent set. Namely, rows
in an echelon matrix with n columns give us set
"
of linearly independent vectors in ℝ .
 1
Example 3. a) Given vectors 𝑣! = 1 , 𝑣" =
0
1 4
3 , 𝑣& = 9 . We check whether they are linearly
2 5
dependent or not by echelon matrix (from Lecture 3
we know that they are linearly dependent). First we
1 1 0
write them as rows of matrix and have 1 3 2 .
4 9 5
Perform the following sequence of elementary
operations 𝑅$ → −𝑅! + 𝑅$ , 𝑅# → −4𝑅! + 𝑅# ,
𝑅# → −5𝑅$ + 2𝑅# :
1 1 0 1 1 0 1 1 0
1 3 2 ~ 0 2 2 ~ 0 2 2 ~
4 9 5 4 9 5 0 5 5
1 1 0
1 1 0
~ 0 2 2 ~ .
0 2 2
0 0 0
Finally, we have the echelon matrix with two
nonzero rows. It means the third row is a linear
combination of the first and second rows, namely,
𝑣# is a linear combination of 𝑣! and 𝑣$ . Hence,
they are linearly dependent in ℝ# .
 1 3
b) Given vectors 𝑣" = 2 , 𝑣# = −1 , 𝑣! =
1 −1
−2
!
2 of ℝ . We want to check whether the set
3
{𝑣" , 𝑣# , 𝑣! } is a basis of ℝ! . First, we write them as
rows of matrix and have
1 2 1
3 −1 −1 .
−2 2 3
After applying the sequence of following
elementary operations 𝑅$ → −3𝑅! +
𝑅$ , 𝑅# → 2𝑅! + 𝑅# , 𝑅# → 7𝑅# , 𝑅# →
6𝑅$ + 𝑅# one can obtain the following echelon
matrix
1 2 1
0 −7 −4 .
0 0 11
As we see there are 3 nonzero rows in the
echelon matrix and therefore these rows define
#
3 linearly independent vectors in ℝ obtained
from 𝑣! , 𝑣$ , 𝑣# . Then 𝑣! , 𝑣$ , 𝑣# are linearly
independent vectors in three dimensional vector
space. Hence the set {𝑣! , 𝑣$ , 𝑣# } is a basis of
ℝ# .
с) Given set 𝑆 = {𝑣! , 𝑣" , 𝑣& , 𝑣' }, where
1 1 2 3
3 4 3 8
𝑣! = 1 , 𝑣" = 3 , 𝑣& = −4 , 𝑣' = 1 .
−2 −1 −7 −7
−3 −4 −3 −8
We will extend the set 𝑆 to a basis of vectors of ℝ( . First
of all, we need to determine whether they are linearly
independent or not. If not, we will delete dependent
vectors from 𝑆.
 1 3 1 −2 −3
1 4 3 −1 −4
~
2 3 −4 −7 −3
3 8 1 −7 −8
1 3 1 −2 −3 1 3 1 −2 −3
0 1 2 1 −1 0 1 2 1 −1
~ ~ ~
0 −3 −6 −3 3 0 0 0 0 0
0 −1 −2 −1 1 0 0 0 0 0
1 3 1 −2 −3
~
0 1 2 1 −1
So we observe that 𝑣! , 𝑣" , 𝑣& , 𝑣' are linearly dependent
and only two vectors are linearly independent. To have
a basis of ℝ( we need three more vectors so that new
five vectors become linearly independent with
1 0
3 1
{ 1 , 2 }.
−2 1
−3 −1
We add three more rows
1 3 1 −2 −3
0 1 2 1 −1
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1
and have an echelon matrix with 5 nonzero rows.
Now we have five linearly independent vectors.
The set of vectors
1 0 0 0 0
3 1 0 0 0
{ 1 , 2 , 1 , 0 , 0 }.
−2 1 0 1 0
−3 −1 0 0 1
is a basis of ℝ( . We note that this extension of course
is not unique. We couldtake other three vectors so that
a set of five vectors form a basis of ℝ( .
Definition 3. The rank of a matrix 𝑨 , written
𝑟𝑎𝑛𝑘(𝐴), is equal to number of rows in its echelon form
matrix.
1 1 0
Example 4. Let 𝐴 = 1 3 2 .
4 9 5
Then rank of 𝐴 is 2, write 𝑟𝑎𝑛𝑘(𝐴) = 2, since its
echelon matrix has two nonzero rows
1 1 0
.
0 2 2
We consider an application of rank in space of
solutions of a homogeneous system of linear
equations in 𝑛 unknowns. Recall that
homogeneous systems have either infinitely many
solutions or only zero solutions. The set of
solutions of systems forms a vector space with
respect to the operations in ℝ" .
Theorem 3. Let 𝑊 be space of solutions of a
homogeneous system of linear equations in 𝑛
unknowns and 𝐴 be a matrix of coefficients of
unknowns with 𝑟𝑎𝑛𝑘(𝐴) = 𝑟. Then 𝑑𝑖𝑚 𝑊 =
𝑛 − 𝑟.
 𝑥+𝑦−𝑧 =0
Example 5. Consider V2𝑥 − 3𝑦 + 𝑧 = 0.
𝑥 − 4𝑦 + 2𝑧 = 0
1 1 −1
Then 𝐴 = 2 −3 1 . An echelon matrix
1 −4 2
1 1 −1
of 𝐴 is and 𝑟𝑎𝑛𝑘(𝐴) = 2.
0 −5 3
Then 𝑑𝑖𝑚 𝑊 = 3 − 2 = 1, that is, the space
of solutions has dimension one. Of course, the
system has infinitely many solutions of the form
$ #
𝑥, 𝑦, 𝑧 = ( 𝑧, 𝑧, 𝑧), but as a vector space all
( (
solutions are linear combination of one vector
2/5
3/5 .
1
Definition 4. Let 𝑈 and 𝑊 be subsets of a vector
space 𝑉 . The sum of U and W, written 𝑈 + 𝑊 ,
consists of all sums 𝑢 + 𝑤 where 𝑢 ∈ 𝑈 and 𝑤 ∈ 𝑊.
Theorem 4. Suppose 𝑈 and 𝑊 are subspaces of 𝑉.
Then 𝑈 + 𝑊 and 𝑈 ∩ 𝑊 are subspaces of 𝑉.
Theorem 5. Suppose 𝑈 and 𝑊 are finite dimensional
subspaces of 𝑉. Then
𝑑𝑖𝑚 𝑈 + 𝑊
= 𝑑𝑖𝑚 𝑈 + 𝑑𝑖𝑚 𝑊 − 𝑑𝑖𝑚 𝑈 ∩ 𝑊 .
Example 6. Given
1 1 2
3 4 3
𝑣! = −2 , 𝑣" = −3 , 𝑣& = −1 ,
2 4 −2
3 2 9
1 1 2
3 5 5
𝑤! = 0 , 𝑤" = −6 , 𝑤& = 3 .
2 6 2
1 3 1
Let 𝑈 be a space spanned by 𝑢! , 𝑢$ , 𝑢# and 𝑊 be
a space spanned by 𝑤! , 𝑤$ , 𝑤# . Namely, any
vector of 𝑈 and 𝑊 is a linear combination of
𝑢! , 𝑢$ , 𝑢# and 𝑤! , 𝑤$ , 𝑤# , respectively. They are
(
subspaces of ℝ .
We will find bases and dimensions of 𝑈, 𝑊, 𝑈 +
𝑊 and 𝑈 ∩ 𝑊.
To construct a basis of 𝑈 we need to derive
linearly independent vectors from 𝑢! , 𝑢$ , 𝑢# . We
write them as rows of matrix and find its echelon
form matrix.
 1 3 −2 2 3 1 3 −2 2 3
1 4 −3 4 2 ~ 0 1 −1 2 −1
2 3 −1 −2 9 0 −3 3 −6 3

1 3 −2 2 3
1 3 −2 2 3
~ 0 1 −1 2 −1 ~ .
0 1 −1 2 −1
0 0 0 0 0
So there are only two nonzero rows or rank of the
matrix is two. Thus,
1 0
3 1
𝑢! = −2 , 𝑢′" = −1
2 2
3 −1
are linearly independent and they span 𝑈, therefore
)
{𝑢! , 𝑢" } is a basis and 𝑑𝑖𝑚 𝑈 = 2.
In a similar way, we find a basis of 𝑊 and its
dimension.
1 3 0 2 1 1 3 0 2 1
1 5 −6 6 3 ~ 0 2 −6 4 2 ~
2 5 3 2 1 0 −1 3 −2 −1
1 3 0 2 1
~ 0 1 −3 2 1 ~
0 −1 3 −2 −1
1 3 0 2 1
1 3 0 2 1
◦ 0 1 −3 2 1 ~ .
0 1 −3 2 1
0 0 0 0 0
So there are only two nonzero rows or rank of the
matrix is two. Thus,
1 0
3 1
𝑤! = 0 , 𝑤′" = −3
2 2
1 1
are linearly independent and they span 𝑊, therefore
{𝑤! , 𝑤′" } is a basis and 𝑑𝑖𝑚 𝑊 = 2.
By the definition of 𝑈 + 𝑊 , the vectors
𝑢! , 𝑢") , 𝑤! , 𝑤′" span 𝑈 + 𝑊.
1 3 −2 2 3 1 3 −2 2 3
0 1 −1 2 −1 ~ 0 1 −1 2 −1 ~
1 3 0 2 1 0 0 2 0 −2
0 1 −3 2 1 0 0 −2 0 −2
1 3 −2 2 3
1 3 −2 2 3
0 1 −1 2 −1
~ ~ 0 1 −1 2 −1 .
0 0 2 0 −2
0 0 1 0 −1
0 0 0 0 0
So, there are only three nonzero rows or rank of the
matrix is three. Thus,
1 0 0
3 1 0
𝑢! = −2 , 𝑢′" = −1 , 𝑤′! = 1
2 2 0
3 −1 −1
are linearly independent and they span 𝑈 + 𝑊 ,
)
therefore {𝑢! , 𝑢" , 𝑤′! } is a basis and 𝑑𝑖𝑚 (𝑈 +
𝑊 ) = 3.
To find the dimension of 𝑈 ∩ 𝑊 we use the
formula given above and have 𝑑𝑖𝑚(𝑈 ∩ 𝑊 ) =
𝑑𝑖𝑚 𝑈 + 𝑑𝑖𝑚 𝑊 − 𝑑𝑖𝑚(𝑈 + 𝑊 ) = 2 + 2 −
3 = 1.
Now we find a basis of 𝑈 ∩ 𝑊 and it consists of
𝑎!
𝑎"
one vector. Let 𝑣 = 𝑎& and 𝑣 ∈ 𝑈 ∩ 𝑊.
𝑎'
𝑎(
Then 𝑣 must be written as a linear combination of
their basis vectors.
𝑣 = 𝜆! 𝑢! + 𝜆" 𝑢") = 𝜇! 𝑤! + 𝜇" 𝑤′" .
1 0 1 0
3 1 3 1
𝜆! −2 + 𝜆" −1 = 𝜇! 0 + 𝜇" −3 .
2 2 2 2
3 −1 1 1
We obtain a system of five linear equations in
unknowns 𝜆! , 𝜆" , 𝜇! , 𝜇" :
𝜆! − 𝜇" = 0
3𝜆! + 𝜆" − 3𝜇! − 𝜇" = 0
−2𝜆! − 𝜆" + 3𝜇" = 0
2𝜆! + 2𝜆" − 2𝜇! − 2𝜇" = 0
3𝜆! − 𝜆" − 𝜇! − 𝜇" = 0
The solution is {(𝜆! , 𝜆" , 𝜇! , 𝜇" ) =
(𝜇" , 𝜇" , 𝜇" , 𝜇" )|𝜇" ∈ 𝑅}. Let 𝜇" = 1 and
1
4
)
𝑣 = 𝑢! + 𝑢" = −3 is the vector in 𝑈 ∩ 𝑊 and
4
2
span it. Hence it is a basis of 𝑈 ∩ 𝑊.
Definition 5. The vector space 𝑉 is said to be
the direct sum of its subspaces of 𝑈 and 𝑊,
denoted by 𝑉 = 𝑈 ⊕ 𝑊 if every 𝑣 ∈ 𝑉 can be
written in one and only one way as 𝑣 = 𝑢 + 𝑤
where 𝑢 ∈ 𝑈 and 𝑤 ∈ 𝑊.
Theorem 6. The vector space 𝑉 is the direct sum
of its subspaces of 𝑈 and 𝑊 if and only if

𝑉 = 𝑈 + 𝑊.
𝑈 ∩ 𝑊 = 0.
 𝑎
#
Example 7. a) Let 𝑉 = ℝ and 𝑈 = { 𝑏 |𝑎 =
𝑐
𝑎
𝑐} and 𝑊 = { 𝑏 |𝑎 + 𝑏 + 𝑐 = 0}. We show
𝑐
that 𝑉 = 𝑈 + 𝑊 but the sum is not direct.
𝑎
Suppose 𝑣 = 𝑏 ∈ 𝑉.
𝑐
Then
𝑎 𝑐 𝑎−𝑐
𝑏 = 𝑎+𝑏−𝑐 + 𝑐−𝑎 .
𝑐 𝑐 0
Then 𝑉 = 𝑈 + 𝑊. We note that
𝑎 𝑐−1 𝑎−𝑐+1
𝑏 = 𝑎+𝑏−𝑐+2 + 𝑐−𝑎−2 .
𝑐 𝑐−1 1
There are two ways of expressing 𝑣 as a linear
combination of vectors 𝑈 and 𝑊. Therefore, the
sum is not direct.
Example 3. Let 𝑉 = ℝ# . Vectors
1 0 0
𝑒! = 0 , 𝑒$ = 1 , 𝑒# = 0
0 0 1
𝑎
# 𝑏 #
span ℝ . For any 𝑣 = ∈ ℝ one can have
𝑐
𝑣 = 𝑎𝑒! + 𝑏𝑒$ + 𝑐𝑒# .
 #
b) Let 𝑉 = ℝ and
𝑎 0
𝑈 = { 𝑏 |𝑎 = 𝑐} and 𝑊 = { 0 |𝑐 ∈ ℝ}.
𝑐 𝑐
We show that 𝑉 = 𝑈 ⊕ 𝑊. Note that
𝑎 𝑎 0
𝑏 = 𝑏 + 0 .
𝑐 𝑎 𝑐−𝑎
Then 𝑉 = 𝑈 ⊕ 𝑊. Let 𝑣 ∈ 𝑈 ∩ 𝑊. It implies
𝑎 = 𝑐 and 𝑎 = 𝑏 = 0. Then 𝑎 = 𝑏 = 𝑐 = 0.
0
Thus, 𝑣 = 0 and 𝑈 ∩ 𝑊 = {0}. Hence
0
𝑉 = 𝑈 ⊕ 𝑊.