Topic 054

Linear Operators
Operator: In the case of vector spaces and in particular normed
spaces, a mapping is called an operator.
Linear Operator: A linear operator T is an operator such that:
(i) The domain D(T) of T is a vector space and range R(T)
lies in a vector space over the same field.
(ii) For all x , y ∈ D(T) and scalars 𝛼
T(x + y) = Tx + Ty
T(𝛼x) = 𝛼 Tx
We can also combine these two conditions as:
T(𝛼x + 𝛽y) = 𝛼Tx + 𝛽Ty
Some notations are:
D(T) denotes domain of T.
R(T) denotes range of T.
N(T) denotes the null space of T.
Null Space: The set of all points x of domain D(T) such that:
T(x) = 0
In other words, we say that null space is same as Kernel (T).
Let D(T)⊂ X and R(T)⊂ Y where X and Y are vector spaces
(Real or Complex). Then T is an operator from D(T) onto R(T).
 T: D(T) → R(T)
Or D(T) into Y i.e T: D(T) → Y as R(T)⊂ Y
If D(T) is the whole space X, then we write:
T: X→ Y
If 𝛼 = 0 then T(0) = 0
So, T(𝛼x + 𝛽y) = 𝛼Tx + 𝛽Ty is same as the condition of
homomorphism.
Hence T is a homomorphism when it is a linear operator.

Topic 055
Linear Operators (Examples)
Identity Operator: Ix : X → X
Ix (x) = x ∀ 𝑥∈X
Zero Operator: 0 : X → Y
0(x) = 0 ∀ 𝑥∈X
Differentiation: Let X be a vector space of all polynomials on
[a,b].
Tx(t) = x’(t) ∀ x(t) ∈ X
T maps X onto itself.
Integration: Linear operator T for C[a,b] into itself can be
defined as:
𝑡
Tx(𝒯) = ∫𝑎 𝑥(𝒯)𝑑𝒯

Multiplication by t: Defined on C[a,b],

Tx(t) = t x(t)
This linear operator plays important role in quantum theory in
physics.
Elementary vector algebra: We have different maps in it.
For example:
T1 : R3→ R3 (cross product of two vectors is again a vector)
T1(x) = a × x
T2 : R3→ R (dot product of two vectors is a scalar)
T2 (x) = a1.x1 + a2.x2+a3+x3 ∈ R
Matrices: A = (𝛼jk) (r rows, n columns)
T: Rn→Rr
Topic 056
Theorem (Range and null space)
Theorem: Let T be a linear operator. Then,
➢ The range R(T) is a vector space.
➢ If dim D(T) = n < ∞, then dim R(T) ≤ 𝑛.
➢ The null space N(T) is a vector space.
These are three different conditions, which we have to prove one
by one.
Proof: We have to prove that the range R(T) is a vector space.
i.e if y1 , y2 ∈ R(T) then 𝛼y1 + 𝛽y2 ∈ R(T) where 𝛼 and 𝛽 are
scalars.
Since, y1 , y2 ∈ R(T) and T : D(T) → Y
⟹ x1 , x2 ∈ D(T)
⟹ y1 = Tx1 , y2 = Tx2
Since D(T) is a vector space then 𝛼x1 + 𝛽x2 ∈ D(T).
T is linear then T(𝛼x1 + 𝛽x2) = 𝛼Tx1 + 𝛽Tx2 = 𝛼y1 + 𝛽y2 ∈ R(T)
Hence, R(T) is a vector space.
Now,
If dim D(T) = n < ∞, we have to prove that dim R(T) ≤ 𝑛.
Let y1 , y2 ,….,yn , y+1 ∈ R(T)
⟹ ∃ x1 , x2, …., xn , xn+1 ∈ D(T) such that
y1 = Tx1 , y2 = Tx2 ,…., yn+1 = Txn+1
dim D(T) = n < ∞
⟹ { x1 , x2, …., xn , xn+1} must be linearly dependent.
𝛼1x1 + 𝛼2x2 + ……+ 𝛼n+1xn+1 = 0
𝑓or scalars, 𝛼1 , 𝛼2 , ….., 𝛼n+1 not all zero.
T is linear. ⟹ T(0) = 0
T(𝛼1x1 + 𝛼2x2 + ……+ 𝛼n+1xn+1 ) = T(0) = 0
𝛼1 Tx1 + 𝛼2 Tx2 + ……+ 𝛼n+1 Txn+1 = 0
𝛼1y1 + 𝛼2y2 + ……+ 𝛼n+1yn+1 = 0
⟹ dim R(T) ≤ 𝑛.
If we apply linear operator on a linearly dependent set, we will
get linearly dependent set.
Linear operators preserve linear dependance.
Now, we have to prove that null space is a vector space.
Let x1 , x2 ∈ N(T) ⟹ Tx1 = Tx2 = 0
T(𝛼x1 + 𝛽x2) = 𝛼Tx1 + 𝛽Tx2
= 𝛼(0) + 𝛽(0) ⟹ = 0
Hence, 𝛼x1 + 𝛽x2 ∈ N(T) ⟹ N(T) is a vector space.
Topic 057
Inverse Operator
T: D(T) → Y is said to be injective or one to one if for every
x1 , x2 ∈ D(T) such that x1 ≠ x2 ⟹ Tx1 ≠ Tx2
⟺ Tx1 = Tx2 ⟹ x1 = x2
T: D(T) → R(T)
There exist a mapping T -1 : R(T) → D(T) ⟹ y0 → x0
T -1 is called the inverse of T.

T -1 T (x) = x ∀ 𝑥 ∈ 𝐷(𝑇)
T T -1 (y) = y ∀ 𝑦 ∈ 𝑅(𝑇)

Topic 058
Theorem (Inverse Operator)
Theorem: Let X , Y be vector spaces , both real or both
complex. Let T: D(T) → Y be a linear operator with domain
D(T) ⊂ X and range R(T) ⊂ Y. Then:
 (a) The inverse T -1 : R(T) → D(T) exists if and only if
Tx = 0 ⟹ x = 0.
(b) If T -1 exists, it is linear operator.
(c) If dimension of D(T) = n < ∞ and T -1 exists then
dimension of R(T) = dimension of D(T).
Proof:
(a). Let T(x) = 0 ⟹ x = 0 , we have to prove that T inverse
exists.
For this, we just need to prove that T is one-one.
Let Tx1 = Tx2 ⟹ Tx1 – Tx2 = 0 ⟹ T(x1 – x2) = 0
As T(x) = 0 ⟹ x = 0
⟹ x1 – x2 = 0 ⟹ x1 = x 2
Hence, T is one-one.
⟹ T -1 exists because T is one-one and onto.
Conversely, let T -1 exists.
We need to show that T(x) = 0 ⟹ x = 0.
Let x1 = 0
⟹ Tx1 = T0 = 0 ⟹ x1 = 0
Proved!
(b). We assume T -1 exists. We need to show that T -1 is a linear
operator.
The domain of T -1 is R(T) and R(T) is a vector space.
Let x1 , x2 ∈ D(T) ⟹ y1 = Tx1 and y2 = Tx2
⟹ x1 = T -1y1 and x2 = T -1y2
Since T is linear, then for any scalars 𝛼 𝑎𝑛𝑑 𝛽
𝛼 y1 + 𝛽 y2 = 𝛼 Tx1 + 𝛽 Tx2
=T ( x1 + 𝛽 x2 )
Now apply T -1:
T -1 (𝛼 y1 + 𝛽 y2 ) = T -1 T ( 𝛼 x1 + 𝛽 x2 )
T -1 (𝛼 y1 + 𝛽 y2 ) = 𝛼 x1 + 𝛽 x2
T -1 (𝛼 y1 + 𝛽 y2 ) = 𝛼 T -1y1 + 𝛽 T -1y2

Hence, T -1 is a linear operator.

(c). If dimension of D(T) = n < ∞ and T -1 exists then dimension

of R(T) = dimension of D(T).
As we have proved a theorem:
If dimension of D(T) = n < ∞ then dimension of R(T) ≤ n.
⟹ dimension of R(T) ≤ dimension of D(T) (1)
If we apply this reversely, we get:

T -1 : R(T) → D(T)
⟹ dimension of D(T) ≤ dimension of R(T) (2)
By (1) and (2)
dimension of R(T) = dimension of D(T) Proved!

Topic 059
Lemma (Inverse of product)
Lemma: Let T: X → Y and S : Y → Z be bijective linear
operators, where X, Y , Z are vector spaces. Then
The inverse (ST)-1 : Z → X of the product (the composite) exists
and (ST) -1 = T -1 S -1.
Proof by diagram:

Mathematical Proof:
Since S and T are both bijective then ST is also bijective.
⟹ (ST) -1 exists. i.e ST(ST) -1 = I
S -1 ST(ST) -1 = S -1I ⟹ T -1T(ST) -1 = T -1 S -1
⟹ (ST) -1 = T -1 S -1
Proved!
Topic 060
Bounded Linear operator
Definition: Let X and Y be normed spaces and T:D(T)→Y a
linear operator, where D(T) ⊂ X. The operator T is said to be
bounded if there is a real number c such that for all x ∈ D(T),
||Tx|| ≤ c ||x||.
||𝑇𝑥||
c≥ for x ∈ D(T)-{0}
||𝑥||

||𝑇𝑥 ||
smallest possible value of c must be supremum of ||𝑥 ||

If D(T) = {0} , then ||T|| = 0

||𝑇𝑥||
c = ||T|| = 𝑠𝑢𝑝𝑥∈𝐷(𝑇)−{0} ||𝑥||

||Tx|| ≤ ||T|| ||x|| we will use this expression for bounded linear
operator.

Topic 061
Lemma (Norm)
Lemma: Let T be a bounded linear operator as defined before.
An alternative formula for the norm of T is
||T|| = 𝑠𝑢𝑝𝑥∈𝐷(𝑇),||𝑥||=1 ||Tx||

The norm defined on T satisfies (N1) to (N4).

Proof: We have defined bounded linear operator as:
||𝑇𝑥||
||Tx|| ≤ c ||x|| and c = ||T|| = 𝑠𝑢𝑝𝑥∈𝐷(𝑇)−{0} ||𝑥||

We have to prove that value of c is equivalent to

c = ||T|| = 𝑠𝑢𝑝𝑥∈𝐷(𝑇),||𝑥||=1 ||Tx||
𝑥
Let ||x|| = a ; set y = 𝑎 , x ≠ 0
||𝑥||
||y|| = = 1. Now,
𝑎
||𝑇𝑥|| 1
||T|| = 𝑠𝑢𝑝𝑥∈𝐷(𝑇)−{0} = 𝑠𝑢𝑝𝑥∈𝐷(𝑇)−{0} ||𝑇 𝑎 𝑥||
𝑎

= 𝑠𝑢𝑝𝑦∈𝐷(𝑇),||𝑦||=1 ||Ty||

Variables can be x or y or anyother.

Hence Proved!
||𝑇𝑥||
||T|| = 𝑠𝑢𝑝𝑥∈𝐷(𝑇)−{0} ≅ 𝑠𝑢𝑝𝑥∈𝐷(𝑇),||𝑥||=1 ||Tx||
||𝑥||

N1: ||T|| > 0 is obvious.

N2: ||T|| = 0 ⟹ T = 0
||T|| = 0 ⟹ ||Tx|| = 0 ∀ 𝑥 ∈ D(T)
⟹T=0
N3: ||𝛼T|| = 𝑠𝑢𝑝𝑥∈𝐷(𝑇),||𝑥||=1 ||𝛼Tx||= 𝑠𝑢𝑝𝑥∈𝐷(𝑇),||𝑥||=1 |𝛼|||Tx||

= |𝛼|𝑠𝑢𝑝𝑥∈𝐷(𝑇),||𝑥||=1 ||Tx|| = |𝛼| ||T||

N4: ||T1 + T2|| ≤ ||T1|| + ||T2||
||T1 + T2||= 𝑠𝑢𝑝𝑥∈𝐷(𝑇),||𝑥||=1 ||(T1 + T2)x||

= 𝑠𝑢𝑝𝑥∈𝐷(𝑇),||𝑥||=1 ||(T1x + T2x||

≤ 𝑠𝑢𝑝𝑥∈𝐷(𝑇),||𝑥||=1 ||T1 x|| + ||T2 x||

= 𝑠𝑢𝑝𝑥∈𝐷(𝑇),||𝑥||=1 ||T1 x|| + 𝑠𝑢𝑝𝑥∈𝐷(𝑇),||𝑥||=1 ||T2 x||

||T1 + T2|| ≤ ||T1|| + ||T2||

Proved!

Topic 062
Examples (Bounded Linear Operators)
➢ Identity Operator
➢ Zero Operator
➢ Differentiation Operator
➢ Integral Operator
Identity Operator:
I:X→X such that Ix = x x ≠ {0} normed space
||𝑇𝑥||
||I|| = 𝑠𝑢𝑝𝑥∈𝐷(𝐼)−{0} ||𝑥||

||𝑥||
||I|| = 𝑠𝑢𝑝𝑥∈𝐷(𝐼)−{0} =1
||𝑥||

Zero Operator:
0:X→Y ⟹ 0x = 0
||𝑇𝑥||
||0|| = 𝑠𝑢𝑝𝑥∈𝐷(0)−{0} =0
||𝑥||

Zero Operator is bounded.

Differentiation Operator:
Differentiation operator is defined by normed spaces of all
polynomials on J=[0,1].
||x|| = max {x(t) , t ∈ J}
T x(t) = x’(t)
This operator T is bounded if ||T x(t)|| ≤ c ||x(t)||
We claim that T is not bounded.
Let xn (t) = t n n∈N
||xn (t)|| = max {x(t) , t ∈ [0,1]} = 1
T xn(t) = nt n-1
|| T xn(t)|| = max | nt n-1| t ∈ [0,1]
|| T xn(t)|| = n.1 = n
||𝑇𝑥𝑛 || 𝑛
=1 =c n∈N
||𝑥𝑛 ||

||𝑇𝑥𝑛 ||
There does not exist any c such that ≤ c.
||𝑥𝑛 ||

Hence, differentiation operator is not bounded.

Integral Operator:
T: C[0,1] → C[0,1] ⟹ y = T(x)
1
y(t) = ∫0 𝑘(𝑡, 𝒯 )x(𝒯) d(𝒯)

𝑘(𝑡, 𝒯) is fixed for every operator and called kernel of T. It is a

continuous and hence bounded function.
𝑘(𝑡, 𝒯) ≤ k0 k0 ∈ R 𝑡, 𝒯 ∈ J × J
|x(t)| ≤ 𝑚𝑎𝑥𝑡∈𝐽 |x(t)| = ||x||
1
||y|| = ||Tx||=𝑚𝑎𝑥𝑡∈𝐽 |∫0 𝑘(𝑡, 𝒯 )x(𝒯) d(𝒯)|
1
≤ 𝑚𝑎𝑥𝑡∈𝐽 ∫0 |𝑘(𝑡, 𝒯 )| |x(𝒯)| d(𝒯)

||Tx|| ≤ k0 ||x||
Hence, integral is a bounded linear operator.

Topic 063
Examples (Bounded Linear Operators)
➢ Matrix
As we know, matrices are linear operator. We have to just find
that these are bounded or not.
Let T: R n → R r
𝜂j = ∑𝑛𝑘=1 𝛼𝑗𝑘 𝜉𝑘
T is linear because of the properties of matrices.
Let x ∈ Rn and y ∈ Rr
1
||x|| = (∑𝑛𝑚=1 𝜉𝑚 2 )2
1
||y|| = (∑𝑟𝑗=1 𝜂𝑗 2 )2
1
||Tx|| = ||y|| = (∑𝑟𝑗=1 𝜂𝑗 2 )2

||𝑇𝑥 ||2 = ∑𝑟𝑗=1 𝜂𝑗 2

2
||𝑇𝑥 ||2 = ∑𝑟𝑗=1(∑𝑛𝑘=1 𝛼𝑗𝑘 𝜉𝑘 )

Using Cauchy Schwartz inequality:

1 1 2
≤ ∑𝑟𝑗=1 ((∑𝑛𝑘=1 𝛼𝑗𝑘 2 ) (∑𝑛𝑚=1 𝜉𝑚 2 ) )
2 2

= ||x||2 (∑𝑟𝑗=1 ∑𝑛𝑘=1 𝛼𝑗𝑘 2 )

||𝑇𝑥 ||2 ≤ c2 ||x||2

||Tx|| ≤ c ||x||
We get the value of c that is fixed and depend upon the value of
T. So, it is satisfying the definition of boundedness.
Hence,
T is a bounded linear operator.
Topic 064
Finite Dimension
Theorem: If a normed space X is finite dimensional, then every
linear operator on X is bounded.
Proof: X is finite dimensional. We have to prove that every
linear operator on X is bounded.
Let dim X = n ⟹ Let {e1 , e2 ,…, en} is a basis for X.
Let 𝑥 ∈ 𝑋 ⟹ x = ∑𝑛𝑗=1 𝜉𝑗 𝑒𝑗

Let T be any linear operator on X.

We have to find a number c such that:
||Tx|| ≤ 𝑐||𝑥 ||
||Tx|| = ||T(∑𝑛𝑗=1 𝜉𝑗 𝑒𝑗 )|| = ||∑𝑛𝑗=1 𝜉𝑗 T 𝑒𝑗 ||

≤ ∑𝑛𝑗=1 |𝜉𝑗 | || T 𝑒𝑗 ||

≤ 𝑚𝑎𝑥𝑘 || 𝑇 𝑒𝑘 || ∑𝑛𝑗=1 |𝜉𝑗 |

Using:
||∑𝑛𝑗=1 𝜉𝑗 𝑒𝑗 || ≥ c ∑𝑛𝑗=1 |𝜉𝑗 |
1 1
∑𝑛𝑗=1 |𝜉𝑗 | ≤ ||∑𝑛𝑗=1 𝜉𝑗 𝑒𝑗 || = ||𝑥 ||
𝑐 𝑐
1
||Tx|| ≤ 𝑚𝑎𝑥𝑘 || 𝑇 𝑒𝑘 || 𝑐 ||𝑥 ||
 1
𝑚𝑎𝑥𝑘 || 𝑇 𝑒𝑘 || 𝑐 = 𝑐′

⟹ ||Tx|| ≤ 𝑐 ′ ||𝑥 ||
⟹ T is bounded. Proved!

Topic 065
Continuity and Boundedness
Theorem: Let T : D(T ) → Y be a linear operator where D (T )  X

and X , Y are normed spaces. Then

T is continuous if and only if T is bounded.
Proof: Let T is continuous at x  D (T)
0

Given 𝜀 > 0, ∃ 𝛿 > 0 such that

||Tx - T𝑥0 || ≤ 𝜀 for all x ∈ D(T) satisfying ||x – x0|| < 𝛿.
We have to prove that T is bounded.
Let y ≠ 0 ∈ D(T) , we need to show ||Ty|| ≤ c ||y||
𝛿
Set x – x0 = ||𝑦|| 𝑦

Since T is linear:
𝛿 𝛿
||Tx - T𝑥0 || = ||T(x - 𝑥0 )||= ||T ||𝑦|| 𝑦||= ||𝑦|| ||Ty||

𝛿 𝜀
||𝑦||
||Ty|| ≤ 𝜀 ⟹ ||Ty|| ≤ 𝛿 ||𝑦||
𝜀
=𝑐
𝛿

⟹ ||Ty|| ≤ 𝑐 ||𝑦|| Proved!

Conversely: Let T is bounded. Let x  D (T).
0

To prove continuity, given 𝜀 > 0 , we nee to find 𝛿 > 0 such that

||x – x0|| < 𝛿 implies || Tx – Tx0|| < 𝜀
|| Tx – Tx0|| = ||T ( x – x0) ||
Since T is bounded.
≤ || T || ||x – x0||
≤ || T || 𝛿
|| T || 𝛿 = 𝜀
|| Tx – Tx0|| ≤ 𝜀
Hence,
T is continuous. Proved!

Topic 066
Continuity, null space
Corollary: Let T be a bounded linear operator.
Then for xn , x ∈ D(T):
(a) xn → x ⟹Txn → Tx
(b) The null space N(T) is closed.
Proof:
(a). We are given that T is linear and bounded operator also
xn → x. We need to show that Txn → Tx.
||Txn – Tx|| = ||T(xn – x)|| ≤ ||T|| ||xn – x||
As xn → x
||Txn – Tx|| ≤ 0
By definition: 0 ≤ ||Txn – Tx||
⟹ Txn – Tx → 0 ⟹ Txn → Tx as xn → x Proved!
(b). We have to prove that N(T) is closed i.e we need to show
that N(T) = ̅̅̅̅̅̅̅
𝑁(𝑇).
We will show this by proving that N(T) ⊂ ̅̅̅̅̅̅̅
𝑁(𝑇) and
̅̅̅̅̅̅̅
𝑁(𝑇) ⊂ 𝑁(𝑇).
̅̅̅̅̅̅̅
𝑁(𝑇) = the set itself along with all its limit points.
⟹ N(T) ⊂ ̅̅̅̅̅̅̅
𝑁(𝑇)
Now, let x ∈ ̅̅̅̅̅̅̅
𝑁(𝑇)
∃ (xn) such that xn ∈ N(T) and xn → x
By (a). Txn → Tx
xn ∈ N(T) ⟹ Txn = 0 ⟹ Tx = 0 ⟹ x ∈ N(T)
⟹ ̅̅̅̅̅̅̅
𝑁(𝑇) ⊂ 𝑁(𝑇)
Hence, N(T) = ̅̅̅̅̅̅̅
𝑁(𝑇). Proved!

Topic 067
Types of Operators
Equal Operators: T1 and T2 are said to be equal operators if
they have the same domain i.e D(T1) = D(T2)
also ∀ x ∈ D(T1) = D(T2) ⟹ Tx1 = Tx2

Restriction of an operator:

The restriction of an operator T: D(T) → Y to a subset B ⊂ D(T)

is denoted by

𝑇|𝐵

and is the operator defined by

𝑇|𝐵 : B → Y 𝑇|𝐵 𝑥 = 𝑇𝑥 for all 𝑥 ∈ 𝐵.

Extension of an operator:

An extension of T to a set M  D(T ) is an operator

such that 𝑇̃ |𝐷(𝑇) = 𝑇, that is 𝑇̃𝑥 = 𝑇𝑥

~
T :M →Y for
all 𝑥 ∈ 𝐷(𝑇).

Hence, T is the restriction of 𝑇̃ to D(T).

Topic 068
Bounded Linear Extension
Theorem: Let T:D(T) → Y be a bounded linear operator where
D(T) lies in a normed space X and Y is a Banach space.
Then T has an extension 𝑇̃: ̅̅̅̅̅̅̅
𝐷(𝑇) → 𝑌. Where 𝑇̃ is a bounded
linear operator of norm ||𝑇̃|| = ||T||.

Proof: Let x ∈ ̅̅̅̅̅̅̅

𝐷(𝑇)
⟹ by definition ∃ (xn ) in D(T) such that xn → x.
Since T is linear and bounded, we have
||Txn – Txm|| = ||T(xn – xm)||
≤ ||T|| ||xn – x m||
⟹ (Txn) is a Cauchy sequence in Y as (xn) converges.
Let (Txn) → 𝑦 ∈ 𝑌
Define
𝑇̃ x = y
Let xn → x , zn → x
(vm) = 〈𝑥1 , 𝑧1 , 𝑥2 , 𝑧2 , … 〉
⟹ vm → x by previous result, (vm) converges.
⟹ subsets (Txn) , (Tzn) also converges to some y.
Now we have to prove that 𝑇̃ is bounded.
𝑇̃ x = T x ∀ x ∈ D(T)
Also T is bounded , xn ∈ D(T)
⟹ ||Txn|| ≤ ||T|| ||xn||
As n→ ∞ , 𝑇𝑥𝑛 → 𝑦 = 𝑇̃ x
||𝑇̃ x|| ≤ ||T|| ||x||
⟹ 𝑇̃ is bounded.
⟹||𝑇̃|| ≤ ||T||
But, ||𝑇̃|| ≥ ||T||
Hence, ||𝑇̃|| = ||T||
Proved!

Topic 069
Linear Functional
A linear functional f is a linear operator with domain in a vector
space X and range in the scalar field K of X; thus,
f : D(f) → K
where K = R if X is real and K = C if X is complete.
We denote linear functional by small letters i.e f , g , h
Bounded Linear Functional:
 A bounded linear functional f is a bounded linear operator with
range in scalar field of the normed space X.
∃ c such that |f(x)| ≤ c ||x||
|𝑓𝑥|
||f ||= 𝑠𝑢𝑝𝑥∈𝐷(𝑓)𝑥≠0 ||𝑥||
= 𝑠𝑢𝑝𝑥∈𝐷(𝑓)||𝑥||=1 |fx|

|fx| ≤ ||f || ||x||

Theorem: A linear functional f with domain D(f) in a normed
space is continuous if and only if f is bounded.

Topic 070
Linear Functionals (Examples)
Norm: || || : X → R
It is functional but not linear.
Dot Product: f : R3 → R
We have some fixed number a ∈ R3 such that ∀ 𝑥 ∈ R3
f(x) = x . a
x = {𝜉1 , 𝜉2 , 𝜉3 } , a = {a1 , a2 , a3}
f(x) = x . a = 𝜉1 a1 + 𝜉2 a2 + 𝜉3 a3
⟹ f is linear.
Now,
Is f bounded?
f(x) = x . a ⟹ |f(x)| = |x . a| ≤ ||x|| ||a||
Let ||x|| = 1 ⟹ ||f || ≤ ||a|| (1)
As we have defined:
|f(x)| ≤ ||f || ||x|| if x = a
|f(a)| ≤ ||f || ||a||
f(a) = a.a = ||a||2
||𝑎||2
||f || ≥ = ||a|| (2)
||𝑎||

By (1) and (2)

||f || = ||a|| ⟹ f is bounded.
Definite Integral:
Defined on C[a,b] where J = [a,b] such that
𝑏
f(x) = ∫𝑎 𝑥(𝑡)𝑑𝑡 the answer will be a field.

Hence it is a linear functional.

𝑏
|f(x)| = | ∫𝑎 𝑥(𝑡)𝑑𝑡| ≤ (𝑏 − 𝑎)𝑚𝑎𝑥𝑡∈𝐽 |𝑥(𝑡)|

= (b – a) ||x||
If ||x|| = 1 ⟹ ||f || ≤ b – a
Now we have to show that ||f || ≥ b – a
Let x = x0 = 1 ⟹ ||x0|| = 1
Now we will use |f(x)| ≤ ||f || ||x||
|𝑓(𝑥0 )| 𝑏 𝑏
||f || ≥ = |f(x0)| = ∫𝑎 1𝑑𝑡 = ∫𝑎 𝑑𝑡 = b – a
||𝑥0 ||

⟹ ||f || ≥ b – a
Hence, ||f || = b – a ⟹ f is bounded.

Topic 071
Linear Functionals (Examples)
Space C[a,b]: We have defined linear functional on this space
C[a,b] in the following way that we fixed an element say t0 ∈ J.
We defined functional operator as x(t) ∈ C[a,b]
f1 (x) = x(t0) ⟹ f1 is linear and bounded.
To find its norm we take:
| f1 (x)| = | x(t0)| ≤ ||x||
If ||x|| = 1 ⟹ ||f1 || ≤ 1 (1)
Now we need to have ||f1 || ≥ 1
|f1(x0)| ≤ ||f1 || ||x0||
1 ≤ ||f1 || 1 ⟹ ||f1 || ≥ 1 (2)
By (1) and (2) : ||f1 || = 1
Space l 2: We choose a fixed element a = (aj)∈ l 2
f(x) = ∑∞
𝑗=1 𝜉𝑗 𝑎𝑗 x ∈ l2

This series is absolutely converges and its value is an element of

field.
Now, it is absolutely convergent as well as this operation is
linear and also it is bounded as well.
|f(x)| = | ∑∞ ∞
𝑗=1 𝜉𝑗 𝑎𝑗 | ≤ ∑𝑗=1 |𝜉𝑗 𝑎𝑗 |

Using Cauchy-Schewartz inequality:

≤ √∑∞ 2 ∞
𝑗=1 |𝜉𝑗 | √∑𝑗=1 |𝑎𝑗 |
2

= ||x|| ||a||
|f(x)| ≤ ||x|| ||a||
⟹ f is bounded
Hence, the functional defined is linear as well as bounded.

Topic 072
Linear Functionals
Algebraic Dual Space:
The set of all linear functionals defined on a vector space X is
itself a vector space. That vector space is called algebraic dual
space.