0% found this document useful (0 votes)
57 views

0.1 Continuous Functions On Intervals: N N N N N N N N

The document discusses continuous functions on intervals. It begins by defining bounded and continuous functions, and proves that a continuous function on a closed bounded interval is bounded. It then discusses properties like the maximum-minimum theorem, intermediate value theorem, and shows that the range of a continuous function on an interval is an interval. It also introduces uniformly continuous functions and Lipschitz functions, proving certain functions are uniformly continuous.
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)
57 views

0.1 Continuous Functions On Intervals: N N N N N N N N

The document discusses continuous functions on intervals. It begins by defining bounded and continuous functions, and proves that a continuous function on a closed bounded interval is bounded. It then discusses properties like the maximum-minimum theorem, intermediate value theorem, and shows that the range of a continuous function on an interval is an interval. It also introduces uniformly continuous functions and Lipschitz functions, proving certain functions are uniformly continuous.
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/ 6

0.

1 Continuous Functions on Intervals


Definition 0.1.1. A function f : A → R is said to be bounded on A if there
exists a constant M > 0 such that |f (x)| ≤ M for all x ∈ A.

Remark 0.1.2. A function is bounded if the range of the function is a


bounded set of R. A continuous function is not necessarily bounded. For
example, f (x) = 1/x with A = (0, ∞). But it is bounded on [1, ∞).

Theorem 0.1.3. Let I = [a, b] be a closed bounded interval, and f : I → R


be continuous on I. Then f is bounded on I.

Proof. Suppose that f is not bounded on I. Then for each n ∈ N, there


exists xn ∈ I such that |f (xn )| > n. As xn ∈ I, so {xn } is bounded, by
Bolzano-Weierstrass Theorem, there exists an accumulation point of {xn },
so there exists a subsequence of {xnk } so that xnk → x. Since a ≤ xnk ≤ b,
we also have a ≤ x ≤ b., i.e., x ∈ I. Since f is continuous on I, we must have
f (xnk ) → f (x). But this is a contradiction since |f (xnk | > nk ≥ k, k ∈ N.

Remark 0.1.4. In the proof, we use the result of earlier homework: if xn ≤ b


for all n ∈ N, then limn→∞ xn ≤ b. Similarly, if xn ≥ a for all n, then
limn→∞ xn ≥ a.

Suppose that f : S → R. Then define the supremum of f on S, denoted


supS f , to be
sup = sup{f (x) : x ∈ S},
S

similarly, the infimum of f on S is defined by

inf f = inf{f (x) : x ∈}.


S

Note that supS f could be ∞ and inf S f could be −∞, depending on the
function f and S. For example, f (x) = x2 and S = R. Then supS f = +∞
and inf S f = 0. Now if S = (0, 2), then supS f = 4 and inf S f = 0. There
are no points in (0, 2) where f takes 0 and 4.

Remark 0.1.5. If there is c ∈ S such that supS f = f (c), then f has an


absolute maximum on S at c. Similarly for absolute minimum.

1
Theorem 0.1.6. (Maximau-Minimum Theorem) Let ) = [a, b] be a closed ad
bounded interval, and f : I → R be continuous on I. Then f has an absolute
maximum and minimum, i.e., there exist points c, d ∈ I such that

f (c) = sup f (x) and f (d) = inf f.


I I

Proof. As f is continuous on I, so it is bounded. Hence both inf f and sup f


exist. Thus, for each n ∈ N, there is xn ∈ I such that
1
sup f − < f (xn ) ≤ sup f.
I n I

So we have a sequence {xn } ⊂ I, by Bolzana-Weierstrass, there exists a


subsequence xnk → c, so f (xnk ) → f (c). But the limit is unique. Hence
f (c) = supI f . Similarly we can prove that f (d) = inf I f .
Theorem 0.1.7. (Bolzano’s Intermediate Value Theorem) Let f be a con-
tinuous function on [a, b] such that f (a) 6= f (b). Let y be any real number
between f (a) and f (b) . Then there is a c ∈ (a, b) such that f (c) = y.
Proof. Without loss of generality, consider f (a) < y < f (b). First define a
set
S = {x ∈ [a, b] : f (x) < y}.
Thus, S 6= ∅ as a ∈ S. It is clear that S is bounded. So sup S exists, let
c = sup S. Now we prove that f (c) = y. It is clear that a ≤ c ≤ b.
Suppose that c = a. As a ∈ S and f is continuous at a, so f (a) < y which
implies there exists a δ > 0 such that ∀x ∈ [a, a + δ) =⇒ f (x) < y. Each
point in this neighborhood is in S. This is a contradiction to c = sup S.
As c = sup S, there exists a sequence {xn } in S such that xn → c. From
f (xn ) < y =⇒ f (c) ≤ y, as f is continuous at c.
If f (c) < y, again by f being continuous at c, there is a neighborhood
of c, (c − δ, c + δ), such that f (x) < y for all x ∈ (c − δ, c + δ), which is
contradiction, as c = sup S.
Example 0.1.8. Consider f (x) = x2 − 2 on [0, 2]. So f (0) = −2, f (2) = 2.
√ from the theorem, there exists c ∈ (0, 2) such that f (c) = 0,
Let y = 0. Then
in fact, c = 2.
One note about this: if we only consider the set of rationals, then the
graph of x2 − 2 would cross the x-axis without meeting it. Another example
of the set of real numbers complete (axiom 12).

2
Corollary 0.1.9. Let f be continuous function on [a, b] and define m =
inf I f and M = supI f . Then the range of f is the interval [m, M ], i.e.,
f ([a, b]) = [m, M ].
Proof. We know from above, there exists c, d ∈ [a, b] such that f (c) =
m, f (d) = M . And any number y ∈ (m, M ), there is c ∈ (a, b) such that
f (c) = y. From the definition m, M , f does not have values outside [m, M ].
So the range equals [m, M ].
Theorem 0.1.10. Let I be an interval and let f : IR be continuous on I. If
α < β are numbers in I such that f (α) < 0 < f (β) (or f (α) > 0 > f (β)),
then there exists a number c ∈ (α, β) such that f (c) = 0.
Proof. As f is continuous on I, so f is continuous on [α, β]. Apply the
Intermediate Value Theorem with y = 0.
1
Example 0.1.11. Let f (x) = x2 +1
.
1. I1 = (−1, 1). f (I1 ) = ( 21 , 1].
2. I2 = [0, ∞), f (I2 ) = (0, 1].
Lemma 0.1.12. Let S ⊆ R be a nonempty set with the property if x, y ∈ S
with x < y, then [x, y] ⊆ S. Then S is an interval.
Theorem 0.1.13. Let I be an interval and let f : I → R be continuous on
I. Then f (I) is an interval.

0.2 Uniform Continuity


First recall the definition of f being continuous at x0 : ∀ > 0∃δ > 0 3 ∀x :
|x − x0 | <  =⇒ |f (x) − f (x0 )| < .
In general, δ depends on both  and x0 , as function changes rapidly at
some points and flat at some other points. We start some examples to look
into this.
Example 0.2.1. Let f : R → R and f (x) = 2x. Let x0 ∈ R. Consider

|f (x) − f (x0 )| = |2x − 2x0 | = 2|x − x0 |.

From this we can see if we choose δ = /2, we have |x − x0 | < δ =⇒


|f (x) − f (x0 )| < . In this case, δ depends only on , it works for all x0 ∈ R.

3
Example 0.2.2. Let f : (0, ∞) → R with f (x) = 1/x. Let x0 = u > 0.
Consider
|x − u|
|f (x) − f (u)| = .
xu
As x → u, so consider only |x − u| < u/2, i.e., u/2 < x < 3u/2. Then
1/x < 2/u. Hence 1/ux < (1/u)(2/u) = 2/u2 . Now given  > 0, choose
δ = min{u/2, u2 /2}. So when |x − u| < δ =⇒ |f (x) − f (u)| < .
Here δ depends on both  and u. In fact, there is no δ for all u > 0, as
then δ = 0.
See the graph of f (x) = 1/x.

Definition 0.2.3. Let f : D → R is uniformly continuous on E ⊂ D iff


∀ > 0∃δ > 0 3 ∀x, y ∈ E, |x − y|, δ =⇒ |f (x) − f (y)| < . If f is uniformly
continuous on D, then f is uniformly continuous.

Remark 0.2.4. f uniformly continuous on E implies f is continuous on E.


The converse is not true.
3
Example 0.2.5. 1. f : [2.5, 3] → R defined by f (x) = x−2
.

2. f : (0, 6) → R with f (x) = x2 + 2x − 5.


3
3. f : (2, 3) → R with f (x) = x−2
.

Non-uniform Continuity Criteriia Let A ⊆ R and let f : A → R. Then


the following statements are equivalent.

1. f is not uniformly continuous on A.

2. ∃0 > 0 such that for every δ > 0 there are points xδ , yδ ∈ A such that
|xδ − yδ | < δ and |f (xδ ) − f (yδ )| ≥ 0 .

3. ∃0 > 0 and two sequences {xn } and {yn } in A such that lim(xn −yn ) =
0 and |f (xn ) − f (yn )| ≥ 0 for all n ∈ N.

Example 0.2.6. Let f : (0, ∞) → R with f (x) = 1/x. Now pick 0 = 1/2,
and choose xn = 1/n and yn = 1/(n + 1). Then lim(xn − yn ) = 0 and
|f (xn ) − f (yn )| = 1 > 1/2 for all n.

Theorem 0.2.7. Let I be a closed bounded interval and let f : I → R be


continuous on I. Then f is uniformly continuous on I.

4
Proof. If f is not uniformly continuous on I. From the above, ∃0 > 0 and
xn , yn ∈ I such that xn − yn → 0 and |f (xn ) − f (yn )| ≥ 0 for all n. As I
is bounded, so by Bolzana-Weierstrass, there is a subsequence {xnk } of {xn }
that converges to z ∈ I, as I is closed interval. In addition, from

|ynk − z| ≤ |ynk − xnk | + |xnk − z|

ynk → z as well.
Now as f is continuous at z, so we have f (xnk ) → f (z) and f (ynk ) → f (z).
But this is a contradiction, as |f (xn ) − f (yn )| ≥ 0 for all n.

Lipschitz Functions
Definition 0.2.8. Let A ⊂ R and let f : A → R. If there exists a constant
K > 0 such that

|f (x) − f (y)| ≤ K|x − y|, ∀x, y ∈ A,

then f is said to be a Lipschitz function (or to satisfy a Lipschitz condition)


on A.

Geometrically, f is Lipschitz if and only if the slopes of secant line joining


points (x, f (x)) and (y, f (y)) are bounded by K.

Theorem 0.2.9. Let f : A → R is a Lipschitz function, then f is uniformly


continuous on A.

Proof. Let  > 0, choose δ = /K. Then for all x, y ∈ A with |x − y| < δ, we
have |f (x) − f (y)| < .

Example 0.2.10. Consider f : R → R, f (x) = x2 . f is uniformly continu-


ous on [a, b] but not on R.

Proof. Let c ∈ R. Consider

|f (x) − f (c)| = |x2 − c2 | = |x − c||x + c|.

As x is close to c, we assume that |x − c| < 1. So this implies |x| < 1 + |c|,


thus |x + c| ≤ 1 + 2|c|. Hence

|x − c||x + c| < |x − c|(1 + 2|c|).

5

Now for  > 0, choose δ = min{1, 1+2|c| } such that for all x satisfying |x−c| <
δ, |f (x) − f (c)| < , i.e., f is continuous on R.
As δ depends on both  and c, c is larger and larger, the values of δ is
smaller and smaller (as the graph becomes more steeper). There is no such
δ that works for all points. In fact, inf c δ = 0.
But when we consider only on [−a, a] for a > 0. Then

|x + c| ≤ |x| + |c| ≤ 2a,

hence δ = /2a works for all points on [−a, a], i.e., f is uniformly continuous
on [−a, a].

You might also like