Division of the Humanities

and Social Sciences

Introduction to Point-Set Topology

KC Border
v. 2018.10.03::13.35

Abstract
These notes are gathered from several of my other handouts, and are a terse introduc-
tion to the topological concepts used in economic theory. For further study I recommend
Willard [4] and Wilanksy [3]. You may also be interested in my on-line notes on metric
spaces [2].

Contents
1 Topological spaces 1

2 Relative topologies 2

3 Neighborhoods, interiors, closed sets, closures 2

4 Bases 3

5 Product topology 3

6 Continuous functions 3

7 Homeomorphisms 4

8 Compactness 4

9 Topological vector spaces 6

1 Topological spaces
You should know that the collection of open subsets of Rm is closed under finite intersections
and arbitrary unions. Use that as the motivation for the following definition.

1 Definition A topology τ on a nonempty set X is a family of subsets of X, called open

sets satisfying

1. ∅ ∈ τ and X ∈ τ .

1
KC Border Introduction to Point-Set Topology 2

2. The family τ is closed under finite intersections. That is, if U1 , . . . , Um belong to τ , then
∩m
i=1 Ui belongs to τ .

3. The family τ is closed under arbitrary unions. That is, if Uα , α ∈ A, belong to τ , then
∪
α∈A Uα belongs to τ .

The pair (X, τ ) is a topological space.

The topology τ is a Hausdorff topology if for every two distinct points x, y in X there are
disjoint open sets U , V with x ∈ U and y ∈ V .
The collection of open sets in Rm is a Hausdorff topology. A property of X that can be
expressed in terms of its topology is called a topological property.

2 Relative topologies
2 Definition (Relative topology) If (X, τ ) is a topological space and A ⊂ X, then (A, τA )
is a topological space with its relative topology, where τA = {G ∩ A : G ∈ τ }.
Not that if τ is a Hausdorff topology, then τA is also a Hausdorff topology.

3 Neighborhoods, interiors, closed sets, closures

3 Definition The set A is a neighborhood of x if there is an open set U satisfying x ∈ U ⊂ A.
We also say that x is an interior point of A.
The interior of A, denoted int A, is the set of interior points of A.

4 Lemma A set is open if and only it is a neighborhood of each of it points.

Proof : Clearly an open set is a neighborhood of each of its points. So assume the set G is a
neighborhood of each of it points. That is, for each x ∈ G there is an open set Ux satisfying
∪
x ∈ Ux ⊂ G. Then G = x∈G Ux is open, being a union open sets.

5 Exercise The interior of any set A is open (possibly empty), and is indeed the largest open
set included in A. □

6 Definition A set is closed if its complement is open.

The closure of a set A, denoted A or cl A, is the intersection of all the closed sets that
include A.

7 Exercise The union of finitely many closed sets is closed and the intersection of an arbitrary
family of closed sets is closed. □

8 Exercise The closure of A is the smallest closed set that includes A. □

9 Lemma A point x is not in A, that is, x ∈ (A)c , if and only if there is an open neighborhood U
of x disjoint from A.

v. 2018.10.03::13.35
KC Border Introduction to Point-Set Topology 3

Proof : ( ⇐= ) If x ∈ U , where U is open and U ∩ A = ∅, then the complement U c is a closed

set including Ac , so by definition Ac ⊂ U c . Thus x ∈
/ Ac .
( =⇒ ) Since A is closed, if x ∈ c
/ A, then (A) is an open neighborhood of x disjoint from
A, so a fortiori disjoint from A.

10 Definition The boundary of a set A, denoted ∂A, is A ∩ Ac .

11 Corollary ∂A = A \ int A.

Proof : By Lemma 9, int A = (Ac )c . Thus A \ int A = A ∩ Ac = ∂A.

4 Bases
12 Definition A family G of open sets is a base (or basis) for the topology τ if every open set
in τ is a union of sets from G. A neighborhood base at x is a collection N of neighborhoods
of x such that for every neighborhood G of x there is a neighborhood U of x belong to N
satisfying x ∈ U ⊂ G.
In a metric space, the collection of open balls {Bε (x) : ε > 0, x ∈ X} is base for the metric
topology, and {B1/n (x) : n > 0} is a neighborhood base at x.
Given a nonempty family A of subsets of X there is a smallest topology τA on X that
includes A, called the topology generated by A. It consists of arbitrary unions of finite
intersections of members of A. If A is closed under finite intersections, then A is a base for the
topology τA .

5 Product topology
13 Definition If X and Y are topological spaces, the collection sets of the form U × V , where
U is an open set in X and V is an open set in Y , is closed under finite intersections, so it is a
base for the topology it generates on X × Y , called the product topology.

6 Continuous functions
14 Definition Let X and Y be topological spaces and let f : X → Y . Then f is continuous
if the inverse image of open sets are open. That is, if U is an open subset of Y , then f −1 (U ) is
an open subset of X.
This corresponds to the usual ε-δ definition of continuity that you are familiar with.

15 Lemma A function f : X → Y is continuous if and only if the inverse image of every closed
set is closed.

16 Lemma If f : X → Y is continuous, then for every A ⊂ X, we have f (A) ⊂ f (A).

( )
Proof : Since f is continuous and f (A) is closed, f −1 f (A) is a closed set that clearly includes
( )
( ) ( )
A, and so includes its closure A. That is, A ⊂ f −1 f (A) , so f (A) ⊂ f f −1 f (A) = f (A).

v. 2018.10.03::13.35
KC Border Introduction to Point-Set Topology 4

7 Homeomorphisms
17 Definition Let X and Y be topological spaces. A function f : X → Y is a homeomor-
phism if it is a bijection (one-to-one and onto), is continuous, and its inverse is continuous.
If f is homeomorphism U ↔ f (U ) is a one-to-one correspondence between the topologies of
X and Y . Thus X and Y have the same topological properties. They can in effect be viewed
as the same topological space, where f simply renames the points.

8 Compactness
Let K be a subset of a topological space. A family A of sets is a cover of K if
∪
K⊂ A.
A∈A

If each set in the cover A is open, then A is an open cover of K. A family B of sets is a
subcover of A if B ⊂ A and K ⊂ ∪A∈B A.
For example, let K be a subset of R, and for each x ∈ K, let εx > 0. Then the family
A = {(x − εx , x + εx ) : x ∈ K} of open intervals is a open cover of K.

18 Definition A set K in a topological space X is compact if for every family G of open sets
satisfying K ⊂ ∪G (an open cover of K), there is a finite subfamily {G1 , . . . , Gk } ⊂ G with
∪
K ⊂ ki=1 Gi (a finite subcover of K).

19 Lemma If (X, τ ) is a topological space and K ⊂ A ⊂ X, then K is a compact subset of

(A, τA ) if and only if it is a compact subset of (X, τ ).

Proof : Assume K is a compact subset of (X, τ ). Let G be a τA -open cover of K in A. For each
G ∈ G there is some UG ∈ τ with G = UG ∩ A. Then {UG : G ∈ G} is a τ -open cover of K in X,
so it has a finite subcover UG1 , . . . , UGk . But then G1 , . . . , Gk is a finite subcover of K in A.
The converse is similar.

There is an equivalent characterization of compact sets that is sometimes more convenient.

A family A of sets has the finite intersection property if every finite subset {A1 , . . . , An } of
∩
A has a nonempty intersection, ni=1 Ai ̸= ∅.

20 Theorem A set K is compact if and only if every family of closed subsets of K having the
finite intersection property has a nonempty intersection.

Proof : Start with this observation: Let A be an arbitrary family of subsets of K, and define
∩ ∪
A = {K \ A : A ∈ A}. By de Morgan’s Laws A∈A A = ∅ if and only if K = B∈A B. That is,
A has an empty intersection if and only if A covers K.
( =⇒ ) Assume K is compact and let F be a family of closed subsets of K. Then F is
a family of relatively open sets of K. If F has the finite intersection property, by the above
observation, no finite subset of F can cover K. Since K is compact, this implies that F itself
cannot cover K. But then by the observation F has nonempty intersection.

v. 2018.10.03::13.35
KC Border Introduction to Point-Set Topology 5

( ⇐= ) Assume that every family of closed subsets of K having the finite intersection
property has a nonempty intersection, and let G be an open cover of K. Then G is a family of
closed having an empty intersection. Thus G cannot have the fintiie intersection property, so
there is a finite subfamily G0 of G with empty intersection. But then G0 is a finite subfamily of
G that covers K. Thus K is compact.

21 Lemma A closed subset of a compact set is compact.

Proof : Let K be compact and F ⊂ K be closed. Let G be an open cover of F . Then G ∪ {F c }

is an open cover of K. Let {G1 , . . . , Gk , F c } be a finite subcover of K. Then {G1 , . . . , Gk } is a
finite subcover of F .

22 Lemma A compact subset of a Hausdorff space is closed.

Proof : Let K be compact, and let x ∈

/ K. Then by the Hausdorff property, for each y ∈ K there
are disjoint open sets Uy and Vy with y ∈ Uy and x ∈ Vy . By compactness there are y1 , . . . , yk
∪ ∩
with K ⊂ ki=1 Uyi = U . Then V = ki=1 Vyi is an open set satisfying x ∈ V ⊂ U c ⊂ K c . That
is, K c is a neighborhood of x. Since x is an arbitrary member of K c , we see that K c is open
(Lemma 4), so K is closed.

23 Lemma Let f : X → Y be continuous. If K is a compact subset of X, then f (K) is a

compact subset of Y .

Proof : Let G be an open cover of f (K). Then {f −1 (G) : G ∈ G} is an open cover of K. Let
{f −1 (G1 ), . . . , f −1 (Gk )} be a finite subcover of K. Then {G1 , . . . , Gk } is a finite subcover of
f (K).

24 Lemma Let f : X → Y be one-to-one and continuous, where Y is a Hausdorff space and

X is compact. The f : X → f (X) is a homeomorphism, where f (X) has its relative topology
as a subset of Y .

Proof : We need to show that the function f −1 : f (X) → X is continuous. So let G be any
open subset of X. We must show that (f −1 )−1 (G) = f (G) is open in f (X). Now Gc is a closed
subset of X, and thus compact. Therefore f (Gc ) is compact, and since Y is Hausdorff, so is
f (X), so f (Gc ) is a closed subset of Y . Now f (X)∩f (Gc )c = f (G), so f (G) is open in f (X).

25 Weierstrass’s Theorem If K is compact and f : K → R is continuous, then there exists

a point x∗ in K that maximizes f . That is, ( ∀x ∈ K ) [ f (x∗ ) ⩾ f (x) ].
( )
Proof : Since f is continuous, Fα = {x ∈ K : f (x) ⩾ α} = f −1 [α, ∞) is closed for each α ∈ R,
as inverse images of closed sets are closed for a continuous function. Let A = {f (x) : x ∈ K}
denote the range of f . Then F = {fα : α ∈ A} is family of nonempty closed subsets of K
having the finite intersection property.1 Since K is compact, M = ∩ Fα is nonempty. Let x∗
α∈A
belong to M and let x be any point in K. Set α = f (x). Then x∗ ∈ Fα , which means that
f (x∗ ) ⩾ α = f (x). That is, x∗ maximizes f over K.
1
Let α = max{α1 , . . . , αn }. Then Fα = Fα1 ∩ · · · ∩ Fαn , and Fα ̸= ∅ since α belongs to A, the range of f .

v. 2018.10.03::13.35
KC Border Introduction to Point-Set Topology 6

Note that this proof works if we only assume that each Fα is closed, that is, that f is upper
semicontinuous.
The following result is well known.

26 Heine–Borel–Lebesgue Theorem A subset of Rm is compact if and only if it is both

closed and bounded in the Euclidean metric.

This result is special. In general, a subset of a metric space may be closed and bounded
without being compact. (Consider the coordinate vectors in ℓ∞ .)

9 Topological vector spaces

For a detailed discussion of topological vector spaces, see chapter five of the Hitchhiker’s
Guide [1]. But here are some of the results we will need.

27 Definition A (real) topological vector space is a vector space X together with a topol-
ogy τ where τ has the property that the mappings scalar multiplication and vector addition
are continuous functions. That is, the mappings

(α, x) 7→ αx

from R × X to X and
(x, y) 7→ x + y
from X × X to X are continuous. (Where, of course, R has its usual topology, and R × X and
X × X have their product topologies.)

28 Lemma If V is open, then V + y is open.

Proof : Since f : x 7→ x − y is continuous, V + y = f −1 (V ) is open.

29 Lemma If V is open, and α ̸= 0, then αV is open.

Proof : Since f : x 7→ (1/α)x is continuous, αV = f −1 (V ) is open.

30 Definition A set C in a vector space is circled or radial if αC ⊂ C whenever |α| ⩽ 1.

31 Lemma Let V be a neighborhood of zero. Then there is an open circled neighborhood U

of zero included in V .

Proof : The mapping f : (α, x) 7→ αx is continuous, and f (0, 0) = 0, the inverse image f −1 (V )
is a neighborhood of 0. Thus there is an δ > 0 and an open neighborhood W of 0 such that
(−δ, δ) × W ⊂ f −1 (V ). This implies that for any α with |α| < δ and x ∈ W , we have αx ∈ V .
In other words αW ⊂ V . Set ∪
U= αW
α:0<|α|<δ

Then U ⊂ V , U is circled, and U is open, being the union of the open sets αW .

v. 2018.10.03::13.35
KC Border Introduction to Point-Set Topology 7

32 Lemma Let T : X → Y be a linear transformation between topological vector spaces. Then

T is continuous on X if it is continuous at 0.

Proof : It suffices to prove that T is continuous at each point x. So let V be an open neighbor-
hood of T (x). Then (
V − T (x)
)
is an open neighborhood of 0.( Since T )is continuous at 0, the
inverse image T −1 V − T (x) (
, is a neighborhood
)
of 0, so T −1 V − T (x) + x is a neighborhood
of x. But by linearity, T −1 −1
V − T (x) + x = T (V ), and we are done.

References
[1] C. D. Aliprantis and K. C. Border. 2006. Infinite dimensional analysis: A hitchhiker’s guide,
3d. ed. Berlin: Springer–Verlag.

[2] K. C. Border. 2013. What to remember about metric spaces.

[Link]

[3] A. Wilansky. 1998. Topology for analysis. Mineola, NY: Dover. Unabridged republication
of the work originally published by Ginn and Company, Waltham, MA in 1970.

[4] S. Willard. 1970. General topology. Reading, Massachusetts: Addison Wesley.

v. 2018.10.03::13.35