Chapter6

RealVectorSpaces

Objective:
l Studymanyimportantpropertiesof
n

vectorsin R
l Acarefullyconstructedgeneralizationof
Rn

tomanyotherimportantvectorspace.
l Applicationofvectorspacetothelinear
systemandmatrix.

6.1.VectorSpaces
DefinitionofVectorSpace
Definition:
ArealvectorspaceVisasetofelements
togetherwithtwooperations,addition +and
scalarmultiplication .,satisfyingthe
followingproperties:
Letu,v,andwbevectorsinV,andletcand dbe
scalars.

Addition:
( a )Ifu andvareanyelementsofV,then

u +v

isinV(closed).
1.

u +v = v+ u.

2.

u +(v+ w)= (u+ v)+ w.

3. Vhasazerovector 0 suchthatforeveryu
inV,

u +0= 0+ u = u.

4. Forevery uinV,thereisanelement
V,

u+(-u)= 0.

-u

in

Scalarmultiplication:
( b )IfuisanyelementofVandcisanyreal
number, then cu isinV(closed).
5.
6.
7.

c(du)=(cd)u.
(c +d)u= cu+ du.

c(u+v)= cu+ cv.

8. 1u = u.

Example:
V1 Rn together with standard vector addition and scalar

multiplication.Then, V1
is a vector space since for any two vectors u and v in Rn
and scalar c, both u +v and

cu are in Rn. Therefore,

conditions ( a ) and ( b ) are satisfied. In addition, the

conditions (1) to (8) are satisfied (see the previous
subsection).

Example:
V2 thesetconsistingofall m n matricestogetherwith

standard matrix addition and scalar multiplication. V2 is a

vector space since for any two m n matrices u and v
and scalar c, both u +v and cu are m n matrices.
Thatis,both

u +v

and

cu

arestillin V2 .Therefore,

conditions ( a ) and ( b ) are satisfied. In addition, the

conditions(1)to(8)aresatisfied(seesection2).

Example:
V3 thesetconsistingofallpolynomialsofdegree2orlesswith
theformtogetherwithstandardpolynomialadditionandscalar
multiplication.Is V3 avectorspace?
We need to examine whether the conditions ( a ), ( b ), and the
conditions(1)to(8)aresatisfied.Let

u = a2 x2 + a1x + a0 ,

v = b2x2 + b1x + b0 ,
w = c2x2 + c1x + c0
andletcanddbescalars.Then,

Solution:
Addition:
( a ):
u +v= (a2x2 + a1x+ a0)+ (b2x2 + b1x+ b0)
= (a2 + b2)x2 + (a1 + b1)x+ (a0 + b0) V3

since u +v isapolynomialofdegree2orless.

Solution:
(1):
u +v= (a2 x2 + a1x+ a0)+ (b2x2 + b1x+ b0)
= (a2 + b2)x2 + (a1 + b1)x+ (a0 + b0)
2

= (b2 + a2)x + (b1 + a1)x+ (b0 + a0)

= (b2x2 + b1x+ b0)+ (a2x2 + a1x+ a0)= v+ u

Solution:
(2):

u + (v+ w)= (a2x2+ a1x+ a0)+[(b2 + c2)x2 + (b1+ c1)x+ (b0 + c0)]
= (a2 + b2 + c2)x2 + (a1+ b1+ c1)x+ (a0 + b0 + c0)
2

=[(a2 + b2)x + (a1+ b1)x+ (a0 + b0)]+ c2x + c1x+ c0

= (u+ v)+ w

Solution:
(3):
Let 0=0x2 + 0x+ 0.Then,
u+0=(a2+0)x2+(a1+0)x+(a0+0)=(0+a2)x2+(0+a1)x+(0+a0)=0+u
=a2x2+a1x+a0=u

(4):
Let

-u = (-a2)x2 + (-a1)x+ (- a0).

Then,

u +-u=[a2 + (-a2)]x2 +[a1+ (-a1)]x+[a0 + (-a0)]= 0x2 + 0x+ 0= 0

Solution:
Scalarmultiplication:
( b ):
2

cu =(ca2)x + (ca1)x+ (ca0) V3

since cu isapolynomialofdegree2orless.
(5):
c(u+v)= [c(a2 + b2)]x2 + [c(a1 + b1)]x+ [c(a0 + b0)]
= (ca2 + cb2)x2 + (ca1 + cb1)x+ (ca0 + cb0)= cu+ cv

Solution:
(6):
(c +d)u= [(c+ d)a2]x2 + [(c+ d)a1]x+ [(c+ d)a0]
= [(ca2)x2 + (ca1)x+ (ca0)]+ [(da2)x2 + (da1)x+ (da0)]= cu+ du

(7):
c(du)=c[(da2 )x2 + (da1)x+ (da0)]= c(da2)x2 + c(da1)x+ c(da0)
= (cd)a2x2 + (cd)a1x+ (cd)a0 = (cd)u

(8):
2

1u =1a2x + 1a1x+ 1a0 = a2x + a1x+ a0 = u

Note:
Pn thesetconsistingofallpolynomialsofdegreen
orlesswiththeformtogetherwithstandard
polynomialadditionandscalarmultiplication.Then,
Pn isa vectorspace.Inaddition, P the set

consistingofallpolynomialswiththeformtogether
withstandardpolynomialadditionandscalar
multiplication.Then,

isalsoa vectorspace.

Example:
V4 thesetconsistingof all realvaluedcontinuous

functions definedontheentirereallinetogetherwith
standardadditionandscalarmultiplication.Is V4 avector
space?
Weneedtoexaminewhethertheconditions( a ),( b ),and
theconditions (1)to(8)aresatisfied.Let

u = f(x),v = g(x),w = h(x)

andlet c and d bescalars.Then,

Solution:
Addition:
( a ):

u +v= f(x)+ g(x) V4

since f (x)+g(x) isstillacontinuousfunction.
(1):

u +v= f(x)+ g(x)= g(x)+ f(x)= v+ u

(2):
u +(v+ w)= f(x)+ [g(x)+ h(x)] = [ f(x)+ g(x)]+ h(x)
= (u+ v)+ w

Solution:
(3):
Letthezerovector 0 = 0 .Then,
u +0= f(x)+ 0= f(x)= u

(4):
Let -u = - f(x).Then,

u +(-u)= f(x)+ [- f(x)]= f(x)- f(x)= 0

Solution:
( b ):

cu =cf(x) V4
since cf(x) isstillacontinuousfunction.
(5):

c(u+v)= c[ f(x)+ g(x)]= cf(x)+ cg(x)= cu+ cv

(6):

(c +d)u= (c+ d)f(x)= cf(x)+ df(x)= cu+ du

Solution:
(7):

c(du)=c[df(x)]= cdf(x)= (cd)f(x)= (cd)u

(8):

1(u)=1[ f(x)]= f(x)= u

Note:
Let

V4* the

set of all differentiable functions

definedontheentirereallineand

V4** thesetof

allintegrablefunctionsdefinedontheentirereal
**
*
V
V
line. Both 4 and 4 are vector space under

standardadditionandscalarmultiplication.

Example:
V5 the set consisting of all integers with standard

addition and scalar multiplication. Is V5 a vector

space?
V5 is not a real vector space since for
c=0.7,
cu = 0.7 1= 0.7 V5 ,

condition ( b ) isnotsatisfied.

u =1 V5,

and

Example:
V6 the set consisting of all vectors in R2 with

standard addition and

multiplicationdefinedby

nonstandard

x1 cx1
c =
x2 0

.Is V6 avectorspace?

scalar

Solution:
V6 is not a real vector space since for

c1 ,c2 R,c2 0,
c1 c1 c1
1u = 1 = = u
,
c2 0 c2

condition(8)isnotsatisfied.

c1
u = R2
,
c2

Example:
V7

the set consisting of only second degree

polynomials with standard addition and scalar

multiplication.Is V7 avectorspace?
V7 isnotarealvectorspacesincefor u = x2 and

v =-x2

u + v = x2 + (- x2 ) = 0 V7 ,

condition( a )isnotsatisfied.

Example:
V8 the set consisting of all realvalued continuous
functions such that f(1)=3. Suppose the operations are

standardadditionandscalarmultiplication.Is
space?

V8 avector

Solution:
V8

isnotarealvectorspacesincefor

u = f (x) V8 , v =g(x)V8
f (1)+ g(1) = 3+ 3= 6 3
u+ v = f (x)+ g(x) V8 ,

condition( a )isnotsatisfied.
Example19Page273276

Example1p273
n

ConsiderthesetRn togetherwiththe
operationsofvectoradditionand
scalarmultiplicationasdefinedin
Section4.2.Theorem4.2inSection
4.2establishedthefactthatRn isa
vectorspaceundertheoperationsof
additionandscalarmultiplicationsofn
vector.

Example2p273
n

ConsiderthesetVofallorderedtriples
ofrealnumberoftheform(x,yx,0)and
definetheoperations and by
(x,y,0) (x,y,0)=(x+x,y+y,0)
c (x,y,0)=(cx,cy,0)

Example3p273
ConsiderthesetVofallorderedtriplesof
realnumbers(x,y,z)anddefinethe
operations and by
1.(x,y,z)(x,y,z)=(x+x,y+y,z+z).
2.c (x,y,z)=(cx,y,z).
3.c [(x,y,z)(x,y,z)]=(c(x+x),y+y,z+z).
4.c (x,y,z)d (x,y,z)=((c+d)x,2y,2z).
n

Example4p274
n

ConsiderthesetM23ofall23
matrices.
Similarlytheserofallmnmatrices
undertheusualoperationsofmatrix
additionandscalarmultiplicationisa
vectorspacewillbedenotedbyMmn.

Example5p274
n

LetF[a,b]bethesetofallrealvalued
functionsthataredefinedontheinterval
[a,b].IffandgareinV,wedefinefgand
cfby
(fg)(t)f(t)+g(t).
(cf)(t)cf(t).

Apolynomial(int)isafunctionthatis
expressibleas
p(t)antn+ an1tn1++a1t+a0
a0,a0,a1,arerealnumbers.

Example6p274
n
n
n
n
n

Thefollowfunctionsarepolynomials
p1(t)3t42t25t1.
p2(t)2t1.
p3(t)4.
f4(t) 2 t 6andf5(t) 1/t22t1are
notpolynomials.

Example7p275
n
n
n

p1(t):degree4
p2(t):degree1
p3(t):degree0

Example8p275
n
n
n

Ifp(t)antn+ an1tn1++a1t+a0.
q(t)bntn+ bn1tn1++b1t+b0 .
p(t) q(t)

(an+bn)tn+(an1+bn1)tn1++(a1+b1)t+a0+b0
n cp(t) cantn+can1tn1++ca1t+ca0.
n (c+d)p(t)cp(t) dp(t).

Example9p276
n
n

Letuv uv,cucu,
Ifu3,v2,
uv1.vu1
Ifu4,c2,d3
(c+d) (2+3) 420
c u d v8 124

Theorem6.1:
(ImportantResult)
IfVisavectorspaceandletubeanyelementofa
realvectorspaceV.Then,
(a) 0 u = 0.
(b) c 0 = 0,c R, 0 V.
(c) cu =0. c = 0 or u = 0.
(d) (-1)u = -u.

6.2.Subspace
Definitionofsubspace:
WiscalledasubspaceofarealvectorspaceVif
1. WisasubsetofthevectorspaceV.
2. W is a vector space with respect to the
operationsinV.
Example1(Page279)
Example2(Page279)

Example1p279
n

Everyvectorspacehasatleasttwo
subspaces,itselfandsubspace{0}.
Thesubspace{0}iscalledthezero
subspace.

Example2p279
n

Letu(a1,b1,0),v (a2,b2,0)
uv(a1+a2,b1+b2,0)
cu(ca1,cb1,0)

Theorem6.2:(Important
Result)
WisasubspaceofarealvectorspaceV

1.IfuandvareanyvectorsinW,then u +v W .
2.If cisanyrealnumberand u is any vectorin W,
then cu W .

Example:
W1 the subset of

consisting of all vectors of the

form,
a
0 , a R

,
0

togetherwithstandardadditionandscalarmultiplication.
3
W
Is 1 asubspaceof R ?

Solution:
Weneedtocheckiftheconditions(1)and(2)aresatisfied.
Let
a1
a2
u = 0, v= 0, c R
.
0
0

Then,
(1):
.

a1 a2 a1 + a2
u + v = 0 + 0 = 0 W1
.
0 0 0

Solution:
(2):
ca1
cu = 0 W1
.
0

W1 isasubspaceof .
R3

Example:
LettherealvectorspaceVbethesetconsistingofall
n n matricestogetherwiththestandardadditionand
scalarmultiplication.Let
W2 thesubsetofVconsistingofall n n diagonal
matrices.
Is W2 asubspaceofV?
Let
a11 0
0 a
22
u =
M
M

0 0

0
L 0
W2
O M
,

L ann

b11 0
0 b
22
v=
M
M

0 0

c R.

L 0
L 0
W2
O M
,and

L bnn

Solution:
(1):
0
a11 + b11
0
a22 + b22

u+ v =
M
M

0
0

L
0
W2

O
M

L ann + bnn

since u +v isstilladiagonalmatrix.

Solution:
(2):
ca11 0
0 ca
22
cu=
M
M

0
0

L
L

0
0
W2

O
M

L cann

since cu isstilladiagonalmatrix.
W2 isasubspaceofV.

Example:
Pn thesetconsistingofallpolynomialsofdegreenor

less with the form together with standard polynomial

addition and scalar multiplication. Pn is a vector
space. P the set consisting of all polynomials with
the form together with standard polynomial addition and
scalar multiplication. Then, P is also a vector space.
Then, Pn isasubspaceof P .

Example:
V4 the set consisting of all realvalued continuous

functions defined on the entire real line together with

*
V
standard addition and scalar multiplication. Let 4 the

setofalldifferentiablefunctionsdefinedontheentirereal
linetogetherwithstandard
addition and scalar multiplication. Then, V4 is a real
vectorspace.Also.

V4*

isasubspaceof V4.

Example:
W3 thesubsetof

consistingofallvectorsoftheform,

R3

a
a 2 , a, b R

,
b

togetherwithstandardadditionandscalarmultiplication.Is
W3

asubspaceof

Solution:
(1):
a1
u +v = a12 +
b1

a2 a1 + a2 a1 + a2
a2 = a2 + a2 (a + a )2 W
2
2
3
2 1
1
.
b2 b1 + b2 b1 + b2

Therefore, u +v W3.

W3

is not asubspaceof R .

Example:
V3

thesetconsistingof allpolynomialsofdegree2orless
with the form together with standard polynomial addition
andscalarmultiplication. V3 isavectorspace.Let
W4 thesubsetof V3 consistingofallpolynomialsofthe
form
2

ax +bx+ c, a+ b+ c= 2.
Is W4 asubspaceof V3?

Solution:
Let

u =a2x2 + a1x+ a0 W4
and

v =b2x2 + b1x+ b0 W4.

Then, a2 +a1 + a0 = 2 and b2 +b1 + b0 = 2.Thus,
u + v= (a2x2 + a1x+ a0)+ (b2x2 + b1x+ b0)
= (a2 + b2)x2 + (a1 + b1)x+ (a0 + b0) W4

Solution:
since

(a2 +b2) +(a1+b1) +(a0+b0) =(a2+a1+a0)+(b2+b1+b0)=2+2=4.

W4

is notasubspaceof

V3.

Example310(Page280282)

Example3p280
a1 b1 0
u =

0 c1 d1

a2 b2 0
v =

0 c2 d2

b1 + b2 0
a1 + a2
u +v=

c1 + c2 d1 + d2
0
ka1 kb1 0
ku =

0 kc1 kd1

Example5p282
n

Letu(a1,b1,1),v (a2,b2,1)
uv(a1+a2,b1+b2,2)

Example6p282
n

WeletPn denotethevectorspace
consistingofallpolynomialsofdefree
nandthezeropolynomial.AndletP
denotethevectorspaceofakk
polynomials.ItiseasytoverifythatP2
isasubspaceofP3 andingeneral
thatPnisasubspacesofPn+1.Pn isa
subspaceofP.

Example7p282
n

LetVbethesetallpolynomialsof
degreeexactly 2.VisasubsetP2 but
itisnotasubspaceofP2 ,sincethe
sumofpolynomials2t2+3t+1and
2t2+t+2,apolynomialofdegree1,is
notinV.

Example8p282
n

LetC[a,b]denotethesetofallreal
valuescontinuousfunctionsthatare
definedontheinterval[a,b].Iffandg
areinC[a,b]thenf+gisinC[a,b].

Example9p282
n

Considerthehomogeneoussystem
Ax=0,Aisanmnmatrix.Asolutionconsistsofa
vectorxinRn.LetWbethesubsetofRn consisting
ofallsolutionstothehomogeneoussystem.Since
A0=0,weconcludethatWisnotempty.Tocheck
thatWusasubspaceofRn,weverifyproperties()
and()ofThem6.2.Thusletxandybesolutions.
Ax=0andAy=0
A(x+y)=Ax+Ay=0
A(cx)=c(Ax)=0

Example10p282
n

Letv1 andv2befixedvectorsina
vectorspaceVandletWbethesetof
alllinearcombinationsofv1 andv2.
w1=a1v1+a2v2
w2=b1v1+b2v2
w1+w2=(a1+b1)v1+(a2+b2)v2
cw1=(ca1)v1+(ca2)v2

Definitionoflinearcombination:
Let v1,v2,K,vk bevectorsinarealvectorspaceV.A
vector v in V is called a linear combination of
v1,v2,K
,vk if

v = c1v1 + c2v2 +L+ ckvk,

where c1,c2,K,ck arerealnumbers.

Example:
Let
0 2
- 1 3
- 2 0
0 8
v1 =
, v2 =
, v3 =
, v=

1
0
1
2
1
3
2
1

bevectorsinthevectorspaceconsistingofall

2 2

matrices.Then,
0
v =
2

Thatis,

8
0
= 1

1
1

2
-1
+ 2

0
1

3
- 2
+ (-1)

2
1

0
= v1 + 2v2 - v3

.
3

isalinearcombinationof v1,v2,v3.

Example:
Forlinearsystem
1 0 - 1 x1 1
Ax = 2 1 0 x2 = 1 = b
.
3 2 1 x3 1
2
x = - 3
1

Thus,

isasolutionfortheabovelinearsystem.

Example:
2 1 0 - 1 2
1
0
- 1
A -3 = 2 1 0 - 3 = 2 2 - 3 1 + 1 0
1 3 2 1 1
3
2
1
1
= 2col1(A)- 3col2(A)+ col1(A)= 1 = b
1

Thatis,
ofA,

isalinearcombinationofthecolumnvectors

col1(A),col2 (A),col3(A).

Note:
For a linear system Am n xn 1 = bm 1 , the linear
system has solution or solutions
combinationofthecolumnvectorsofA,

col1(A),col2(A),L,coln (A).
Forexample,if

c1
c
2
c =
M

cn

Am n xn 1 = bm 1 ,

isasolutionof

is a linear

Note:
then

c1col1(A)+c2col2(A)+ L+ cn coln(A)= b.
Ontheotherhand,thelinearsystemhasnosolution

b isnotalinearcombinationofthecolumn
vectorsofA

Example:
4
v= 5
Isthevector
5 alinearcombinationofthevectors

1
-1
3

v1 =2,v2 = 1,v3 = 3
3
4
2 .

Solution:
Weneedtofindtheconstants

c1,c2,c3

suchthat

4
1
- 1
3
v =5 = c12 + c2 1 + c33 = c1v1 + c2v2 + c3v3
.
5
3
4
2

weneedtosolveforthelinearsystem
c1 1 - 1 3 c1 4
Ac2 = 2 1 3 c2 = 5
c3 3 4 2 c3 5 .

Solution:
Thesolutionsare

c1 =-2t+ 3, c2 = t-1, c3 = t, t R.
Thus,

v =(- 2t+ 3)v1 + (t-1)v2 + tv3, t R

v isalinearcombinationof v,v ,v
1

ofexpressions.

withinfinitenumber

Example:
3
v = - 4
Isthevector
- 6 alinearcombinationofthevectors

1
-1
1

v1 =2, v2 = -1, v3 = 4
3
- 2
5 .

Solution:
Weneedtofindtheconstants

c1,c2,c3

suchthat

3
1
- 1
1
v =- 4 = c12 + c2 - 1 + c34 = c1v1 + c2v2 + c3v3
.
- 6
3
- 2
5

Solution:
weneedtosolveforthelinearsystem
c1 1 - 1 1 c1 3
Ac2 = 2 - 1 4 c2 = - 4
c3 3 - 2 5 c3 - 6 .

Thelinearsystemhas no solution.

v is notalinearcombinationof v1,v2,v3

Note:
,vn and
Let v1,v2,K

be vectors in

Rm

and let Am*n be

the matrix with column vectors colj(A)=vj,j=1,2,K,n.

Thus,

Ax =v

has solution or solutions

v is a linear

,vn.
combination of v1,v2,K

Ax =v

has no solution

,vn.
combination of v1,v2,K

v is

not a linear

Definitionofspanningset:
Let S={v1,v2,K,vk} be a set of vectors in a real vector
space V.Then,thespan of S, denoted by span ( S) ,is
the set consisting of all the vectors that are linear

,vk.Thatis,
combinationsof v1,v2,K
span(S)={c1v1 + c2v2 + L+ ck vk |c1,c2,K,ck R}.
If span (S) =V ,itissaidthatVisspannedbySorS
spansV.

Example:
Let
1
0
e1 = 0 , e2 = 1 , e3 =
0
0

0
0 , and S = {e ,e ,e }
1
2
3

.
1

Then,

c1

span(S)=c1e1 + c2e2 + c3e3 = c2 |c1,c2,c3 R = R3

c
3

Example:Example1(Page292)
Let
1
v1 = 2 , v2 =
1

1
0 , v =
3
2

3
span
(
S
)
=
R
Does
?

1
1 , and S = {v ,v ,v }
1
2
3

.
0

Solution:
span (S) = R3 Foranyvector

a
v = b R3
c

existrealnumbers c1,c2,c3 suchthat

a
1
1
1
v = b = c1 2 + c2 0 + c31 = c1v1 + c2v2 + c3v3
c
1
2
0

weneedtosolveforthelinearsystem

,there

Solution:
1 1 1c1 a
2 0 1c = b

2
1 2 0c3 c .
Thesolutionis
- 2a+ 2b+ c
a-b+ c
4a-b- 2c
c1 =
, c2 =
, c3 =
3
3
3 .
- 2a+ 2b+ c
a- b+ c
4a- b- 2c
v =
v
+
v
+
1
2
v3
3
3
3

Thatis,everyvectorin
v1,v2,v3

R3

canbealinearcombinationof

Example10(Page282)
Example11(Page283)
Example12(Page284)
Theorem6.3:(Importantresult)
Let S={v1,v2,K,vk} beasetofvectorsinarealvector
spaceV.Then, span ( S) isasubspaceofV.
Example13(Page285)

Example11p283
v1=(1,2,1),v2=(1,0,2),v3=(1,1,0),v=(2,1,5)
n Ifc1v1+c2v2+c3v3=v,findc1c2c3?
n c1(1,2,1)+c2(1,0,2)+c3(1,1,0)=(2,1,5)
c1+c2+c3=2
2c1+0c2+c3=1
c1+2c2+0c3=5
Soc1=1,c2=2,c3=1
n

Example12p284

1 0 0 0 1 0 0 0 0 0 0 0
S =
,
,
,

0 0 0 0 0 0 0 1 0 0 0 1
1 0 0 0 1 0 0 0 0
0 0 0 a b 0
a
+ b
+ c
+ d
=

0
0
0
0
0
0
0
1
0
0
0
1
0
c
d

Wherea,b,c,anddarerealnumber

Section6.3

LinearIndependence

Definition:
Thevectorsv1,v2,, vk inavectorspaceVaresaidtospan Vif
everyvectorinVisalinearcombinationofv1, v2,, vk. Moreover,
ifS={v1, v2,,vk},thenwealsosaythatthesetSspans V,orthat
{v1, v2,, vk}spans V,orthat Visspannedby S,orinthe
languageofSection6.2,spanS= V.
Theproceduretocheckifthevectorsv1,v2,, vk spanthevector
spaceVisasfollows.

Definition:
Step1.Chooseanarbitraryvector vin V.
Step2.Determineif visalinearcombinationofthegiven
vectors.Ifitis,thenthegivenvectorsspan V. Ifitisnot,
theydonotspan V.

Example26(Page292294)
Question:
Let S={v1,v2,K,vk} and span (S) =W . Is it possible to
find a smaller (or even smallest) set, for example,

S ={v1,v2,K,vk-1},suchthat
*

span ( S) =W = span (S ) ?
Toanswerthisquestion,weneedtointroducetheconceptof
linearindependenceandlineardependence.

LinearIndependence
Definition of linear dependence and linear
independence:
The vectors v1,v2,K,vk in a vector space V are
calledlinearlydependentifthereexistconstants,
c1,c2,K,ck,notall0,suchthat

c1v1 +c2v2 + L+ ckvk = 0.

Definitionoflineardependence
andlinearindependence:
v1,v2,K,vk arelinearlyindependentif

c1v1 +c2v2 + L+ ckvk = 0 c1 = c2 = L = ck = 0

,vk are
Theproceduretodetermineif v1,v2,K
linearlydependentorlinearlyindependent:

Definitionoflineardependence
andlinearindependence:
1. Form equation c1v1+c2v2 +L+ ckvk = 0, which lead to a
homogeneoussystem.
2. Ifthehomogeneoussystemhasonlythetrivialsolution,
then the given vectors are linearly independent if it
has a nontrivial solution, then the vectors are linearly
dependent.

Example:
1
0
e1 = 0 , e2 = 1 , e3 =
0
0

0
0 , and S = {e ,e ,e }
1
2
3

1

e1, e2 and e3 linearlyindependent?

Are

Solution:
1
0
0 1 0 0 c1
c1e1 +c2e2 + c3e3 = c10 + c21 + c30 = 0 1 0 c2 = 0
0
0
1 0 0 1 c3

Therefore,
independent.

e1, e2

c 1
c
2
c 3

and

0
0

0

e3

.
are

linearly

Example:
1
v1 = 2 , v2 =
3

- 2
1 , v =

3
1

8
6 .

10

linearlyindependent?

. Are v1, v2

and v3

Solution:
1
- 2
8 1 - 2 8 c1
c1v1 +c2v2 + c3v3 = c12 + c2 1 + c3 6 = 2 1 6 c2 = 0
3
1
10 3 1 10 c3

c 1
c
2
c 3

4
= t - 2 , t

- 1

Therefore, v1, v2 and v3 arelinearlydependent.

Example:
Determine whether the following set of vectors in
the vector space consisting of all 2 2 matrices is
linearlyindependentorlinearlydependent.
2 1 3 0 1 0
S ={v1,v2,v3} =
,
,

0 1 2 1 2 0 .

Solution:
2
c1v1 +c2v2 + c3v3 = c1
0

1
3
+ c2

1
2

0
1
+ c3

1
2

0 0
=

0 0

0
0

Thus,
2c1 +3c2 + c3 = 0
c1

= 0
2c2 + 2c3 = 0

c1 + c2

= 0

2
3
1 0
1
0
0 0
c1 + c2 + c3 =
0
2
2 0


.
1
1
0 0

Solution:
Thehomogeneoussystemis
2 3 1
0
c

1 0 0 1
0

c2 =
0 2 2
0

3

1
1
0

Theassociatedhomogeneoussystemhasonlythetrivial
solution
c1 0
c = 0
2
c3 0

Therefore, v1, v2 and v3 arelinearlyindependent.

Example:
Determine whether the following set of vectors in
the vector space consisting of all polynomials of
degree n is linearly independent or linearly
dependent.

S ={v1 ,v2,v3} = x2 + x+ 2,2x2 + x,3x2 + 2x+ 2 .

Solution:

) (

) (

c1v1+c2v2 +c3v3 = c1 x2+ x+2+c2 2x2+ x+c3 3x2+ 2x+ 2= 0.

Thus,
c 1 + 2c2 + 3c3 = 0
c1 + c2 + 2c3 = 0
2c1 +

+ 2c3 = 0

1
2
3 0
c11 + c21 + c3 2 = 0
2
0
2 0 .

Solution:
Theassociatedhomogeneoussystemis
1
1

2
1
0

3 c1 0
2 c2 = 0
2 c3 0

Thehomogeneoussystemhasinfinitenumberofsolutions,
c 1
1
c = t 1 , t R.
2

c3
- 1

Therefore, v1, v2 and v3 arelinearlydependentsince

tv1 +tv2 - tv3 = 0, t R.

Example712(Page295296)

Example7p295
-1
- 2
1
0
and
0
1


0
1

-1
- 2 0
1
0 0
c1 + c2 =
0
1 0


0
1 0

c1=c2=0.thevectorsare
linearindependent.

Example8p295
n

Arethevectorsv1=(1,0,1,2),
v2=(0,1,1,2), v3=(1,1,1,3)independent?
c1+c3=0
Theonly
c2+c3=0
solutions
c1=c2=c3 =0.the
c1+c2+c3=0
vectorsare
2c1+2c2+3c3=0
linear
independent.

Example9p295
n

Arethevectorsv1=(1,2,1),v2=(1,2,1),
v3=(3,2,1),v4=(2,0,0),independent?
c1+c2+3c3+2c4=0
2c1+2c2+2c3=0
c1+c2c3=0

Theonlysolutions
c1=1,c2=2,c3 =1.
c4 =0or
c1=1,c2=1,c3=0.
c4=1sothe
vectorsarelinear
dependent.

Example10p296
n

Thevectore1 ande2 inR2 ,definedinEX4,are

linearlyindependentsince
c1(1,0)+c2(0,1)=(0,0)
onlyifc1=c2=0

1 0
A =

0 1

WeformthematrixA,whosecolumnsarethe
givennvectors.Thenthegivenvectorsare
linearlyindependentifandonlyifdet(A) 0.

Example11p296
n
n
n
n
n

Considerthevectors
p1(t)=t2+t+2,p2(t)=2t2+t,p3(t)=3t2+2t+2
c1+2c2+3c3=0
c1=1,c2=1,c3=1.
c1+c2+2c3=0
Lineardependent
2c1+2c3=0

Example12p296
n

Ifv1,v2,v3,vk arekvectorsinandy
vectorspaceandvi isthezerovector.
ThenS={v1,v2,v3,vk }islinearly
dependent.

Note:
In the examples with

1
v1 = 2 , v2 =
3

- 2
1 , v =

3
1

8
6 ,

10

or with

S ={v1 ,v2,v3} = {x2 + x+ 2,2x2 + x,3x2 + 2x+ 2}, v1, v2 and

v3

are linearly dependent. Observe that v3 in both

examplesarelinearcombinationsof v1, v2 ,

Note:
8
1
- 2
v 3 = 6 = 4 2 - 2 1 = 4v3 - 2v2
10
3
1

and
2

) (

v3 =3x + 2x+ 2= x + x+ 2+ 2x + x = v1+ v2.

Asamatteroffact,wehavethefollowinggeneral
result.

Theorem6.4:(Important
result)
The nonzero vectors v1,v2,K,vk in a vector space V
are linearly dependent if and only if one of the
vectors vj, j2, is a linear combination of the
precedingvectors v1,v2,K,vj-1.

Note:
Every set of vectors containing the zero vector is
linearly dependent. That is, v1,v2,K,vk are k vectors
inanyvectorspaceand vi isthezerovector,then
v1,v2,K,vk arelinearlydependent.

Example13onpage298
Example14onpage299

Example13p298
n

Ifv1,v2,v3,v4 areasinEx9,thenwe
findthatv1+v2+0v3v4=0
Sov1,v2,v3,v4 arelineardependent.
Wethenhacev4=v1+v2

Example14p299
1
1
v1 = , v2 =
0

0

1
0
, v3 =
1

0

0
1
,v4 =
1

0

2
1
letw = span S.Since v4 = v1 + v2
1

0

WeconcludetharW=spanS1 where
S1={v1,v2,v3 }

6.4.BasisandDimension
Thevectors v1,v2,K,vk inavectorspaceVaresaidto
formabasisofVif
(a) v1,v2,K,vk spanV(i.e., span(v1,v2,K,vk )=V ).
(b) v1,v2,K,vk arelinearlyindependent.
Example1(Page303)
Naturalbasisorstandardbasis

Example1p303
n

Thevectore1=(1,0)ande2=(0,1)from
abasisforR2.Eachofthesesetof
vectorsiscalledthenaturalbasisor
standardbasis.

Example:
1
0
e1 = 0 , e2 = 1 , e3 =
0
0

0
0 , and S = {e ,e ,e }
1
2
3

.
1

and e3 abasisin R3?

Are e1, e2

Solution:
e1, e2 and e3 formabasisin

R3

since
3

(a) span(S)= span(e1,e2,e3)= R (seetheexample inthe

previoussection).
(b) e1, e2 and e3 are linearly independent (also see the
exampleintheprevioussection).

Example:
1
v1 = , v2 =
0

0
1 , v3 =

3
4 .Are v1, v2

and v3 abasisin R2?

[solution:]
and v3 are not abasisof
linearlydependent,
v1, v2

R2

since

v1, v2

3v1 + 4v2 - v3 = 0.
2
span
(
v
,
v
,
v
)
=
R
1
2
3
Notethat
.

and v3 are

Example:
1
v1 = 2 , v2 =
3

- 2
1 , v =

3
1

3
R
basisin
?

8
6 .

10 .

Are v1, v2 and

v3

Solution:
v1, v2 and v3 are not a basis in R3 since v1, v2 and

arelinearlydependent,
8
1
- 2
v 3 = 6 = 4 2 - 2 1 = 4v3 - 2v2
.
10
3
1

Example2(Page303)
Example3(Page304)
Example4(Page304)

v3

Example2p303
n
n
n
n

S={v1,v2,v3,v4}where
v1=(1,0,1,0),v2=(0,1,1,2)
v3=(0,2,2,1),v4=(1,0,0,1)
ShowthatSisabasisforR4

part1
n
n
n
n

c1+c4=0
c2+2c3=0
c1c2+2c3=0
2c2+c3+c4=0

Theonlysolutions
c1=c2=c3=c4=0,
ShowingthatSis
linearindepeddent

part2
n
n

Letv=(a,b,c,d)beanyvectorinR4.
Wenowseekconstantsk1,k2,k3,k4
suchthatk1v1+k2v2+ k3v3+ k4v4=v
Substitutingforv1v2v3v4andvwefind
asolutionfork1,k2,k3,k4totheresulting
linearsystemforanya,b,c,d.HenceS
spansR4 andisavasisforR4.

Example3p304part1
n

n
n
n
n

ToshowS={t2+1,t
1,2t+2}isabasis
forthevector
spaceP2.
at2+bt+c=a1t2+(a2+
a3)t+(a1a2+2a3)
a1=a
a2+2a3=b
a1a2+2a3=c

a 1 = a,a2 =

a + b - c
c + b - a
,a3 =
2
4

given 2t2 + 6t + 13
a1=2,a2=5/2,a3=17/4
a1(t2+1)+a2(t1)+a3(2t+2)=0

Example3p304part2
Thena1t2+(a2+2a3)t+(a1a2+2a3)=0

a1=0
a2+2a3=0
a1a2+2a3=0
Theonlysolutionsa1=a2=a3=0,
Sislinearindepeddent

Example4p304
n

n
n
n
n
n

FindabasisforthesubspaceVof
P2 ,consistingofallvectoroftheform
at2+bt+c,wherec=ab.
at2+bt+ab
a(t2+1)+b(t1)+c
Sothet2+1andt1spanV.
a(t2+1)+b(t1)=0
Sincethisequationidtoholdallvaluesoft,
wemusthavea1=a2=0.

Example:
Let
1
v1 = 2 , v2 =
1

1
0 , v =
3
2
3

AreSabasisin R ?

1
1 , and S = {v ,v ,v }
1
2
3

.
0

Solution:
(a)
span (S) = R3 Foranyvector

a
v = b R3

c

,there

existrealnumbers c1,c2,c3 suchthat

a
1
1
1
v = b = c12 + c20 + c31 = c1v1 + c2v2 + c3v3
.
c
1
2
0

Solution:
weneedtosolveforthelinearsystem
1 1 1c1 a
2 0 1c = b

2
1 2 0c3 c .

Thesolutionis

- 2a+ 2b+ c
a-b+ c
4a-b- 2c
c1 =
,c2 =
, c3 =
3
3
3 .

Solution:
Thus,
- 2a+ 2b+ c
a- b+ c
4a- b- 2c
v =
v1 +
v2 +
v3
3
3
3

.
3
R
Thatis,everyvectorin
canbealinear

combinationof

v1,v2,v3

and

span (S) = R3.

Solution:
(b)
Since
c1 + c2 + c3 0
c1v1 +c2v2 + c3v3 = 2c1 + c3 = 0 c1 = c2 = c3 = 0
,
c1 + 2c2 0

v1,v2,v3 arelinearlyindependent.
3
v
,
v
,
v
R
.
By(a)and(b), 1 2 3 areabasisof

Theorem6.5:(Importantresult)
If S={v1,v2,K,vk} is a basis for a vector space V, then
every vector in V can be written in an unique(one
and only one) way as a linear combination of the
vectorsinS.

Example:
1
0
e1 = 0 , e2 = 1 , e3 =
0
0

0
0 , and S = {e ,e ,e }
1
2
3

.
1

S is a basis of

v= b
R3.Then,foranyvector
c ,
a
1
0
0
v = b = a 0 + b1 + c 0 = ae1 + be2 + ce3
c
0
0
1

isuniquelydetermined.

Theorem6.6:(Important
result)
Let S={v1,v2,K,vk} be a set of nonzero vectors in a
vectorspaceVandlet

W =span{v1,v2,K ,vk}.Then,

somesubsetofSisabasisofW.

Example:
Let
1 0 2 0 1

S={e1,e2,a1,e3,a2} = 0, 1, 3, 0, 3
0 0 0 1 2

and span(S)= R .PleasefindsubsetsofSwhich

3

formabasisof R3.

Solution:
We first check if e1 and e2 are linearly independent.
Sincetheyarelinearlyindependent,wecontinuetocheckif
e1 , e2 and a1 arelinearlyindependent. Since

2e1 +3e2 - a1 = 0,

Solution:
wedelete a1 fromSandformanewset S , S1 ={e1,e2,e3,a2}.
Then,wecontinuetocheckif e1, e2 and e arelinearly
independent. Theyarelinearlyindependent.Thus,wefinally
1

check if e1, e2 e3 and a2

arelinearlyindependent. Since

e1 +3e2 + 2e3 - a2 = 0,
wedelete a fromSandformanewset S2 ,
Therefore,
1

S 2 ={e1,e2,e3}.

= {e1 ,e2 ,e3 }

3
R
isthesubsetofSwhichformabasisofformabasisof .

Howtofindabasis(subsetofS)ofW?

Therearetwomethods:
Method1:
Theprocedurebasedontheproofoftheabove
importantresult.
Method2:
Step1:Formequation

c1v1 +c2v2 + L+ ckvk = 0.

Howtofindabasis(subsetofS)ofW?
Step2:Constructtheaugmentedmatrixassociatedwiththe
equationinstep1andtransformthisaugmentedmatrix
tothereducedrowechelonform.
Step 3: The vectors corresponding to the columns
containingtheleading1sformabasis.Forexample,if

k =6 andthereducedrowechelonmatrixis

Howtofindabasis(subsetofS)ofW?
1
0

0
M

0
0
0

0
M

thenthe1st,the3nd,andthe4thcolumnscontain
aleading1andthus
v1,v3,v4

areabasisof W =span{v1,v2,K,v6}.

Example:
Let
1 0 2 0 1

S={e1,e2,a1,e3,a2} = 0, 1, 3, 0, 3
0 0 0 1 2

3
(
)
span
S
=
R
and
.PleasefindsubsetsofSwhich

formabasisof R3.

Solution:
Method1:
We first check if e1 and e2 are linearly independent.
Sincetheyarelinearlyindependent,wecontinuetocheckif
e1 , e2 and a1 arelinearlyindependent. Since
2e1 +3e2 - a1 = 0,

Solution:
wedelete a1 fromSandformanewset S1,

S1 ={e1,e2,e3,a2}.

Then,wecontinuetocheckif e1 , e2 and e3 arelinearly

independent.Theyarelinearlyindependent.Thus,we
finallycheckif e1, e2 e3 and a2 arelinearly
independent.Since

e1 +3e2 + 2e3 - a2 = 0,

Solution:
wedelete

a1

from

S1

andformanewset S ,
2

S 2 ={e1,e2,e3}.Therefore,

= {e1 ,e2 ,e3 }

isthesubsetofSwhichformabasisofforma
3

basisofR .

Solution:
Method2:
Step1:
Theequationis
1
0
2
0
1
c1 0 + c21 + c33 + c40 + c52 = 0
.
0
0
0
1
3

Solution:
Step2:
Theaugmentedmatrixanditsreducedrowechelonmatrixis

1 0 2 0 1 0
0 1 3 0 2 0

0 0 0 1 3 0 .
The1st,the2ndand4thcolumnscontaintheleading1s.
Thus,

{e1 ,e2 ,e3 } formsabasis.Example5(Page 309)

Example5p309
n

n
n
n

S={v1,v2,v3,v4,v5},v1=(1,2,2,1),
v2=(3,0,4,3),v3=(2,1,1,1),v4=(3,3,
9,6),v5=(9,3,7,6).
FindasubsetofSthatisabasisfor
W=spanS.
Step1:
c1(1,2,2,1)+c2(3,0,4,3)+c3(2,1,1,
1)+c4(3,3,9,6)+c5(9,3,7,6)=0

Step2:

0
0

1
0
2
1
1 -
2
0 0
0

3 3

0
2 2

3
5
0
2
2
0 0 0

0 0 0

Step3:
n

Theleading1sappearincolumns1
and2,so{v1,v2}isabasisforW=span
S.

Theorem6.7:(Importantresult)
Let S={v1,v2,K,vn} be a basis for a vector space V and
let

T ={w1,w2,K,wr} is a linear independent set of

vectorsinV.Then, r n .

Corollary6.1:
Let S={v1,v2,K,vn} and T ={w1,w2,K ,wm} be two
basesforavectorspaceV.Then,

n = m

Note:
ForavectorspaceV,thereareinfinitebases.But
the number of vectors in two different bases are
thesame.

Example:
3
R
Forthevectorspace
,

1
v1 = 2 , v2 =
1
3

1
0 , v =
3
2

1
1 , S = {v ,v ,v }
1
2
3

isabasis
0

for R (9094).

Example:
previousexample).Also,
1
e1 = 0 , e2 =
0

0
1 , e =
3
0

0
0 , T = {e ,e ,e }
1
2
3

1

isbasisfor R .

Thereare3vectorsinbothSandT.

Definitionofdimension:
ThedimensionofavectorspaceVisthenumber
ofvectorsinabasisforV. Weoftenwritedim(V)
forthedimensionofV. dim({0})=0.

Example:
1
e1 = 0 , e2 =
0

0
1 , e =
3
0

0
0 , T = {e ,e ,e }
1
2
3

1

Thedimensionof R3is3.

Example68(Page310)

isbasisfor R3.

Example6p310
n
n
n

ThedimensionofR2 is2
ThedimensionofR3 is3
ThedimensionofRn isn.

Example7p310
n
n
n

ThedimensionofP2 is3
ThedimensionofP3 is4
ThedimensionofPn isn+1

Example8p310
n

ThesubspacewofR4consideredin
Ex5hasdimension2.

Theorem6.8:(Importantresult)
IfS isalinearlyindependentsetofvectors ina
finitedimensionalvectorspaceV,thenthereisa
basisTforV,whichcontainsS.
Example9(Page311)

Example9p311
n
n
n
n
n

FindthebasisforR4 thatcontainsthe
vectorv1=(1,0,1,0)andv2=(1,1,1,0).
Let{e1,e2,e3,e4,}bethenaturalbasis
forR4 where
e1=(1,0,0,0),e2=(0,1,0,0)
e3=(0,0,1,0)e4=(0,0,0,0)
c1v1+c2v2+c3e1+c4e2+c5e3+c6e4=0

Example9p311

1
0

0 0 1 1 0 0

1 0 1 0 0 0
0 1 0 -1 0 0

0 0 0 0 1 0

Since leading1sappearincolumns1,2,3and6we
concludethat{v1,v2,e1.e4}isabasisforR4 containing
v1andv2

Theorem6.9:(Importantresult)
Let V be an ndimensional vector space, and let
S={v1,v2,K,vn} beasetofnvectorsinV.

(a) IfSislinearlyindependent,thenSisabasisfor
V.
(b)IfSspansV,thenSisabasisforV.

Example:
Is

1
v1 = 2 , v2 =
1
3

for R ?

1
0 , v =
3
2

1
1 , S = {v ,v ,v }
1
2
3

0

a basis

Note:
n

Solution:
3
R
Since
is a 3dimensional vector space, not like

in the previous example, we only need to examine

3

whether S is linearly independent or S spans R .

We dont need to examine S being both linearly
independentandspansV.

Example:
1
v1 = 3 , v2 =
- 1

Is
3

R ?

2
1 , v =
3
0

4
2 , S = {v ,v ,v }
1
2
3

1

a basis for

Solution:
3
R
Since
isa3dimensionalvectorspace,weonlyneedto
examine whether S is linearly independent or S spans R3.
Because

c1+ 2c2 + 4c3 0

c1v1 +c2v2 + c3v3 = 3c1+ c2 + 2c3 = 0 c1 = c2 = c3 = 0
,
- c1+ c3 0

v1,v2,v3 arelinearlyindependent.Therefore, v1,v2,v3 area

3
R
basisof
.

Example10(Page 312)

Example10p312
n

InEx5.W=spanSisasubspaceofR4,so
dimW 4.SinceScontainsfivevectorswe
concludebyCorollary6.1thatSisnota
basisforW.InEx2,sincedimR4 =4,and
thesetScontainsfourvectors,itispossible
forStobeabasisforR4.IfSislinearly
independentorspansR4,itisabasis
otherwiseitisnotabasis.Thusweneed
onlycheckoneoftheconditionsinThm9.6
notboth.