Definition Let
S
S
S be a set. An order on
S
S
S is a relation, denote
by
<
<
<, with the following two properties:
-
If x ∈ S x\in S x∈S and y ∈ S y\in S y∈S then one and only one of the statements
x < y , x = y , y < x x<y, x=y, y<x x<y,x=y,y<x is true. -
If x , y , z ∈ S x,y,z\in S x,y,z∈S, if x < y x<y x<y and y < x y<x y<x, then x < z x<z x<z.
Definition An ordered set is a set S S S in which an order is defined.
Definition Suppose
S
S
S is an ordered set,
E
⊂
S
E\subset S
E⊂S, and
E
E
E is
bounded above. Supose there exists an
α
∈
S
\alpha\in S
α∈S with the following
properties:
-
α \alpha α is an upper bound of E E E
-
If γ < α \gamma<\alpha γ<α then γ \gamma γ is not an upper bound of E E E.
Then
α
\alpha
α is called the least upper bound of
E
E
E or the
supremum of
E
E
E
α
=
sup
E
\alpha=\sup{E}
α=supE
Definition Suppose
S
S
S is an ordered set,
E
⊂
S
E\subset S
E⊂S, and
E
E
E is
bounded blow. Suppose there exists an
α
∈
S
\alpha\in S
α∈S with the following
properties:
-
α \alpha α is a lower bound of E E E
-
If γ > α \gamma>\alpha γ>α then γ \gamma γ is not a lower bound of E E E.
Then
α
\alpha
α is called the greatest lower bound of
E
E
E, or the
infimum of
E
E
E
α = inf E \alpha=\inf{E} α=infE
Definition An ordered set
S
S
S is said to have the
least-upper-bound property if the following is true:
If
E
⊂
S
E\subset S
E⊂S,
E
E
E is not empty, and
E
E
E is bounded above, then
sup
E
\sup{E}
supE exists in
S
S
S.
Theorem Suppose
S
S
S is an ordered set with the least-upper-boudn
property,
B
⊂
S
B\subset S
B⊂S,
B
B
B is not empty, and
B
B
B is bounded below. Let
L
L
L be the set of all lower bounds of
B
B
B. Then
α
=
sup
L
\alpha=\sup{L}
α=supL
exists in
S
S
S, and
α
=
inf
B
\alpha=\inf{B}
α=infB
Proof.\
-
∀ α ( a ∈ B ) \forall \alpha (a\in B) ∀α(a∈B), a a a is an upper bound of L L L. Since B B B
is bounded blow, L L Lis not empty. From the definition of a set with
the least-upper-bound property, we get that L L L has a supremum value
in S S S. We call it α \alpha α. α = sup L \alpha=\sup{L} α=supL -
-
∀ γ ( γ < α ) \forall\gamma (\gamma<\alpha) ∀γ(γ<α), γ \gamma γ is not an upper bound
of L L L, which means that γ \gamma γ is a lower bound of B B B and
γ ∈ L \gamma\in L γ∈L. In another word,
∀ γ ( γ < α ) → γ ∈ L \forall\gamma (\gamma<\alpha)\rightarrow \gamma\in L ∀γ(γ<α)→γ∈L. -
∀ β ( β > α ) \forall \beta (\beta>\alpha) ∀β(β>α), β ∉ L \beta\notin L β∈/L, because
α \alpha α is the least upper bound of L L L. β ∉ L \beta\notin L β∈/L means
that β \beta β is not a lower bound of B B B. Hence,
∃ δ ( δ ∈ B ) → δ < β \exists \delta(\delta\in B)\rightarrow \delta<\beta ∃δ(δ∈B)→δ<β.
From the definition of the greatest lower bound, we have shown that
α = inf B \alpha=\inf{B} α=infB -