0% found this document useful (0 votes)
162 views47 pages

Unit 1-Linear Algebra-I

This document discusses linear algebra concepts including vector spaces, subspaces, and linear transformations. It provides 5 examples of vector spaces, such as the set of all n-tuples over a field F and the set of polynomials of degree ≤ n with coefficients in a field. It also defines what qualifies as a subspace and provides 2 exercises asking to prove certain sets are vector spaces. The overall document introduces fundamental linear algebra topics and provides illustrative examples.

Uploaded by

Anjali Sinha
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
162 views47 pages

Unit 1-Linear Algebra-I

This document discusses linear algebra concepts including vector spaces, subspaces, and linear transformations. It provides 5 examples of vector spaces, such as the set of all n-tuples over a field F and the set of polynomials of degree ≤ n with coefficients in a field. It also defines what qualifies as a subspace and provides 2 exercises asking to prove certain sets are vector spaces. The overall document introduces fundamental linear algebra topics and provides illustrative examples.

Uploaded by

Anjali Sinha
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 47

Department of Mathematics

UNIT - I
LINEAR ALGEBRA - I

Topic Learning Objectives:

CE
Upon Completion of this unit, students will be able to:
• Understand the concept of vector spaces and subspaces, basis and
dimension of these vector spaces and subspaces.
• Find the four fundamental subspaces, the rank and nullity of the
matrix and hence verify the rank-nullity theorem.
• Study Linear transformations, with projection, rotation and re-
flection matrices as special cases.
RV
• Obtain the matrix representation, Kernel and image of a linear
transformation.

Introduction:
This unit deals with the vectors, which are functions arising in engineer-
ing systems. We study the linear combination of these vectors that is
the combination of addition of two vectors and multiplication of a vector
by a scalar. In applications of linear algebra, subspaces usually arise in
one of two ways-the set of all solutions to a system of homogeneous lin-
ear equations or as the set of all linear combinations of certain specified
vectors. We study null spaces and column spaces which describe these
situations. Next we study the transformation of vectors from one vector
space to another, whose applications can be seen in computer graphics
and signal processing.
Department of Mathematics

Vector Spaces:
Let F be a field , V be a non-empty set. For every ordered pair α, β ∈ V ,
let there be defined uniquely a sum α + β and for every α ∈ V , and c
in F a scalar product c.α ∈ V . The set V is called a vector space over
the field F , if the following axioms are satisfied for every α, β, γ ∈ V
and for every c, c0 ∈ F .
(i) α + β ∈ V ,
(ii) (α + β) + γ = α + (β + γ),
(iii) Identity element w.r.t addition exists.
i.e, ∃ e ∈ V s.t. α + e = e + α = α,
(iv) Inverse element w.r.t addition exists.

CE
i.e, ∃ α−1 ∈ V s.t. α + α−1 = e = α−1 + α,
(v) α + β = β + α,
(vi) c.(α + β) = c.α + c.β,
(vii) (c + c0 ).α = c.α + c0 .α,
(viii) (c.c0 ).α = c.(c0 .α)
(ix) 1.α = α, ∀α ∈ V , where 1 is the unit element of F .
Examples:
RV
1. Let F be a field and n be a positive integer. Let Vn (F ) be the set
of all ordered n tupples of the elements of the field F . i.e, Vn (F ) =
{(x1 , x2 , ..., xn )/xi ∈ F }.
Define addition and scalar multiplication as below:
(a) α + β = (x1 , x2 , ..., xn ) + (y1 , y2 , ..., yn ) = (x1 + y1 , x2 + y2 , ..., xn + yn )
(b) c.α = c.(x1 , x2 , ..., xn ) = (c.x1 , c.x2 , ..., c.xn ), ∀c ∈ F .
Solution:
(i) α + β = (x1 + y1 , x2 + y2 , ..., xn + yn ) ∈ Vn (F )
(ii) (α + β) + γ
= (x1 + y1 , x2 + y2 , ..., xn + yn ) + (z1 , z2 , ..., zn )
= ((x1 + y1 ) + z1 , (x2 + y2 ) + z2 , ..., (xn + yn ) + zn )
= (x1 + (y1 + z1 ), x2 + (y2 + z2 ), ..., xn + (yn + zn ))
(x1 , x2 , ..., xn ) + (y1 + z1 , y2 + z2 , ..., yn + zn )
= α + (β + γ)
(iii) α + 0 = (x1 , x2 , ..., xn ) + (0, 0, ..., 0)
= (x1 , x2 , ..., xn )
= (0, 0, ..., 0) + (x1 , x2 , ..., xn )
Department of Mathematics

=0+α
∴ 0 = (0, 0, ..., 0) is the identity.
(iv) α + (−α) = (x1 , x2 , ..., xn ) + (−x1 , −x2 , ..., −xn )
= (x1 − x1 , x2 − x2 , ..., xn − xn )
= (0, 0, ..., 0)
=0
∴ −α = (−x1 , −x2 , ..., −xn ) is the additive inverse of α = (x1 , x2 , ..., xn )
(v) α + β = (x1 , x2 , ..., xn ) + (y1 , y2 , ..., yn )
= (x1 + y1 , x2 + y2 , ..., xn + yn )
= (y1 + x1 , y2 + x2 , ..., yn + xn )
= (y1 , y2 , ..., yn ) + (x1 , x2 , ..., xn )
=β+α
CE
(vi) c.(α + β) = c.(x1 + y1 , x2 + y2 , ..., xn + yn )
= (c.(x1 + y1 ), c.(x2 + y2 ), ..., c.(xn + yn ))
= (c.x1 , c.x2 , ..., c.xn ) + (c.y1 , c.y2 , ..., c.yn )
= c.(x1 , x2 , ..., xn ) + c.(y1 , y2 , ..., yn )
= c.α + c.β
(vii) (c + c0 )α = (c + c0 )(x1 , x2 , ..., xn )
= ((c + c0 )x1 , (c + c0 )x2 , ..., (c + c0 )xn )
RV
= (cx1 + c0 x1 , cx2 + c0 x2 , ..., cxn + c0 xn )
= (c.x1 , c.x2 , ..., c.xn ) + (c0 .x1 , c0 .x2 , ..., c0 .xn )
= c.(x1 , x2 , ..., xn ) + c0 .(x1 , x2 , ..., xn )
= cα + c0 α
(viii) (c.c0 ).α = (c.c0 ).(x1 , x2 , ..., xn )
= ((c.c0 ).x1 , (c.c0 ).x2 , ..., (c.c0 ).xn )
= (c.(c0 .x1 ), c.(c0 .x2 ), ..., c.(c0 .xn ))
= c.(c0 .x1 , c0 .x2 , ..., c0 .xn )
= c.(c0 .(x1 , x2 , ..., xn )
= c.(c0 .α)
(ix) 1.α = 1.(x1 , x2 , ..., xn )
= (1.x1 , 1.x2 , ..., 1.xn )
= (x1 , x2 , ..., xn )

Thus Vn (F ) is a vector space over the field F .
Department of Mathematics

Note:
(i) With F = R, V1 (R), V2 (R), V3 (R) are all vector spaces. They are also
denoted as R1 , R2 , R3 respectively. The elements of R1 , R2 , R3 are real
numbers, plane vectors and space vectors respectively.
(ii) If F = R, Vn (R) is denoted as Rn .
If F = C, Vn (C) is denoted as Cn .

2. Show that V = {a + b 2/a, b ∈ Q}, where Q is the set of all
rationals, is a vector space under usual addition and scalar multiplica-
tion.
Solution:

CE √
Let α = a1 + b1 2, √

(i) α + β = (a1 + b1 2) +√(a2 + b2 2) ∈ √V



β = a2 + b2 2,√γ = a3 + b3 2 ∈ V and c, c0 ∈ Q

(ii) (α + β) + γ = ((a1 + b1 2) + (a2 + b2 2)) + (a3 + b3 2) = α + (β + γ)
(iii) 0 is the additive√identity, as 0 + α = α = α + 0.
(iv) −α = −a1 − b1 2 is the additive inverse of α as α + (−α) = 0 =
(−α) + α. √ √
(v) α + β = (a1 + b1 2) + √ (a2 + b2 2) = √β + α.
RV
(vi) c.(α + β) = c.((a1 + b1 2) + √ (a2 + b2 2))0 = c.α + c.β.
(vii) (c + c0 )α = (c + c0 )(a1 +√ 1 2) = 0c.α + c α.
b
0 0
(viii) (c.c )α = (c.c )(a
√ 1 + b1 2) =√c.(c .α).
(ix)1.α = 1.(a1 + b1 2) = a1 + b1 2 = α.
Thus V is a vector space over Q .

3. Let V be the set of all polynomials of degree ≤ n, with coefficients


in the field F , together with zero polynomial. Then show that V is a
vector space under addition of polynomials and scalar multiplication of
polynomials with the scalar c ∈ F defined by c(a0 + a1 x + ... + an xn ) =
ca0 + ca1 x + ... + can xn .
Solution:
(i) Sum of polynomials is again a polynomial
(ii) Sum of polynomials will be associative.
(iii) 0 is the additive identity.
(iv) If α = a0 + a1 x + ... + an xn then −α = −a0 − a1 x − ... − an xn is
the additive inverse.
Department of Mathematics

(v) Sum of polynomials is commutative.


(vi) c.(α + β) = c.α + c.β will hold.
(vii) (c + c0 ).α = c.α + c0 .α will hold.
(viii) (c.c0 ).α = c.(c0 .α) will hold.
(ix) 1.α = 1.(a0 + a1 x + ... + an xn ) = a0 + a1 x + ... + an xn = α.
Thus V is a vector space over F .
 
x y
4. Let V = { /x, y ∈ C} under usual addition and scalar
−y x
multiplication, with field C of complex numbers. Show that V is a vec-
tor space.

CE
Solution:
Let α =

x1 y1
−y1 x1

,β=

x2 y2
−y2 x2
and c1 = a1 + b1 i, c2 = a2 + b2 i ∈ C.
(i) α + β ∈ V ,
(ii) (α + β) + γ = α + (β  + γ),

,γ=

x3 y 3
−y3 x3

∈V

0 0 0 0
(iii) α + =α= + α.
 0 0 0 0
RV

0 0
∴ is the additive identity.
0 0  
−x1 −y1
(iv) −α = is the additive inverse.
y1 −x1
(v) α + β = β + α will hold.
(vi) c.(α + β) = c.α + c.β will hold.
(vii) (c + c0 ).α = c.α + c0 .α will hold.
(viii) (c.c0 ).α = c.(c0 .α) willhold. 
x1 y 1 x1 y 1
(ix) 1.α = 1. = = α.
−y1 x1 −y1 x1
Thus V is a vector space over C.

5. Let R+ be the set of all positive integers. Define the operations


of addition and scalar multiplication as below:
α + β = αβ, ∀ α, β ∈ R+
c.α = αc , α ∈ R+ and c ∈ R
Show that R+ is a vector space over the real field.
Department of Mathematics

Solution:
(i) α + β = αβ ∈ R+ .
(ii) (α + β) + γ = (αβ) + γ = (αβ)γ = α(βγ) = α + βγ = α + (β + γ).
(iii) α + 1 = α.1 = α = 1.α = 1 + α.
1 1 1 1
(iv) α + = α. = 1 = .α = + α.
α α α α
1
∴ is the additive inverse of α.
α
(v) α + β = α.β = β.α = β + α.
(vi) c.(α + β) = c.(αβ) = (αβ)c = αc β c = αc + β c = c.α + c.β.
0 0
(vii) (c + c0 )α = α(c+c ) = αc αc = αc + αc0 = c.α + c0 .α.
0 0 0 0
(viii) (c.c0 ).α = α(c.c ) = α(c .c) = (αc )c = c(αc ) = c.(c0 α)

CE
(ix) 1.α = α1 = α, where 1 is the unit element of R+ .

Exercise:
1. If R is the field of real numbers and V is the set of vectors in a plane,
further if addition of vectors is the internal binary composition in V
and the multiplication of elements of R with those of V as the external
composition, prove that V (R) is a vector space.
RV
2. Let R be the field of real numbers and let P3 be the set of all poly-
nomials of degree at most 3, over the field R. Prove that P3 is a vector
space over the field R.

Subspaces:
A non empty subset W of a vector space V over a field F is called a
subspace of V , if W is itself a vector space over F , under the same
operations of addition and scalar multiplication as defined in V .
Note:
(i) The set {0} consisting of zero vector of V is a subspace of V .
(ii) The whole vector space V , itself is a subspace of V .
These two subspaces are called trivial or improper subspaces of V .
Any subspace W of V different from {0} and V is called a proper sub-
space of V .

Theorem: A non empty subset W of a vector space V over a field


F is a subspace of V , if and only if (i) ∀α, β ∈ W , α + β ∈ W (ii)
Department of Mathematics

∀c ∈ F, α ∈ W, c.α ∈ W .

Examples:
1. Verify whether W = {f (x)/2f (0) = f (1)} over 0 ≤ x ≤ 1, is a
subspace of V = {all functions} over the field R.
Solution:
Let f1 , f2 ∈ W .
Thus 2f1 (0) = f1 (1) and 2f2 (0) = f2 (1)
Consider, 2(f1 + f2 )(0) = 2[f1 (0) + f2 (0)]
= 2f1 (0) + 2f2 (0)
= f1 (1) + f2 (1)

CE
= (f1 + f2 )(1)
Thus, f1 + f2 ∈ W . i.e., W is closed under vector addition.
Consider, 2(cf1 )(0) = (2c)f1 (0)
= c.2f1 (0)
= c.f1 (1)
= (cf1 )(1).
Thus cf1 ∈ W i.e., W is closed under scalar multiplication.
RV
Hence W is a subspace.

2. Is the subset W = {(x1 , x2 , x3 )/x21 + x22 + x33 ≤ 1} of V3 (R) a


subspace of V3 (R)?
Solution:
Let α = (1, 0, 0), where 12 + 02 + 02 = 1 and β = (0, 1, 0), where
02 + 12 + 02 = 1 be two vectors in W .
Consider α+β = (1, 0, 0)+(0, 1, 0) = (1, 1, 0), where, 12 +12 +02 = 2  1.
Hence α + β 6∈ W . ∴ W is not a subspace.

3. Show that the subset W = {(x1 , x2 , x3 )/x1 + x2 + x3 = 0} of the


vector space V3 (R) is a subspace of V3 (R).
Solution:
Let α = (x1 , x2 , x3 ), β = (y1 , y2 , y3 ) be any two elements of W .
Then, x1 + x2 + x3 = 0 and y1 + y2 + y3 = 0.
Consider, c1 α + c2 β = c1 (x1 , x2 , x3 ) + c2 (y1 , y2 , y3 )
= (c1 x1 , c1 x2 , c1 x3 ) + (c2 y1 , c2 y2 , c2 y3 )
Department of Mathematics

= (c1 x1 + c2 y1 , c1 x2 + c2 y2 , c1 x3 + c2 y3 )
To show that c1 α + c2 β ∈ W , we have to show that the sum of the
components of c1 α + c2 β is zero.
∴ consider c1 x1 + c2 y1 + c1 x2 + c2 y2 + c1 x3 + c2 y3
= c1 (x1 + x2 + x3 ) + c2 (y1 + y2 + y3 )
= c1 .0 + c2 .0
=0
∴ c1 α + c2 β ∈ W , hence W ia a subspace of V3 (R).

4. Verify whether W = { Polynomial of degreee three } defined on


0 ≤ x ≤ 1 is a subspace of the vector space V = { all polynomials }

CE
over R.
Solution:
The set of all polynomials of degree three is not a subspace, as the sum
of two polynomials of degree three need not be of degree three.
∵ f1 (x) = 3x3 − 4x2 + 2x + 1, f2 (x) = −3x3 + 3x2 + 2x + 5
=⇒ f1 (x) + f2 (x) = −x2 + 4x + 6 which is not a polynomial of degree
three.
RV
Hence W is not a subspace of V

5. Let V = R3 , the vector space of all ordered triplets of real


numbers, over the field of real numbers. Show that the subset W =
{(x, 0, 0)/x ∈ R} is a subspace of V = R3 .
Solution:
The element 0 = (0, 0, 0) ∈ W .
Thus W is non-empty.
Let α1 = (x1 , 0, 0) and α2 + (x2 , 0, 0) be any two elements of W .
Then α1 + α2 = (x1 , 0, 0) + (x2 , 0, 0) = (x1 + x2 , 0, 0) ∈ W .
∴ W is closed under addition.
Again, for any scalar c ∈ R,
c.α1 = c.(x1 , 0, 0) = (cx1 , 0, 0) ∈ W
∴ W is closed under scalar multiplication.
Hence W is a subspace of R3 .

Exercise:
Department of Mathematics

1. Let H = {a−3b, b−a, a, b} where a and b are arbitrary scalars. Show


that H is a subspace of R4 .
2. Let V be a vector space of all 2 x 2 matrices over reals. Determine
whether W is a sub-space of V or not, where
(i) W consists of all matrices with non-zero determinant.
2
(ii) W consists
 of all
 matrices A such that A = A.
a 0
(iii) W = { , where a, b ∈ R}.
0 b
Answer: (i) W is not a subspace. (ii) W is not a subspace. (iii) W is
a subspace.

CE
Linear Combination
Let V be a vector space over the field F and α1 , α2 , ..., αn be any
n vectors of V . The vector of the form, c1 α1 + c2 α2 + ... + cn αn ,
where c1 , c2 , ..., cn ∈ F , is called a linear combination of the vectors
α1 , α2 , ..., αn .
Linear Span of S
Let S be a non empty subset of a vector space V (F ). The set of all
RV
linear combinations of finite number of elements of S is called the linear
span of S and is denoted by L[S].
i.e, L[S] = {c1 α1 +c2 α2 +...+cn αn /ci ∈ F, αi ∈ S, i = 1, 2, ..., n and n is any pos
If α ∈ L[S], then α is of the form, α = c1 α1 + c2 α2 + ... + cn αn , for some
scalars c1 , c2 , ..., cn ∈ F .
Theorem: Let S be a non-empty subset of a vector space V [F ]. Then
(i) L[S] is a subspace of V
(ii) S ⊆ L[S]
(iii) L[S] is the smallest subspace of V containing S.
Linear Dependence and Independence
A set {α1 , α2 , ..., αn } of vectors of a vector space V [F ] is said to be lin-
early dependent if there exists scalars c1 , c2 , ..., cn ∈ F , not all zero such
that c1 α1 + c2 α2 + ... + cn αn = 0.
A set {α1 , α2 , ..., αn } of vectors of a vector space V [F ] is said to be lin-
early independent if c1 α1 + c2 α2 + ... + cn αn = 0 =⇒ c1 = c2 = ... =
cn = 0.
Department of Mathematics

Examples:
1. Show that the vectors e1 = (1, 0, 0, ..., 0), e2 = (0, 1, 0, ...0), ...en =
(0, 0, 0, ..., 1) of the vector space Vn (R) are linearly independent.
Solution:
Let c1 , c2 , ..., cn ∈ R
Consider c1 e1 + c2 e2 + ... + cn en = 0
=⇒ c1 (1, 0, 0, ..., 0)+c2 (0, 1, 0, ...0)+...+cn (0, 0, 0, ..., 1) = (0, 0, 0, ..., 0)
=⇒ (c1 , c2 , ..., cn ) = (0, 0, 0, ..., 0)
=⇒ c1 = 0, c2 = 0, ..., cn = 0
i.e., e1 , e2 , e3 , ..., en are linearly independent.

CE
2. Show that the set S = {(1, 0, 1), (1, 1, 0), (−1, 0, −1)} is linearly
dependent in V3 (R).
Solution:
Consider c1 (1, 0, 1) + c2 (1, 1, 0) + c3 (−1, 0, −1) = (0, 0, 0)
=⇒ (c1 + c2 − c3 , c2 , c1 − c3 ) = (0, 0, 0)
=⇒ c1 + c2 − c3 = 0, c2 = 0, c1 − c3 = 0
=⇒ c1 = 1, c2 = 0, c3 = 1
RV
Thus there exists, not all zeros, scalars, such that c1 (1, 0, 1)+c2 (1, 1, 0)+
c3 (−1, 0, −1) = (0, 0, 0)
∴ S is linearly dependent.

Note:
1.The set {(x1 , x2 , x3 ), (y1 , y2 , y3 ), (z1 , z2 , z3 )} of vectors of the vector
x1 x2 x3

space V3 (R) is linearly dependent iff y1 y2 y3 = 0
z1 z2 z3
2. Two vectors α, β ∈ V2 (R) are linearly dependent iff α = kβ for some
non zero k ∈ R
3. A set of vectors of V , containing the zero vector is linearly depen-
dent.
4. The set consisting of a single vector α of V is linearly independent
iff α 6= 0.

Exercise:
Department of Mathematics

1. Determine if the following sets  of vectors


 are linearly
 independent

1 0 0 0 0 1
or linearly dependent. v1 = , v2 = , v1 =
  0 0 0 0 0 0
0 0 0
.
0 1 0
Answer: Linearly independent.
2. Check
   forthe linear  independence/dependence of the column vectors
1 2 1
 2 ,  6 ,  3 
0 2 5
Answer: Linearly independent.

CE
3. Let V be a vector space of real valued derivable functions on (0, ∞),
then show that the set S = {x2 ex , xex , (x2 + x − 1)ex } is linearly inde-
pendent.
4. In a P2 let v1 = 2t2 + t + 2; v2 = t2 − 2t; v3 = 5t2 − 5t + 2;
v4 = −t2 − 3t − 2. Determine if the vector u = t2 + t + 2 belongs to
the span{v1 , v2 , v3 , v4 }. (Here P2 is the vector space of all polynomials
of degree at the most n over the field of real numbers).
Answer: Yes, u = t2 + t + 2 belongs to the span{v1 , v2 , v3 , v4 }.
RV
Basis
A subset B of a vector space V [F ] is called a basis of V if
(i) B is a linearly independent set
(ii) L[B] = V
i.e., a basis of a vector space V [F ] is linearly independent subset which
spans the whole space.

Note: The zero vector cannot be an element of a basis of a vector


space because a set of vectors with zero vector is always linearly depen-
dent.

Examples:
1. Show that the vectors e1 = (1, 0, 0, ..., 0), e2 = (0, 1, 0, ...0), ...en =
(0, 0, 0, ..., 1) of the vector space Vn (R) form a basis of Vn (R).
Solution:
Department of Mathematics

Consider S = {e1 , e2 , ..., en }


c1 e1 + c2 e2 + ... + cn en = 0
=⇒ c1 (1, 0, ..., 0) + c2 (0, 1, ..., 0) + ... + cn (0, 0, ..., 1) = (0, 0, ..., 0)
=⇒ (c1 , c2 , ..., cn ) = (0, 0, ..., 0)
c1 = 0, c2 = 0, ..., cn = 0
Hence S is linearly independent.
Further, any vector (x1 , x2 , ..., xn ) ∈ Vn (R) can be expressed as a linear
combination of the elements of S, as (x1 , x2 , ..., xn ) = x1 e1 + x2 e2 + ... +
xn en . Hence L[S] = Vn (R)
∴ S is a basis of Vn (R).

CE
Standard Basis
The basis S = {e1 , e2 , ..., en } of the vector space Vn (R) is called the
standard basis.
example: The vectors e1 = (1, 0, 0), e2 = (0, 1, 0), e3 = (0, 0, 1) of V3 (R)
form a basis of V3 (R), and is called the standard basis.

2. Show that the set B = {(1, 1, 0), (1, 0, 1), (0, 1, 1)} is a basis of
RV
the vector space V3 (R).
Solution:
Let B = {(1, 1, 0), (1, 0, 1), (0, 1, 1)}
Consider, c1 (1, 1, 0) + c2 (1, 0, 1) + c3 (0, 1, 1) = (0, 0, 0)
=⇒ (c1 + c2 , c + 1 + c3 , c2 + c3 ) = (0, 0, 0)
=⇒ c1 + c2 = 0, c1 + c3 = 0, c2 + c3 = 0
=⇒ c1 = 0, c2 = 0, c3 = 0
∴ B is linearly independent.
Let (x1 , x2 , x3 ) ∈ V3 (R) be arbitrary.
Let c1 , c2 , c3 ∈ R, such that
(x1 , x2 , x3 ) = c1 (1, 1, 0) + c2 (1, 0, 1) + c3 (0, 1, 1)
(x1 , x2 , x3 ) = (c1 + c2 , c1 + c3 , c2 + c3 )
=⇒ x1 = c1 + c2 , x2 = c1 + c3 , x3 = c2 + c3
x1 + x2 − x3 x1 − x2 + x3 −x1 + x2 + x3
=⇒ c1 = , c2 = , c3 =
2 2 2
∴ (x1 , x2 , x3 ) = x1 +x22 −x3 (1, 1, 0) + x1 −x22 +x3 (1, 0, 1) + −x1 +x2 2 +x3 (0, 1, 1)
∴ L[B] = V3 (R)
Department of Mathematics

Exercise:
1. Let W be a subspace of R4 spanned by the vectors α1 = (1, 2, 2, 1);
α2 = (0, 2, 0, 1); α3 = (−2, 0, −4, 3). Prove that α1 , α2 , α3 form a basis
for W.
2. Find the basis for the subspace spanned by the vectors
 T  T  T
v1 = 1 0 0 −1 , v2 = 1 0 −1 0 , v3 = 1 −1 0 0 ,
 T  T  T
v4 = 0 1 −1 0 , v5 = 0 0 1 −1 , v6 = 0 1 0 −1 .
Answer: Basis= {v1 , v2 , v3 }.
3. Show that the vectors α1 = (1, 0, −1), α2 = (1, 2, 1), α3 = (0, −3, 2)
form a basis for R3 . Express each of the standard basis vector as a linear

CE
combination of α1 , α2 , α3 .
Answer: e1 = 0.7α1 + 0.3α2 + 0.2α3 , e2 = −0.2α1 + 0.2α2 − 0.2α3 ,
e1 = −0.3α1 + 0.3α2 + 0.2α3 .

Dimension of a vector space V


The dimension of a finite dimensional vector space V over F is the num-
ber of elements in any basis of V and is denoted by d[V ].
RV
example: Vn (R) is a n dimensional space.
V3 (R) is a 3 dimensional space.

Finite dimensional space


A vector space V [F ] is said to be a finite dimensional space if it has a
finite basis.

Note:
(i) Any two bases of a finite dimensional vector space V have the same
finite number of elements.
(ii) A vector space which is not finitely generated may be called an in-
finite dimensional space.
(iii) In an n dimensional vector space V (F )
(a) any n + 1 elements of V are linearly dependent.
(b) no set of n − 1 elements can span V .
(iv) In an n dimensional vector space V (F ) any set of n linearly inde-
pendent vectors is a basis.
Department of Mathematics

(v) Any linearly independent set of elememts of a finite dimensional


vector space V is a part of a basis.
(vi) For n vectors of n−dimensional vector space V , to be a basis, it is
sufficient that they span V or that they are Linearly independent.

Examples:
1. Let A = {(1, −2, 5), (2, 3, 1)} be a linearly independent subset of
V3 (R). Extend this to a basis of V3 (R).
Solution:
Let α1 = (1, −2, 5), α2 = (2, 3, 1)

CE
Let S be the subspace spanned by {α1 , α2 }
∴ S = {c1 α1 + c2 α2 /c1 , c2 ∈ R}
c1 α1 + c2 α2 = c1 (1, −2, 5) + c2 (2, 3, 1)
= (c1 + 2c2 , −2c1 + 3c2 , 5c1 + c2 )
∴ S = {(c1 + 2c2 , −2c1 + 3c2 , 5c1 + c2 )/c1 , c2 ∈ R}
Chose a vector of V3 (R), outside of S.
(1, 0, 0) 6∈ S
∴ the set A = {(1, −2, 5), (2, 3, 1), (1, 0, 0)} is a basis of V3 (R).
RV
2. Given two linearly independent vectors (2, 1, 4, 3)&(2, 1, 2, 0), find
a basis of V4 (R) that includes these two vectors.
Solution:
α1 = (2, 1, 4, 3), α2 = (2, 1, 2, 0)
S = {c1 α1 + c2 α2 /c1 , c2 ∈ R}
Choose α3 = (1, 0, 0, 0) & α4 = (0, 1, 0, 0) 6∈ S
∴ {α1 , α2 , α3 , α4 } is a basis of V4 (R).

The non-zero rows of a row-reduced echelon form of a matrix are


linearly independent.

3. Test the following set of vectors for linear dependence in V3 (R).


{(1, 0, 1), (0, 2, 2), (3, 7, 1)}. Do they form a basis?
Solution:
Department of Mathematics

 
1 0 1
Consider the matrix A = 0 2 2
3 7 1
|A| = 1(2 − 14) − 0(0 − 6) + 1(0 − 6) = −18 6= 0.
Therefore the given set in linearly independent.
Any three vectors in V3 (R) which are linearly independent is a basis of
V3 (R).

4. Does the set S = {(1, 2, 3), (3, 1, 0), (−2, 1, 3)} form a basis of R3 .
Solution:  

CE 1 2 3
Consider A =  3 1 0
−2 1 3
|A| = 1(3 − 0) − 2(9 + 0) + 3(3 + 2) = 0.
∴ S is linearly dependent and hence is not a basis of R3 .

5. Show that the vectors (1, 1, 2, 4), (2, −1, −5, 2), (1, −1, −4, 0), (2, 1, 1, 6)
are linearly dependent in R4 and extract a linearly independent subset.
RV
Also find the dimension and a basis of the subspace spanned by them.
Solution:  
1 1 2 4
2 −1 −5 2
Consider A =  1 −1 −4 0

2 1 1 6
R2 =R2 − 2R1 ; R3 = R3 − R1 ; R4 = R4 − 2R1 =⇒
1 1 2 4
0 −3 −9 −6
A= 0 −2 −6 −4

0 −1 −3 −2
R2 =R2 ÷ (−3) =⇒ 
1 1 2 4
0 1 3 2
A= 0 −2 −6 −4

0 −1 −3 −2
R3 = R3 + 2R2 ; R4 = R4 + R2 =⇒
Department of Mathematics

 
1 1 2 4
0 1 3 2
A= 0 0 0 0

0 0 0 0
The final matrix is in echelon form and the rank of A is 2. Therefore
the given vectors are linearly dependent.
The corresponding non-zero rows of the initial matrix are (1, 1, 2, 4) &(2, −1, −
which are linearly independent.
The dimension of the subspace spanned by the vectors is 2. These two
vectors form a basis of the subspace.

CE
6. Let S be the subspace of R3 defined by S = {(a, b, c)/a+b+c = 0}.
Find a basis and dimension of S.
Solution:
S 6= R3 [∵ (1, 2, 3) ∈ R3 but(1, 2, 3) 6∈ S]
α = (1, 0, −1) & β = (1, −1, 0) ∈ S and further they are independent.
∴ d[S] = 2 and hence {α, β} forms a basis of S.
RV
7. Show that the field C of complex numbers is a vector space over
the field R of reals. What is its dimension?
Solution:
C = {a + ib/a, b ∈ R}
C is closed under 0 +0 .
C is associative under 0 +0 .
0 + i0 is the identity w.r.t 0 +0 .
−a − ib is the inverse of a + ib.
C is commutative.
Hence (C, +) is an abelian group.
c.((a1 + ib1 ) + (a2 + ib2 )) = c.(a1 + ib1 ) + c.(a2 + ib2 ) ∈ C
(c1 + c2 ).(a1 + ib1 ) = c1 .(a1 + ib1 ) + c2 .(a1 + ib1 ) ∈ C
0 0
(c.c ).(a1 + ib1 ) = c.(c .(a1 + ib1 )) ∈ C
0 0
1 is the unity
Therefore C is a vector space over R
Let α ∈ C, α = a + ib s.t. a, b ∈ R
∴ α = 1.a + i.b = a.1 + b.i
Department of Mathematics

i.e., every element of C is a linear combination of the elements 1, i. That


is {1, i} generates C.
Further c1 .1 + c2 .i = 0 =⇒ c1 = 0&c2 = 0
∴ {1, i} is linearly independent.
∴ {1, i} is a basis of C and d[C] = 2.

8. Let V be the vector space of 2 × 2 symmetric matrices over the


field F . Show that d[V ] = 3.
Solution:  
a b
Let A = ∈ V, a, b, c ∈ F .
b c

CE
Set a = 1, b = 0, c = 0; a = 0,b = 1,
We get three matrices E1 =
1 0
0 0
 c = 0; a = 0,b = 0, c =
, E2 =
We shall showthat these elements of V form a basis.
Let A =
a b
b c 
∈ V be arbitrary.
   
0 1
1 0


, E3 =


 1 
0 0
0 1

a b 1 0 0 1 0 0
Then, A = =a +b +c
b c 0 0 1 0 0 1
RV
Thus {E1 , E2 , E3 } generates V .
Suppose c1 E1+ c2 E2 + c 0, c1 , c2 , c3 ∈ F 
3 E3 = 
1 0 0 1 0 0 0 0
=⇒ c1 + c2 + c3 =
 0 0  1  0 0 1 0 0
c1 c2 0 0
=⇒ = =⇒ c1 = c2 = c3 = 0.
c2 c3 0 0
∴ {E1 , E2 , E3 } is linearly independent.
Hence {E1 , E2 , E3 } is a basis of V and d[V ] = 3.

9. Find the basis and dimension of the subspace spanned by the


subset       
1 −5 1 1 2 −4 1 −7
S ={ , , , } of the vector space
−4 2 −1 5 −5 7 −5 1
of all 2 × 2 matrices over R.
Solution:
Let α, β, γ, δ are the matrices of S.
Then the coordinates of α, β, γ, δ w.r.t standard basis are
Department of Mathematics

(1, −5, −4,2), (1, 1, −1, 5),(2, −4, −5, 7), (1, −7, −5, 1).
1 −5 −4 2
1 1 −1 5
Consider 2 −4 −5 7

1 −7 −5 1
R2 = R2 − R1 , R3 = R3 − 2R1 , R4 = R4 − R1 =⇒
 
1 −5 −4 2
0 6 3 3
 
0 6 3 3
0 −2 −1 −1


CE
R3 = R3 − R2 , R4 = 3R4 + R2 =⇒

1 −5 −4 2
0 6

0 0
3 3
0 0

0 0 0 0
RV
The final matrix has two non-zero rows.
∴ d(subspace) = 2.
Further the matrices corresponding
  to the non-zero rows in the final
1 −5 0 6
matrix are , .
−4 2 3 3
10. In a vector space V3 (R), let α = (1, 2, 1), β = (3, 1, 5) & γ =
(−1, 3, −3). Show that the subspace spanned by {α, β} & {α, β, γ} are
the same.
Solution:  
1 2 1
Consider, A =  3 1 5 
−1
3 −3
1 2 1

|A| = 3 1 5 = 1(−3−15)−2(−9+5)+1(9+1) = −18+8+10 = 0
−1 3 −3
∴ {α, β, γ} is linearly dependent.
Let γ = c1 α + c2 β
Department of Mathematics

(−1, 3, −3) = c1 (1, 2, 1) + c2 (3, 1, 5)


=⇒ (−1, 3, −3) = (c1 + 3c2 , 2c1 + c2 , c1 + 5c2 )
=⇒ c1 + 3c2 = −1, 2c1 + c2 = 3, c1 + 5c2 = −3
Solving these equations, we get c1 = 2, c2 = −1.
∴ γ ∈ subspace spanned by {α, β}.
∴ the subspace spanned by {α, β} & {α, β, γ} are the same.

Exercise:
1. Let V be a space of 2 x2  matrices over
  R and let W be the  sub-space

1 −5 1 1 2 −4 1 −7
generated by , , and .
 −4 2   −1 5 −5 7 −5 1

CE
Show that (i) {
1 −5
−4 2

Four Fundamental Subspaces


Null Space
,
0 2
1 1
} forms a basis set. (ii) dimW = 2.

The null space of a m × n matrix A, written as Nul A, is the set of all


solutions to the homogrneous equation Ax = 0.
RV
Theorem:
The null space of a m × n matrix A, is a subspace of Rn . Equivalently,
the set of all solutions to a system Ax = 0 of m homogeneous linear
equations in n unknowns is a subspace of Rn .
Examples:  
  5
1 −3 −2
1. Let A = and u = 3 

−5 9 1
−2
Determine if u belongs to the null space of A.
Solution:  
  5    
1 −3 −2   5−9+4 0
Au = 3 = =
−5 9 1 −25 + 27 − 2 0
−2
∴ u is in Nul A.  
−3 6 −1 1 −7
2. Find a spanning set for the null space of the matrix  1 −2 2 3 −1
2 −4 5 8 −4
Solution:
Department of Mathematics

Consider Ax = 0.
Reducing
 A to echelon form
−3 6 −1 1 −7
A =  1 −2 2 3 −1,
2 −4 5 8 −4
3R
 2 + R 1 , 3R3 + 2R1 =⇒

−3 6 −1 1 −7
 0 0 5 10 −10
0 0 13 26 −26
 2 ÷ 5, R3 ÷ 13 =⇒
R
−3 6 −1 1 −7

=⇒
CE
 0 0 1 2 −2
0 (0 0 0 0
−3x1 + 6x2 − x3 + x4 − 7x5 = 0
x3 + 2x4 − 2x5 = 0
1 1
=⇒ x1 = 2x2 − 3 (−2x4 + 2x5 ) + 3 x4 − 3 x5 7
=⇒
(
x1 = 2x2 − 31 x3 + 13 x4 − 37 x5
x3 = −2x4 + 2x5

=⇒  x  2 + x4 − 3x5 , with
1 = 2x  x2 , 
x4 , x5 as free
 variables.
    
x1 2x2 + x4 − 3x5 x1 2 1 −3
RV
x2   x x 1 0
2  2 0
        
       
∴ x3  =  −2x4 + 2x5  =⇒ x3  = x2 0+x4 −2+x5  2 
          

x4   x4  x4  0 1 0
x5 x5 x5 0 0 1
= x2 u + x4 v + x5 w.
Every linear combination of u, v, w is an element of NulA. Thus {u, v, w}
is a spanning set for NulA.
Column Space:
The Column space of an m × n matrix A, written as ColA, is the set of
all linear combinations of the columns of A.
If A = [a1 , a2 , ..., an ], then ColA = span{a1 , a2 , ..., an }.
Theorem: The column space of a m × n matrix A is a subspace of Rm .

Examples:
1. Find a matrix A such that W = ColA.
Department of Mathematics

 
6a − b
W = { a + b  : a, b ∈ R}
−7a
Solution:
W can be  written
 as
     
6 −1 6 −1
W = {a 1 + b 1 : a, b ∈ R} = span { 1 , 1 }
      
−7 0 −7 0
Using the vectors
 in the spanning set as the columns of A, we get
6 −1
A= 1 1 .

CE
−7 0
Then W = ColA as desired.
Note: The column space of an m × n matrix A is all of Rm iff the
equation Ax = b has a solution for each b in Rm .


2 4 −2 1

2. Let A = −2 −5 7 3
RV
3 7 −8 6
(a) If the column space of A is a subspace of Rk , what is k?
(b) If the null space of A is a subspace of Rk , what is k?
Solution:
(a) m = 3, ColA is a subspace of Rm , where m = 3. (b) n = 4, NullA
is a subspaceof Rn , where n =4.
2 4 −2 1
3. Let A = −2 −5 7 3, find a non-zero vector in ColA and a

3 7 −8 6
non-zero vector in NullA.
Solution:  
2
Any column of A belongs to ColA. eg. −2 ∈ ColA.

3
Consider
 Ax = 0 
2 4 −2 1 : 0
A = −2 −5 7 3 : 0
3 7 −8 6 : 0
Department of Mathematics

 
2 4 −2 1 : 0
R2 + R1 ; 2R3 − 3R1 =⇒ A = 0 −1 5 4 : 0
 0 2 −10  9 :0
2 4 −2 1 : 0
R3 + 2R2 =⇒ A = 0 −1 5 4 : 0
0 0 0 17 : 0
=⇒ 2x1 + 4x2 − 2x3 + x4 = 0; −x2 + 5x3 + 4x4 = 0; 17x4 = 0
=⇒ x4 = 0, x2 = 5x3 , x1 = −9x3 =⇒ x3 is a free variable.
Let x3 = 1, then x1 = −9, x2 = 5, x4 = 0. The vector x = (−9, 5, 1, 0) ∈
NulA.  
  3  
2 4 −2 1 3

CE
solution:

3


7 −8 6

−2
4. With A = −2 −5 7 3 , u =  , v = −1,

−1
0

 
 3
 
3
(a) Determine if u is in NulA. Could u be in ColA? (b) Determine if v
is in ColA. Could v be in NulA?
   
2 4 −2 1   0 0
−2    
(a) Consider Au = −2 −5 7 3  = −3 6
= 0 .
RV
−1
3 7 −8 6 3 0
0
∴u∈ / NulA.
∵ u has 4 entries and ColA is subspace of R3 , u ∈
/ ColA.
(b)
 Consider [A v] 
2 4 −2 1 : 3
−2 −5 7 3 : −1
3 7 −8 6 : 3
R2 + R1 ; 2R3 − 3R1 =⇒
 
2 4 −2 1 : 3
0 −1 5 4 : 2 
0 2 −10 9 : −6
R3 + 2R2 =⇒
 
2 4 −2 1 : 3
0 −1 5 4 : 2
0 0 0 17 : 0
=⇒ Ax = v is consistent.
Department of Mathematics

∴ v is in ColA.
∵ v has 3 entries and NulA is a subspace of R4 , v ∈ / NulA.
Row Space:
If A is an m × n matrix, each row of A has n entries and this can be
identified with a vector in Rn . The set of all linear combinations of the
row vectors is called the row space of A and is denoted by RowA. Each
row has n entries, so RowA is a subspace of Rn . Since the rows of A are
identified with the columns of AT , we could also write ColAT in place
of RowA.
Examples:  
−2 −5 8 0 −17

CE
1. Let A = 

Solution:
1 3 −5
 3 11 −19
1 7 −13 5 − 3

r1 = (−2, −5, 8, 0, 17)


r2 = (1, 3, −5, 1, 5)
1
7
5 
.
1 

r3 = (3, 11, −19, 7, 1)


r4 = (1, 7, −13, 5, −3)
RV
The row space of A is the subspace of R5 spanned by {r1 , r2 , r3 , r4 }.
That is RowA = Span{r1 , r2 , r3 , r4 }.
Theorem:
If two matrices A and B are equivalent, then their row spaces are the
same. If B is in echelon form, then non-zero rows of B form a basis for
the row space of A as well as for that of B.

2. Find the basis for the row space


 of
−2 −5 8 0 −17
1 3 −5 1 5 
A=  3 11 −19

7 1 
1 7 −13 5 − 3
Solution:
To find the basis for the row space, reduce A to echelon form.
Department of Mathematics

 
−2 −5 8 0 −17
1 3 −5 1 5 
A=  3 11 −19

7 1 
1 7 −13 5 − 3
R1 ←→ R2 =⇒
 
1 3 −5 1 5
−2 −5 8 0 −17
 
 3 11 −19 7 1 
1 7 −13 5 − 3
R
 2 + 2R 1 , R3 − 3R1 ,R4 − R1 =⇒
1 3 −5 1 5


CE
0 1 −2 2 −7 

0 2 −4 4 −14
0 4 −8 4 −8
R3 − 2R2 , R4 − 4R2 =⇒
1 3 −5 1
0 1 −2 2 −7

0 0 0
5


0 0
0 0 0 −4 20
RV
R3 ←→ R4 =⇒ 
1 3 −5 1 5
0 1 −2 2 −7
B= 0 0 0 −4 20 

0 0 0 0 0
The first three rows of B form a basis for the row space of A(as well as
for the row space of B).
Basis for RowA = {(1, 3, −5, 1, 5), (0, 1, −2, 2, −7), (0, 0, 0, −4, 20)}
3. Find
 the basis for the rowspace of
2 −3 6 2 5
−2 3 −3 −3 −4
A=  4 −6 9

5 9
−2 3 3 −4 1
Solution:
R2 + R1 , R3 − 2R1 , R4 + R1 =⇒
Department of Mathematics

 
2 −3 6 2 5
0 0 3 −1 1 
A= 0 0 −3 1 −1

0 0 9 −2 6
R3 +R2 , R4 − 3R2 =⇒
2 −3 6 2 5
0 0 3 −1 1
A= 0 0 0 0 0

0 0 0 1 3
 
2 −3 6 2 5
0 0 3 −1 1
R3 ←→ R4 =⇒ A = 

CE 0 0 0 1 3
0 0 0 0 0

The first three rows of B form the basis for the rowspace of A.
RowA = {(2, −3, 6, 2, 5), (0, 0, 3, −1, 1), (0, 0, 0, 1, 3)}

Left null space


The left null space of an m × n matrix A written as Nul(AT ), is the set
of all solutions to the homogeneous equation AT y = 0.
RV
Theorem
The left null space of an m × n matrix A is a subspace of Rm . Equiv-
alently, the set of all solutions to a system AT y = 0 n homogeneous
linear equations in m unknowns is a subspace of Rm .

Examples:
   
1 2 −3
1. Let A = and v = .
3 6 1
Determine if v belongs to the left null space of A.
Solution:
      
T −3 −3 −3 + 3 0
A v= = =
1 1 −6 + 6 0
T
∴ v is in NulA .
 
1 0 0
2. Find a spanning set for the left null space of the matrix .
0 0 0
Solution:
Department of Mathematics

 
1 0
AT = 0 0
0 0 

1 0    
y 0
A y = 0 0 1 =
T 
y2 0
0 0
=⇒ y1 = 0, y2 is a free
  variable.
y1 0 0
= = y1 = y2 u.
y2 y2 1
∴ every linear combination of u is an element of NulAT . Thus {u} is a
spanning set of NulAT .

CE
Example:
1. Find the dimension
 and basisfor the four fundamental subspaces of
1 3 3 2
RV
the matrix A = 2  6 9 7.
−1 −3 3 4
Solution:
R2 − 2R1 , R
 3 + R1 =⇒
1 3 3 2
0 0 3 3
0 0 6 6
 3 − 2R2 =⇒
R 
1 3 3 2
0 0 3 3
0 0 0 0
 2 ÷ 3 =⇒ 
R
1 3 3 2
0 0 1 1
0 0 0 0
Ax = 0 =⇒ x1 + 3x2 + 3x3 + 2x4 = 0, 3x3 + 3x4 = 0
=⇒ x3 = −x4 , x1 = 3x2 + x4
x2 , x4 are free variables.
Department of Mathematics

       
x1 −3x2 + x4 −3 1
 x2   x2  = x2  1  + x4  0 
    
 =
x3   −x4  0 −1
x4 x4 0 1
∴ {(−3, 1, 0, 0), (1, 0, −1, 1)} forms a basis of NulA and dim(NulA) = 2.
{(1, 2, −1), (3, 9, 3)} forms a basis of ColA and dim(ColA) = 2.
{(1, 3, 
3, 2), (2, 6, 
9, 7)} forms a basis of RowA and dim(Row)A = 2.
1 2 −1
3 6 −3
AT =  3 9 3 

2 7 4
R


CE
 2 − 3R1 , R
1 2 −1
0 0 0 
0 3 6 
0 0 0
 2 ←→ R3=⇒
R
3 − 3R1 , R4 − 2R1 =⇒

1 2 −1
0 3 6 
RV
 
0 0 0 
0 0 0
 
1 2 −1
0 1 2 
R2 ÷ 3 =⇒  0 0 0 

0 0 0
T
A
 y = 0 =⇒ 
1 2 −1    
0 1 2  y1 0
  y2  = 0
0 0 0 
y3 0
0 0 0
=⇒ y1 + 2y2 − y3 = 0, y2 + 2y3 = 0 =⇒ y2 = −2y3 , y1 = −2y2 + y3 =
4y3 
 + y3 = 5y3   
y1 5y3 5
y2  = −2y3  = y3 −2
y3 y3 1
∴ {(5, −2, 1)} forms a basis of Nul(AT ) and dimNul(AT ) = 1.
Department of Mathematics

The dimension of the column space of A, is called the rank of A.


Since RowA is the same as ColAT , the dimension of the row space of A
is the rank of AT .
The dimension of the null space is called the nullity of A.

Rank-Nullity Theorem:
For an m × n matrix A, rank A + nullity A = n.
Examples:
1. If A is a 7 × 9 matrix with two dimensional null space, what is the
rank of A?

CE
Solution:
rank + nullity = 9 =⇒ rank + 2 = 9 =⇒ rank = 7.
2. Could a 6 × 9 matrix have a two-dimensional null space?
Solution:
rank + nullity = 9 =⇒ rank + 2 = 9 =⇒ rank = 7, which contradicts
that basis of ColA is a subspace of R6
∴ a 6 × 9 matrix cannot have a two-dimensional null space.
RV
Exercise:    
1 3 −5 −3
1. Determine if u =  3  is in NulA, where A =  6 −2 0 
−4 −8 4 1
Answer: Yes, u is inNulA.   
5 5 21 19
2. Determine if v = −3 is in NulA, where A = 13 23 2 
2 8 14 1
Answer: Yes, v is in NulA.
3. Find an explicit description of NulA, by listing vectors that span the
null space.  
  1 5 −4 −3 1
1 3 5 0
(i) A = , (ii) A = 0 1 −2 1 0
0 1 4 −2
0 0 0 0 0
Answer: (i) NulA = {c1 u1 + c2 u2 }, where u1 = (7, −4, 1, 0), u2 =
(−6, 2, 0, 1).
(ii) NulA = {c1 u1 + c2 u2 + c3 u3 }, where u1 = (−6, 2, 1, 0, 0), u2 =
Department of Mathematics

(8, −1, 0, 1, 0), u3 = (−1, 0, 0, 0, 1).


4. Find A  such thatthe given set is ColA.  
2s + 3t b−c
r + s − 2t 2b + c + d
(i) A = {  4r + s  : r, s, t real}, (ii) A = { 5c − 4d  : b, c, d real}.
  

3r − s − t d
           
0 2 3 1 −1 0
1 1 −2 2 1 1
Answer: (i) r  4 + s  1  + t  0 , (ii) b 0 + c  5  + d −4
          

3 −1 −1 0 0 1
k
5. Find (a) k such that NulA is a subspace of R , (b) k such that ColA

(i) 
CE
is asubspaceof Rk
2 −6
−1 3 
−4 12 
3 −9
, (ii) 
7 −2 0
−2 0 −5
 0 −5 7 
−5 7 −2

, (iii)

4 5 −2
1 1 0 1 0
6 0

Answer: (i) (a) k = 2, (b) k = 4, (ii) (a) k = 3, (b) k = 4, (iii) (a)



, (iv)

1 −3 9 0

k = 5, (b) k = 2, (iv) (a) k  = 5, (b) k = 1.


2 −6
RV
 
−1 3  1 3 5 0
6. With (i) A =    , (ii) A = , find a non-zero
−4 12  0 1 4 −2
3 −9
vector in NulA and a non-zero vector in ColA.
Answer: (i) (3, 1) ∈ NulA, (2, −1, −4, 3) ∈ ColA.
(ii) c1 (7, −4,1, 0) + c2 (−6, 2, 0, 1) ∈ NulA, c1 (1, 0) + c2 (3, 1) ∈ ColA.
−6 12 2
7. Let A = and u = . Determine if u is in ColA. Is u in
−3 6 1
NulA?
Solution: u∈ ColA, u ∈ NulA.   
−8 −2 −9 2
8. Let A =  6 4 8  and v =  1 . Determine if v is in ColA.
4 0 4 −2
Is v in NulA?
Answer: v ∈ ColA, v ∈ NulA.
9. Find a basis for the set of vectors in R3 in the plane x + 2y + z = 0.
Department of Mathematics

Answer: Basis = {(−2, 1, 0), (−1, 0, 1)}.


10. Find a basis for the set of vectors in R2 on the line y = 5x.
Answer: Basis = {(1, 5)}.
11. Find the basis for NulA and ColA. Assume that A is row equivalent
to B.    
−2 4 −2 −4 1 0 6 5
(i) A =  2 −6 −3 1 , B = 0 2 5 3
−3 8 2 −3 0 0 0 0
   
1 2 −5 11 −3 1 2 0 4 5
2 4 −5 15 2 
, B = 0 0 5 −7 8 
 
(ii) A = 1 2 0 4 5  0 0 0 0 −9

CE 3 6 −5 19 −2 0 0 0 0

basis for ColA = {(−2, 2, −3), (4, −6, 8)}.


0
Answer: (i) Basis for NulA = {(−6, −5/2, 1, 0), (−5, −3/2, 0, 1)} and

(ii) Basis for NulA = {(−2, 1, 0, 0, 0), (−4, 0, 7/5, 1, 0)} and
basis for ColA = {(1, 2, 1, 3), (−5, −5, 0, −5), (−3, 2, 5, −2)}.
12. Find
 the basis for therow space and column space of the matrix
1 5 −2 3 5
 0 0 1 −1 0 
RV
U =  0 0 0
.
0 1 
0 0 0 0 0
Answer: Basis for RowA = {(1, 5, −2, 3, 5), (0, 0, 1, −1, 0), (0, 0, 0, 0, 1)},
basis for ColA = {(1, 0, 0, 0), (−2, 1, 0, 0), (5, 0, 1, 0)}.
13. Determine a basis for the null space and left null space of the ma-
trix,  
2 −4 1 2 −2 −3
A =  −1 2 0 0 1 −1 .
10 −4 −2 4 −2 4
Answer: Basis of NulA = {(−1, −1/2, −2, 1, 0, 0), (0, −1/2, 0, 0, 1, 0), (1, 1, 5,
basis of AT = {}.
14. Find the bases for the null spaces of the matrices 
1 0 −3 2 1 0 −5 1 4
(i) A = 0 1 −5 4 , (ii) A = −2 1 6 −2 −2.
3 −2 1 −2 0 2 −8 1 9
Answer: (i) Basis of NulA = {(3, 5, 1, 0), (−2, −4, 0, 1)},
Department of Mathematics

(ii) Basis of NulA = {(5, 4, 1, 0, 0), (−7, −6, 0, 3, 1)} 


−1 2 −1
 4 −4 −4
15. Find a bases for the four fundamental sub-spaces of A =   2 0 −6
−3 1 7
Answer: Basis of NulA = {(3, 2, 1, 0, 0), (0, −8, 0, 2, 1)},
Basis of RowA = {(−1, 2, −1, 5, 6), (4, −4, −4, −12, −8), (−3, 1, 7, −2, 12)},
Basis of ColA = {(−1, 4, 2, −3), (2, −4, 0, 1), (5, −12, −2, −2)},
Basis of NulAT = {(−2, −1, 1, 0)}.
Transformation:
Consider the matrix equation Ax = b, where A is an m × n matrix, x

CE
ia an n × 1 matrix and b is an m × 1 matrix.
In other words x is a vector in Rn and b is a vector in Rm .
Solving the equation Ax = b amounts to finding all vectors x in Rn that
are transformed into the vector b in Rm under the action of multiplica-
tion by A.
The correspondence from x to Ax is a function from one set of vectors
to another.
This concept generalizes the common notation of a function as a rule
RV
that transforms one real number into another.
A transformation(or function or mapping) T from Rn to Rm is a rule
that assigns to each vector x in Rn a vector T (x) in Rm . The set Rn
is called the domain of T and Rm is called the co domain of T . The
notation T : Rn −→ Rm indicates that the domain of T is Rn and the
co domain is Rm .
For x in Rn , the vector T (x) in Rm is called the image of x. The set of
all images T (x) is called the range of T .

Matrix Transformations:
For each x in Rn , T (x) is computed as Ax, where A is an m × n matrix.
It is also denoted by the matrix transformation x −→ Ax. Observe that
the domain of T is Rn when A has n columns and codomain of T is Rm
when each column of A has m entries.
The range of T is the set of all linear combinations of the columns of A,
because each image T (x) is of the form Ax.
Department of Mathematics

Example:      
1 −3   3 3
2
1. Let A = 3  5 ,u =
 , b = 2 , c = 2 and define a
  
−1
−1 7 −5 5
2 3
transformation
  T : R −→ R by T (x) = Ax, so that T (x) = Ax =
1 −3   x1 − 3x2
x
3 5  1 =  3x1 + 5x2 .
x2
−1 7 −x1 + 7x2
(a) Find T (u), the image of u under the transformation T .
(b) Find an x in R2 whose image under T is b.
(c) Is there more than one x whose image under T is b?

CE
(d) Determine if c is in the range of the transformation T .
Solution:

(a) T (u) = Au =  3

−1 7

1 −3  
5
−1
2


 
5
= −1.
9 
1 −3  
x
 
3
(b) T (x) = b =⇒ Ax = b =⇒  3 5  1 =  2 , which can
x2
 −1 7  −5
RV
1 −3 : 3
be written in the matrix form as  3 5 :2 
−1 7 : −5
Reducing to echelon form as below:
R2 = R2 − 3R1 ; R3= R3 + R1 =⇒

1 −3 : 3
0 14 : −7 
0 4 : −2
R
 3 = 14R 3 − 4R 2 =⇒
1 −3 : 3
0 14 : −7 
0 0 : 0
=⇒  x1 − 3x2 = 3; 14x2 = −7 =⇒ x2 = −1/2, x1 = 3/2. Hence
3/2
x= .
−1/2
(c) From (b) we can see that, the vector x is unique.
Department of Mathematics

   
1 −3   3
x1
(d) Let T (x) = c =⇒ Ax = c =⇒  3 5  = 2 =⇒

x2
  −1 7 5
1 −3 : 3
3 5 :2 
−1 7 : 5
R
 2 = R 2 − 3R 1; R3 = R3 + R1 =⇒
1 −3 : 3
0 14 : −7 
0 4 : 8
R 3 = 14R 3 − 4R 2 =⇒

1 −3

0 0
CE
0 14 : −7
:

inconsistent.
3

140

Third row shows that 0 = 140(which is invalid). Hence the system is

Hence c is not in the range of the transformation.


 
2 0
RV
2. Let A = , define T : R2 −→ R2 by T (x) = Ax. Find the
0 2    
1 a
images under T of u = and v = .
−3 b
Solution:    
2 0 1 2
T (u) = =
 0 2 −3
    −6
2 0 a 2a
T (v) = = .
0 2 b 2b
 
1 3
3. Let A = . Then the transformation T (x) = Ax, trans-
0 1
forms
 the
 square
 with
 vertices
   (0,0), (2,
 0), (2,
 2), (0, 2)
 to   
1 3 0 0 1 3 2 2 1 3 2 8 1 3 0
= , = , = , =
0 1 0 0 0 1 0 0 0 1 2 2 0 1 2
6
.
2
Sketching the above transformation, we see
Department of Mathematics

that the square has been transformed to a parallelogram(In other words,


the square has been stretched to a parallelogram, keeping the base in-
tact).    
5 −2
4. Let u = ,v= .
2 4
−1 0

Au = CE
(i) With A =

−1 0
0 −1 2
0 −1  
5
=
,
−5
−2
, Av =

−1 0
0 −1
   
−2
4
=
2
−4
RV

Reflected about
 origin.

0 1
(ii) With A = ,
 1
   0     
0 1 5 2 0 1 −2 4
Au = = , Av = =
1 0 2 5 1 0 4 −2
Department of Mathematics

Reflected about the line


 y = x.
2 0

Au =
CE
(iii) With A =
   0 2 
2 0 5
0 2 2
=
,
10
4
, Av =


0 2
   
2 0 −2
4
=
−4
8
RV
Stretches the vector.
 
0 0
(iv) With A = ,
   0 2      
0 0 5 0 0 0 −2 0
Au = = , Av = =
0 2 2 4 0 2 4 8
Department of Mathematics

CE
Rotates about origin and projects onto y axis.
If A is an m × n matrix, then the transformation x −→ Ax has the
properties A(u + v) = Au + Av and A(cu) = cA(u) for all u, v in Rn
and all scalars c.
These properties, written in function notation, identify the most impor-
tant class of transformations in Linear Algebra.
RV
Linear transformation:
Let U and V be two vector spaces over the same field F . The mapping
T : U −→ V is said to be a linear transformation, if
(i) T (α + β) = T (α) + T (β) ∀ α, β ∈ U
(ii) T (c.α) = c.T (α) ∀ c ∈ F, α ∈ U.
A Linear transformation T : U −→ V is also called a linear map on U .
Note:
(i) Every matrix transformation is a linear transformation.
(ii) Linear transformations preserve the operations of vector addition
and scalar multiplication.
Theorem:
A mapping T : U −→ V from the vector space U (F ) into V (F ) is a
linear transformation iff
T (c1 α + c2 β) = c1 T (α) + c2 T (β) ∀ c1 , c2 ∈ F and α, β ∈ U .

Stretch, Rotation, Reflection, Projection


Suppose x is an n-dimensional vector. When A multiplies x, it trans-
Department of Mathematics

forms that vector into a new vector Ax. This happens at every point
x of the n-dimensional space Rn . The whole space is transformed or
’mapped into itself” by the matrix A.

Stretch:  
c 0
A multiple of the identity matrix A = cI = stretches every vector
0 c
by the same factor c. The whole space expands or contracts(or goes
through the origin andout the
  opposite
  side when c is negative).
c 0 x cx
α = (x, y), then Aα = = = (cx, cy)
0 c y cy

CE
(x,y)
(cx,cy)
RV
Rotation:
A rotation
 matrix
 turns the whole space around the origin.
0 −1
A = turns all vectors through 90◦ transforming every point
1 0
(x, y) to (−y, x)

(x,y)
(-y,x)

Reflection:
A reflection matrix transforms
 every
 vector into its image on the oppo-
0 −1
site side of a mirror. A = gives the reflection through y = x.
1 0
Department of Mathematics

(y,x)

(x,y)

 
1 0
A= gives the reflection through x−axis.
0 −1

CE (x,y)

(x,-y)
RV
 
−1 0
A= gives the reflection through y−axis.
0 1
 
0 −1
A= gives the reflection through y = −x.
−1 0
 
−1 0
A= gives the reflection through the origin.
0 −1

Projection:
A projection matrix takes the whole space onto a lower-dimensional
subspace(not
  invertible).
1 0
A = transforms each vector (x, y) in the plane to the nearest
0 0
point (x, 0) on the horizontal axis.
Department of Mathematics

(x,y)

(x,0)
 
0 0
A= transforms each vector (x, y) in the plane to the nearest
0 1
point (0, y) on the vertical axis.

CE
(0,y) (x,y)
RV
Rotation
  Q, Projection P, Reflection H
0 −1
Q= Rotation about 90◦
 1 0
1 0
P = Projection onto the x−axis.
 0 0 
0 1
H= Reflection about 45◦
1 0
Rotation:
The family of rotations can be represented in matrix form as
   
cos θ − sin θ c −s
Qθ = =
sin θ cos θ s c

   
cos θ sin θ c s
Q−θ = =
− sin θ cos θ −s c
Department of Mathematics

Rotation in backwards through θ, Q−θ = [Qθ ]T


    
c −s c s 1 0
Qθ .Q−θ = = =I
s c −s c 0 1
    
c −s c −s cos 2θ − sin 2θ
Qθ .Qθ = = = Q2θ
s c s c sin 2θ cos 2θ
   
−1 1 c s c s
Qθ = 2 = = QTθ
c + s2 −s c −s c

CE
Qθ .Qφ =

exists.


sin θ cos θ

Projections:

cos θ − sin θ cos φ − sin φ
sin φ cos φ
 
=
cos(θ + φ) − sin(θ + φ)
sin(θ + φ) cos(θ + φ)
Qθ .Qφ = Qθ+φ . If φ = −θ then Qθ .Q−θ = Qθ−θ = Q0 = I. Inverse

RV
The family of rotations can be represented in matrix form as
   2 
cos2 θ sin θ cos θ c cs
P = =
sin θ cos θ sin2 θ cs s2

|P | = c2 s2 − c2 s2 = 0 =⇒ inverse doesn’t exist.


 2    2 
2 c cs c2 cs c cs
P = = =P
cs s2 cs s2 cs s2

Projection twice on the θ−line is the same as projecting once on θ−line.


Points on the y−axis is projected to (0, 0)
Points on θ− line is projected onto itself.
Reflection:
The family of reflection can be represented in matrix form as
 
cos 2θ sin 2θ
H=
sin 2θ − cos 2θ
Department of Mathematics

H 2 = (2P − I)2 = 4P 2 − 4P + I = 4P − 4P + I(∵ P 2 = P ) =⇒ H 2 = I


Two reflections bring back the original. Reflection of Reflection = Orig-
inal.

H −1 = H, H = 2P − I =⇒ 2P = H + I

.
Problems:
1. If T is a mapping from V3 (R) into V3 (R) defined by T (x1 , x2 , x3 ) =
(0, x2 , x3 ). Show that T is a linear transformation.
Solution:

CE
Let α = (x1 , x2 , x3 ), β = (y1 , y2 , y3 ) ∈ V3 (R)
Consider, T (α + β) = T ((x1 , x2 , x3 ) + (y1 , y2 , y3 ))
= T (x1 + y1 , x2 + y2 , x3 + y3 )
= T (0, x2 + y2 , x3 + y3 )
= (0, x2 , x3 ) + (0, y2 , y3 )
= T (α) + T (β)
Consider, T (cα) = T (c(x1 , x2 , x3 ))
RV
= T (cx1 , cx2 , cx3 )
= (0, cx2 , cx3 )
= c(0, x2 , x3 )
= cT (x1 , x2 , x3 )
= cT (α)
∴ T is a linear transformation.

2. Find the linear transformation T : R2 −→ R3 such that T (1, 1) =


(0, 1, 2), T (−1, 1) = (2, 1, 0).
Solution:
{(1, 1), (−1, 1)} forms a basis of R2 .
Let α = (x, y) ∈ R2
(x, y) = c1 (1, 1) + c2 (−1, 1) =⇒ (x, y) = (c1 − c2 , c1 + c2 )
x+y y−x
=⇒ x = c1 − c2 , y = c1 + c2 =⇒ c1 = , c2 =
2 2
x+y y−x
∴ (x, y) = (1, 1) + (−1, 1)
2 2
∴ the required transformation is
Department of Mathematics

x+y y−x
T (x, y) = (0, 1, 2) + (2, 1, 0)
2 2
T (x, y) = (y − x, y, x + y)

3. Is there a linear map T : R2 −→ R2 for which T (2, 2) =


(4, −6) and T (5, 5) = (2, −3)?
Solution:
The vectors (2, 2) and (5, 5) ∈ R2 are linearly dependent as (5, 5) =
5
(2, 2)
2
5 5 5
If T is linear, T (5, 5) = T ( (2, 2)) = T (2, 2) = (4, −6) = (10, −15)

CE 2
But T (5, 5) = (2, −3) 6= (10, −15).
2

∴ a linear map with the given data doesn’t exist.


2

4. Let M (R) be the vector space of all 2 × 2 matrices over R


and B be a fixed non-zero element of M (R). Show that the mapping
T : M (R) −→ M (R) defined by T (A) = AB − BA, ∀A ∈ M (R) is a
linear map.
RV
Solution:
Let A and C ∈ M (R) be arbitrary.
Consider, T (A + C) = (A + C)B − B(A + C) = AB + CB − BA − BC
= AB − BA + CB − BC = T (A) + T (C)
Let c ∈ R be any scalar.
Consider, T (c.A) = (c.A)B − B(c.A) = c.(AB − BA) = c.T (A)
∴ T is a linear transformation.

5. If T : R2 −→ R2 is a linear transformation such that T (1, 0) =


(1, 1) & T (0, 1) = (−1, 2), show that T maps the square with vertices
(0, 0), (1, 0), (1, 1), (0, 1) into a parallelogram.
Solution:
{(1, 0), (0, 1)} forms a basis of R2
(x, y) = x(1, 0) + y(0, 1)
∴ T (x, y) = xT (1, 0) + yT (0, 1) = x(1, 1) + y(−1, 2) = (x − y, x + 2y)
Now, T (0, 0) = (0, 0) = A, T (1, 0) = (1, 1) = B, T (1, 1) = (0, 3) = C,
T (0, 1) = (−1, 2) = D.
Department of Mathematics

To show that A, B, C, D are vertices of a parallelogram, we shall show


that the diagonals AC and BD bisect each other.
3 3
Midpoint of AC = (0, ), Midpoint of BD = (0, )
2 2
Diagonals bisect each other. Hence ABCD is a parallelogram.

6. If T : V1 (R) −→ V3 (R) is defined by T (x) = (x, x2 , x3 ), verify


whether T is linear or not.
Solution:
Let x, y ∈ V1 (R)
Consider, T (x + y) = (x + y, (x + y)2 , (x + y)3 )

CE
T (x) + T (y) = (x, x2 , x3 ) + (y, y 2 , y 3 )) = (x + y, x2 + y 2 , x3 + y 3 )
We can see that, T (x + y) 6= T (x) + T (y)
∴ T is not a linear transformation.

7. Find the linear transformation f : R2 −→ R2 such that f (1, 0) =


(1, 1) and f (0, 1) = (−1, 2).
Soln: Let (x, y) ∈ R2 . Then (x, y) = x(1, 0) + y(0, 1)
Define f : R2 −→ R2 by f (x, y) = xf (1, 0)+yf (0, 1) = x(1, 1)+y(−1, 2)
RV
Hence f (x, y) = (x − y, x + 2y) ∀ (x, y) ∈ R2

8. Find the linear transformation f : R2 −→ R2 such that f (1, 1) =


(0, 1) and f (−1, 1) = (3, 2).
Solution:
(1, 1), (−1, 1) of R2 forms a basis of R2
Let α = (x, y) ∈ R2 be arbitrary.
Let(x, y) = c1 (1, 1) + c2 (−1, 1) = (c1 − c2 , c1 + c2 )
x+y y−x
=⇒ c1 − c2 = x, c1 + c2 = y =⇒ c1 = , c2 =
2 2
x+y y−x
∴ (x, y) = (1, 1) + (−1, 1)
2 2
x+y y−x
Hence the required transformation is f (x, y) = f (1, 1)+ f (−1, 1)
2 2
x+y y−x
= (0, 1) + (3, 2)
2 2
3y−3x 3y−x
∴ f (x, y) = ( 2 , 2 ).
Department of Mathematics

9. f : V3 (R) −→ V2 (R) is defined by f (x, y, z) = (x + y, y + z), show


that f is a linear transformation.
Solution:
Let α = (x1 , y1 , z1 ) and β = (x2 , y2 , z2 ) be any two elements of V3 (R).
Consider, f (α + β) = f (x1 + x2 , y1 + y2 , z1 + z2 )
= (x1 + x2 + y1 + y2 , y1 + y2 + z1 + z2 ) = ((x1 + y1 ) + (x2 + y2 ), (y1 +
z1 ) + (y2 + z2 ))
= (x1 + y1 , y1 + z1 ) + (x2 + y2 , y2 + z2 ) = f (x1 , y1 , z1 ) + f (x2 , y2 , z2 )
∴ f (α + β) = f (α) + f (β).
Consider, f (c.α) = f (cx1 , cy2 , cz2 )
= (cx1 + cy1 , cy1 + cz2 ) = c(x1 + y1 , y1 + z1 ) = cf (x1 , y1 , z1 )

CE
∴ f (c.α) = cf (α)
Hence f is a linear transformation.

10. If T is a mapping from V2 (R) into V2 (R) defined by T (x, y) =


(x cos θ −y sin θ, x sin θ +y cos θ), show that T is a linear transformation.
Solution:
Let α = (x1 , x2 ), β = (y1 , y2 ) ∈ V2 (R)
RV
Consider, T (α + β) = T (x1 + y1 , x2 + y2 )
= ((x1 + y2 ) cos θ − (x2 + y2 ) sin θ, (x1 + y2 ) sin θ − (x2 + y2 ) cos θ)
= ((x1 cos θ−x2 sin θ)+(y1 cos θ−y2 sin θ), (x1 sin θ+x2 cos θ)+(y1 sin θ+
y2 cos θ))
= (x1 cos θ − x2 sin θ, x1 sin θ + x2 cos θ) + (y1 cos θ − y2 sin θ, y1 sin θ +
y2 cos θ)
= T (x1 , x2 ) + T (y1 , y2 ) ∴ T (α + β) = T (α) + T (β)
Consider, T (c.α) = T (cx1 , cx2 )
= (cx1 cos θ−cx2 sin θ, cx1 sin θ+cx2 cos θ) = c(x1 cos θ−x2 sin θ, x1 sin θ+
x2 cos θ)
= cT (x1 , x2 )
∴ T (c.α) = cT (α)
Hence T is a linear transformation.

Range and Kernel of a Linear transformation


Definition:
Let T : V −→ W be a linear transformation. The range of T is the set
Department of Mathematics

R(T ) = {T (α)/α ∈ V }
Definition:
Let T : V −→ W be a linear transformation. The kernel(or null space)
of T is the set N (T ) = {α ∈ V /T (α) = 0}, where 0 is the zero vector of
W.

Note:
(i) For the identity map I : V −→ V the range is the entire space V
and the kernel is the zero subspace.
(ii) For the zero linear map T : V −→ W defined by T (α) = 0∀α ∈ V ,
the range R(T ) = {0} = zero space of V and the null space N (T ) = V .

CE
Theorem
Let T : V −→ W be a linear transformation.
Then (a) R(T ) is a subspace of W .
(b) N (T ) is a subspace of V .
(c) T is one-one iff N (T ) = {0}, where 0 is the zero vector of W .
RV
Theorem
Let T : V −→ W be a linear transformation. The dimension of the
range space R(T ) is called the rank of the linear transformation T and
is denoted by r(T ). The dimension of the null space N (T ) is called the
nullity of the linear transformation T and is denoted by n(T ).

Theorem
Let T : V −→ W be a linear transformation. If the vectors α1 , α2 , ..., αn
generate V , then the vectors T (α1 ), T (α2 ), ..., T (αn ) generates R(T ).
Theorem(Rank-Nullity theorem)
Let T : V −→ W be a linear transformation and V be a finite dimen-
sional vector space. Then r(T )+n(T ) = d[V ](d[R(T )]+d[N (T )] = d[V ])

Problems:
1. Let T : V −→ W be a linear transformation defined by T (x, y, z) =
(x + y, x − y, 2x + z). Find the range, null space, rank, nullity and hence
verify the rank-nullity theorem.
Department of Mathematics

Solution:

T (e1 ) = T (1, 0, 0) = (1, 1, 2) = α1


T (e2 ) = T (0, 1, 0) = (1, −1, 0) = α2
T (e3 ) = T (0, 0, 1) = (0, 0, 1) = α3
{α1 , α2 , α3 } generates
 R(T)
1 1 2
Consider A = 1 −1 0 |A| = −2 6= 0
0 0 1

CE
∴ {α1 , α2 , α3 } is linearly independent. Thus it is a basis of R(T ).
d[R(T )] = 3.
Let α ∈ R(T )
Then
α = c1 (α1 ) + c2 (α2 ) + c3 (α3 )
= c1 (1, 1, 2) + c2 (1, −1, 0) + c3 (0, 0, 1)
= (c1 + c2 , c1 − c2 , 2c1 + c3 )
RV
∴ R(T ) = {(c1 + c2 , c1 − c2 , 2c1 + c3 )/c1 , c2 , c3 ∈ R}
Suppose
T (x, y, z) = (0, 0, 0)
=⇒ (x + y, x − y, 2x + z) = (0, 0, 0)
=⇒ x + y = 0, x − y = 0, 2x + z = 0
=⇒ x = 0, y = 0, z = 0
∴ N (T ) = {(0, 0, 0)}
d[N (T )] = 0
rank + nullity = 3 + 0 = 3 = d[V3 (R)].
Exercise:
1. Find the range space, null space, rank and nullity of T and verify
the rank-nullity theorem for, T : R3 −→ R3 , defined by T (x, y, z) =
(x + y, x − y, 2x + z).
Answer: Range space = {(1, 1, 2), (1, −1, 0), (0, 0, 1)}, rank = 3, Null
Department of Mathematics

space = {(0, 0, 0)}, nullity = 0. rank + nullity = 3 + 0 = 3 = dimension


of domain.
2. Find the range space, null space, rank and nullity of T and ver-
ify the rank-nullity theorem for, T : V3 (R) −→ V4 (R), defined by
T (e1 ) = (0, 1, 0, 2), T (e2 ) = (0, 1, 1, 0), T (e3 ) = (0, 1, −1, 4).
Answer: Range space = {(0, 1, 0, 2), (0, 1, 1, 0)}, rank = 2, Null space
= {(−2, 1, 1)}, nullity = 1. rank + nullity = 2 + 1 = 3 = dimension of
domain.

CE
RV

You might also like