Closure and Derived Set in Metric Space
📂MetricSpaceClosure and Derived Set in Metric Space
Definitions
Let’s say (X,d) is a metric space. Suppose p∈X and E⊂X.
A set that contains all q satisfying d(q,p)<r is defined as the neighborhood of point p, denoted by Nr(p). In this case, r is called the radius of Nr(p). If it’s permissible to omit the distance, it can also be denoted as Np.
If every neighborhood of p includes a q that is q=p and q∈E, then p is called a limit point of E.
If all limit points of E are included in E, then E is said to be closed.
If there exists a neighborhood N satisfying N⊂E for p, then p is called an interior point of E.
If every point of E is an interior point of E, then E is said to be open.
The set of all limit points of E is called the derived set of E, denoted by E′.
The union of E and E′ is called the closure, denoted by E=E∪E′.
Theorem 1
For A,B⊂X, the following equations hold:
(1a) A⊂B⟹A′⊂B′
(1b) (A∪B)′=A′∪B′
(1c) (A∩B)′⊂A′∩B′
Proof
(1a)
Assume A⊂B. And let p∈A′. Then p is a limit point of A, so by the definition of a limit point, every neighborhood N of p includes a q that is q=p and q∈A. Assuming A⊂B, the statement means that every neighborhood N of p includes a q that is q=p and q∈B. Therefore, by the definition of a limit point, p∈B′ is established.
■
(1b)
part 1. A′∪B′⊂(A∪B)′
Since A⊂A∪B and B⊂A∪B, by (a1), it follows that:
A′⊂(A∪B)′andB′⊂(A∪B)′
Therefore,
A′∪B′⊂(A∪B)′
part 2. (A∪B)′⊂A′∪B′
Let’s say p∈(A∪B)′. Then, by the definition of a limit point, every neighborhood N of p includes a q that is q=p and q∈A∪B. Rewriting q∈A∪B as q∈A or q∈B means p∈A′ or p∈B′. Therefore, p∈A′∪B′, so it follows that:
(A∪B)′⊂A′∪B′
part 3.
Combining the above results yields:
A′∪B′=(A∪B)′
■
(1c)
Since A∩B⊂A and A∩B⊂B, by (1a), it follows that:
(A∩B)′⊂A′and(A∩B)′⊂B′
Therefore,
(A∩B)′⊂A′∩B′
■
Theorem 2
For A,B⊂X, the following equations hold:
(2a) A⊂B⟹A⊂B
(2b) A∪B=A∪B
(2c) A∩B⊂A∩B
Proof
(2a)
Assuming A⊂B, by (1a), A′⊂B′ is established. Therefore,
A=A∪A′⊂B∪B′=B
■
(2b)
part 1. A∪B⊂A∪B
Let’s say p∈A∪B. This means p∈A∪B or p∈(A∪B)′.
case 1-1. p∈A∪B
In this case, p∈A or p∈B. But since A⊂A and B⊂B,
p∈A or p∈B⟹p∈A∪B
case 1-2. p∈(A∪B)′
By (1b), p∈(A∪B)′=A′∪B′ is established. This means p∈A′ or p∈B′. But since A′⊂A and B′⊂B, similarly to the previous case,
p∈A or p∈B⟹p∈A∪B
By case 1-1, 1-2, the following is established:
A∪B⊂A∪B
part 2. A∪B⊂A∪B
Since A⊂A∪B and B⊂A∪B, by (b1), the following is established:
A⊂A∪BandB⊂A∪B
Therefore,
A∪B⊂A∪B
■
(2c)
Let’s say p∈A∩B. Then, p∈A∩B or p∈(A∩B)′.
case 1. p∈A∩B
In this case, p∈A while p∈B. But since A⊂A and B⊂B,
p∈A and p∈B⟹p∈A and p∈B⟹p∈A∩B
case 2. p∈(A∩B)′
By (1a), (A∩B)′⊂A′ and (A∩B)′⊂B′ are established. But since A′⊂A and B′⊂B,
p∈A′ and p∈B′⟹p∈Aandp∈B⟹p∈A∩B
■
Theorem 3
For metric spaces (X,d) and E⊂X, the following facts hold:
(3a) E is closed.
(3b) E=E being equivalent to E being closed.
(3c) For a closed set F⊂X satisfying E⊂F, E⊂F holds.
(3a) and (3c) imply that E is the smallest closed subset of X that includes E.
Proof
(3a)
Let’s say p∈X and p∈/E. In other words, p∈(E)c. Then p is neither a point of E nor of E′. Therefore, by the definition of an accumulation point, p has at least one neighborhood N where N∩E=∅ is true. Therefore, since N⊂(E)c and p was any point of (E)c, by the definition of an interior point, every point of (E)c is an interior point, which means (E)c is an open set. Since (E)c is an open set, E is a closed set.
■
(3b)
(⟹)
Since E=E=E∪E′, all accumulation points of E are elements of E. This is the definition of a closed set, so E is closed. Or it can be immediately understood from the definitions of closure and being closed.
(⟸)
By the definition of a closed set, all accumulation points of E are included in E. Therefore, E=E∪E′=E
■
(3c)
Let F be a closed set with E⊂F⊂X. Then, by (3b), F′⊂F=F is true. Also, by (2a), E′⊂F′⊂F is true. Therefore, the following holds:
E⊂FandE′⊂F
Therefore,
E∪E′=E⊂F
■
Theorem 4
Let E be a non-empty set of real numbers and suppose it is bounded above. Let y=supE. Then, y∈E is true. Moreover, if E is closed, then y∈E is true.
Proof
If y∈E holds, then the subsequent propositions are trivial by (3a), so we will only prove y∈E. The proof is divided into two cases.
case 1. y∈E
y∈E⊂E
Therefore, it holds.
case 2. y∈/E
Then, for every positive number h>0, there exists x∈E satisfying y−h<x<y. This means that every neighborhood of y, which is Nh(y), must contain an element of E. Therefore, by definition, y is an accumulation point of E. Thus, y∈E∪E′=E
■