Unit II
Unit II
1
Since S is connected , either A=ϕ∨B=ϕ .
⟹ Either S= A ⋃ B=B∨S= A ⋃ B= A .
−1 −1
⟹ Either S=f {1 }(¿)S=f {0}.
⟹ Either f (S )=1 for all x ∈S (¿)f ( S)=0 for all x ∈S .
⟹∈either case f is a constant on S .
Conversely , assume that every two−valued funct ionon S is a constant
on S∧assume that S is disconnected .
∴ S= A ⋃ B , where A∧B are nonempty disjoint open sets∈S .
{
Define a function f :S {0 , 1}by f ( x)= 0 if x ∈ A ,
1 if x ∈ B .
We know that every⊂of a discrete metric space { 0 ,1 } is both open
¿ closed ∈{0 , 1}.
Therefore {0 },{1}∧{0 ,1 }are open sets∈{0 ,1 }.
−1 −1 −1
Clearly f ({0 })= A , f ({1 })=B∧f ( {0 , 1})=S are open sets ∈S .
Therefore , by the theorem1.2 , f is a continuous function on S .
Hence f is a two valued function on S . By assumption, f is constant on S .
∴ f (x )=0 for all x ∈S (¿)f (x)=1 for all x ∈S .
⟹ S= A ( ¿ ) S=B ⟹ B=ϕ ( ¿ ) A=ϕ ,
which is a contradiction ¿ the that! A ≠ ϕ∧B ≠ ϕ
Therefore S must be connected∧hence thetheorem .
The following theorem shows that continuous image of a connected set
is connected .
Theorem 2.2 . Let f :S → M be a function ¿ one metric space S ¿ anthor
space M ∧let X be a connected⊂of S . If f is continuous on X , then
f ( X )is a connected⊂of M .
Proof . By theorem 2.3 , ¿ prove f ( X ) is connectedis enough prove that every
two−valued function on f ( X ) is a constant .
Let g be a two valued function on f ( X ) ⟹ g is continuous on f (X ).
Now we define a new function h on X by
h(x )=g [f (x)], for all x∈S .
f is continuous on X∧g is on f ( X)⟹ h=g ∙ f is continuous on X
¿ can take only two values0∧1.
2
Therefore h is a two valued function on X .
Since X is connected ,h is a constant function on X ⟹ g is constant on f ( X )
Hence f ( X )is connected . Hence thetheorem .
Theorem 2.3( Intermediate−value theorem for real continuous functi ons)
Statement : Let f be a real valued∧continuous on a connected⊂S of Rn .
If f takes on two different values∈S , say a∧b , thenfor each real c between
a∧b there exits a x ∈S such that f ( x ) =c .
3
compo nent of S .∈other words , the components of S ¿ a collection of disjoint
sets whose∪is S .
P roof . Suppose that a point x of S belongs two different components of S .
Then , by theorem 2.4 the∪of these two component would be a larger conne cted
set containing x∧hence every point of S belongs¿ a unique component of S .
2.3 UNIFORM CONTINUITY
Suppose f is defined on a metric space (S , d s) , with values∈another metric space
(T ,d T ) ,∧assume that f is continuous on a⊂A of S . Then , given any point
p∈A∧any ε> 0 , there is a δ >0 (depending on p∧on ε ) such that
d T ( f (x) , f ( p) ) < ε whenever x ∈ A∧d S ( x , p)< δ .
¿ general , we cannot expect that for a ¿ ε the same value of δ will serve
equally well for every point p∈A .
Whenhappens , the function is called uniformly continuous on A .
Definition 2.3 Let f :S → T be a function ¿ one metric space ( S , d S )
¿ another metric space (T ,d T ). Then f is said ¿ be uniformly continuous on a
¿ A of S if the following condition holds:
For every ε >0 , there exits a δ> 0(depending only on s ε )such that
d T (f ( x ), f ( p))< ε whenever x ≠ p ∈ A∧d S ( x , p)< δ .
The following examples of real – valued function shows the difference between
continuity∧uniform continuity on A .
1
Example 2.1 L et f ( x )= for x >0∧take A=¿ .
x
First we provethat the function f iscontinuous on A .
Let ε >0 be given∧ p∈ A
1 1
| || |
Now ,|f (x )– f ( p)|= − =
x p
x− p
xp
=¿
4
2
εp
Thus for every ε > 0 ,there is a δ= such that
1+ pε
|f ( x ) – f ( p )|<ε whenever |x− p|< δ .−−−−−−−(1)
Therefore f is continuous at p∧hence f iscontinuous on A .
Next we prove that f isnot uniformly continuous on A .
If f is uniformly continuous on A ,then for every ε >0 , there ∃a δ >0
such that∨f ( x) – f ( p)∨¿ ε whenever x , p ∈ A∧¿ x− p∨¿ δ .
Take ε=10 , x=δ , p=δ /11.Then∨x− p∨¿∨δ−δ / 11∨¿ 10 δ /11< δ .
By(1) ,∨f (x) – f ( p)∨¿ 10.
| ||
1 1 1 11 −10 10
But |f ( x ) – f ( p )|= − = − =
x p δ δ δ
= >10.
δ || |
s ince x ∈ A , which is a contradiction .
Therefore f is not uniformly continuous on A .
Example 2.2 Let f : R R be a function defined by f (x)=x 2 , for all x ∈ R .
Take A=( 0 , 1 ] ∧let ε > 0 be given .
If x , p ∈ A , then
|f ( x ) – f ( p )|=|x 2 – p 2|=|( x + p ) ( x – p )|≤ (|x|+| p|)|x – p|< 2|x− p|.
If |x− p|< δ , then|f ( x ) – f ( p )|<2 δ
ε
Choosing δ= , we findd that |f ( x ) – f ( p )|< ε
2
Hence f is uniformly continuous on A .
¿ the above two examples , we find that uniform continuity on a set A implies
continuity on A but the converse need not be true .
The continuity on a set A implies uniformcontinuity on A when A is compact .
2.4 UNIFORM CONTINUITY ∧COMPCT SETS .
Theorem 2.6 ( Heine ) .
Statement . Let f : S → T be a function ¿ one metric space ( S , dT )
another metric space ( T , d S ) . Let A be a compact set∧assume that f is
continuous on A . The n f is uniformly continuous A .
Proof . Let ε>0 be given . Since f is continuous on A , for each point a A ,
there is an r >0 depending on ε ∧a such that
ε
d T ( f ( x ) , f ( a ) )< whenever x , a A∧d S (x , a)< r .
2
5
ε
i .e . ,d T ( f ( x ) , f ( a ) ) < whenever x ∈ B S ( a ,r ) ∩ A−−−−−−−(1).
2
A ⊆¿ k =1¿ m B S ak , ( r2 ) . k
Let δ=min { r1 r2 r3
, , ,…,
2 2 2
rm
2
. }
Consider two points x , p∈A such that d s ( x , p ) <δ .
We prove that d T ( f ( x ) , f ( p ) ) < ε .
Since x ∈ A ⟹ x ∈ BS a k , ( ) rk
2
for some k ⟹ x ∈ BS ( ak , r k ) for some k
6
Note : A contraction on any metric space isuniformly continuous on S .
'
Since f isa contractionof S∧ p , p are∈S , we have
d ( p , p ) =d ( f ( p ) , f ( p ) ) ≤ α d ( p , p ) , where α <1
' ' '
⟹ ( 1−α ) d ( p , p ) ≤ 0 ⟹ p= p .
' '
≤ α [ α d ( p n−1 , p n−2 ) ]
… … . … .. … .
≤ α n d ( p1 , p0 )
7
( )
m −n
n 1−α
¿c α
1−α
¿c(
1−α )
n
α n
a=α , r=α ∈S n
n
Since α <1 , α ⟶ 0 as n ⟶ ∞∧hence d ( p m , pn ) ⟶ 0 as n ⟶ ∞ .
This shows that { p n } isa Cauchy sequence∈S .
But S is complete , so there is a point p ∈S such that pn → p . , i. e lim pn → p
n→∞
( n→∞ )
f ( p ) =f lim p n =lim f ( p n) =lim pn+1= p
n→∞ n→∞
¿ the ε , δ terminology this means that for every ε > 0 , there is a δ>0
such that∨f ( x) – f ¿
Note that f need not be defined at the point c .
If is defined at c∧if f ¿
¿ the ¿ at c .
We say that A isthe ¿ limit of f at c ( a ,b ] if f ( x ) → A as x → c
through values less than c∧we write
lim ¿
−¿
x→ c f ( x ) = A= f ¿¿
¿ the ε , δ terminology this means that for every ε > 0 , there is a δ>0
such that ¿
If f is defined at c∧if f ( x )=f ¿
¿ the ¿ c .
If a< c< b , thenf is continuous at c if ,∧only if ,
8
f (c )=f ¿
We say that f is discontinuous at c if f is not continuous at c .
¿ this case one of the following caseis satisfied :
a) Either f ¿
b) Both f ¿
c) Both f ¿
¿ the cases ( a )∧( b ) , we call c an irremovable discontinuity .
¿ the case (c) , we call c as a removable discontinuity .
Definition 2.7 Let f be defined on [a ,b ]. If f ¿
a) f (c )=f ¿
b) f (c )=f ¿
c) f ¿
If any one of these three is different ¿ 0 ,then c is called a jump discontinuity of f .
2.7 MONOTONIC FUNCTIONS
Definition 2.8. Let f be a real valued function defined on a⊂S of R .
Then f is said ¿ be increasing ( ¿ non decreasing ) on S if for every pair of
points x an d y ∈S , x < y ⟹ f (x) ≤ f ( y )
If x< y ⟹ f ( x ) < f ( y), we say that f is strictly increasing on S .
Similarly , decreasing functionsare defined .
A function is said ¿ be monotonic if is increasing on S∨decreasing on S .
If f is an increasing function , then – f is a decreasing function .
The following theorem shows that functions which are monotonic on compact
intervals always have finite ¿∧¿ limits .
Theorem 2.8 . If f isincreasing on [a ,b ],then f ¿
(a , b)∧we have
f¿
f (a) ≤ f ¿
9
. A
By completeness axiom , A has a , say α = . Then α ≤ f (c ).
For this , we must show that for every ε >0 ,there is a δ >0 such that
A
Since α = , by approximation property there is an element
Similarly f ¿
Therefore , f ¿
10
Proof . Since f is strictly increasing ,is one−¿−one on S .
Therefore f −1 exits .
−1
Hence f is continuous∧strictly increasing(byTheorem 2.9)on[ f (a), f (b)].
11