0% found this document useful (0 votes)
223 views2 pages

Selected Problems From Royden 9.2

The document discusses several problems regarding metric spaces and properties of open and closed sets. Problem 13 asks if it is possible for two distinct points u and v in a metric space X to have equal open balls B(u,r) and B(v,r). The document proves that this is possible in a discrete metric space, but not in Euclidean space Rn or a normed linear space. Problem 14 proves that if sequences {un} and {vn} converge to points u and v in a metric space (X,ρ), then the sequence of distances {ρ(un,vn)} converges to ρ(u,v). Problem 17 proves that a set is open in one metric

Uploaded by

vinhkhale
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)
223 views2 pages

Selected Problems From Royden 9.2

The document discusses several problems regarding metric spaces and properties of open and closed sets. Problem 13 asks if it is possible for two distinct points u and v in a metric space X to have equal open balls B(u,r) and B(v,r). The document proves that this is possible in a discrete metric space, but not in Euclidean space Rn or a normed linear space. Problem 14 proves that if sequences {un} and {vn} converge to points u and v in a metric space (X,ρ), then the sequence of distances {ρ(un,vn)} converges to ρ(u,v). Problem 17 proves that a set is open in one metric

Uploaded by

vinhkhale
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/ 2

Selected Problems from Chapter 9.

2
Vinh-Kha Le
Problem 13. In a metric space X, is it possible for r > 0 and two distinct
points u and v in X to have B(u, r) = B(v, r)? Is this possible in Euclidean
space Rn ? Is it possible in a normed linear space?
Let us answer the first question. For X a general metric space, it is possible for
B(u, r) = B(v, r). Consider the following example.
Example. Let (X, ) be a discrete metric space of cardinality at least two. Any
two distinct points u and v in X have B(u, r) = B(v, r).
Proof. By definition of the discrete metric, any two points u and u0 in X have
(u, u0 ) = 0 or 1 < 2. This means that B(u, 2) = X for any arbitrary u in X.
Let v be any other member of X. For the same reason, B(v, 2) = X.
Let us answer the second and third questions. It is not possible for two distinct
points u and v in X to have B(u, r) = B(v, r). More generally, this is not
possible in a normed linear space.
Proof. Let X be a linear space under the k k norm. For any u, v, and r, it is
possible to find an x X that belongs to B(u, r) but not B(v, r). Because u 6= v,
we know by the identity of indiscernibles that ku vk > 0. The Archimedian
Property guarantees the existence of an  > 0 for which
0 <  < ku vk so r ku vk < r  < r.
For simplicity, let = r . Now let x be defined as follows.
x=u+

uv
ku vk

Elementary algebra reveals that kx uk < r but kx vk > r.





uv

=<r
kx uk =
ku vk
Similarly,



uv

= ku vk + > r.
kx vk = u v +
ku vk
This completes the proof.
1

Problem 14. Let (X, ) be a metric space in which {un } u and {vn } v.
Show that {(un , vn )} (u, v).
Proof. By the definition of convergence, we know that
lim (un , u) = 0 and lim (vn , v) = 0.

From the Triangle Inequality, we also know that


(u, v) (un , u) (vn , v) (un , vn ) (u, v) + (un , u) + (vn , v).
Taking the limits of every above expression as n gives us
(u, v) lim (un , vn ) (u, v)
n

by monotonicity of the limit. By the squeeze theorem, we can conclude that


lim (un , vn ) = (u, v).

This was what was to be proven.


Problem 17. Prove Proposition 7.
Proposition 7. Let and be equivalent metrics on a nonempty set X. Then
a subset of X is open in the metric space (X, ) if and only if it is open in the
metric space (X, ).
Proof. It is sufficient to prove the condition going forwards. Suppose that O is
open in (X, ). For every x O, there exists an r > 0 such that B (x, r) X.
The equivalence of and guarantees the existence of positive numbers c1 and
c2 such that
c1 c2 everywhere on X, so B (x, r/c2 ) B (x, r) X.
We have a B (x, r/c2 ) X for all x O. Therefore, O is open in (X, ).
Problem 24. Show that for a subset E of a metric space X, the closure of E
is the intersection of all closed subsets of X that contain E.
Proof. We can show that an element x X is in the closure of E if and only
if it belongs to all closed subsets of X that contain E. Let us first prove the
condition going forwards. Suppose that x is in the closure of E. Let F be
some closed subset of X that completely contains E. Every open neighborhood
around x contains an element e of E. Because F completely contains E, e is
an element of F . Therefore, every open neighborhood around x contains an
element belonging to F , so x is a point of closure of F . Because F is closed,
it contains all its points of closure, meaning that x belongs to any closed set F
that completely contains E.
Now let us prove the condition going backwards. Suppose that x is not in
the closure of E. Then x has an open neighborhood O that does not intersect
with E. Notice that Oc is a closed subset of X that completely contains E and
does not contain x. Therefore, x does not belong to all closed subsets fo X that
contain E.

You might also like