2
$\begingroup$

I must proof the following:

  • Prop.: let be $\Bbb{R}$ a complete ordered field $$\emptyset \neq A \subset \Bbb{R} \wedge A \mbox{ is bounded below } \to \exists x (x \doteq \inf(A))$$
  • Proof: by contradiction I have $ A \mbox{ has not least element }$ therefore $$\nexists x (x \doteq \inf(A)) \equiv$$$$\equiv \nexists x(x=\max(m(A)))\equiv$$$$\equiv \nexists x(x \in m(A) \wedge x \in M(m(A))\equiv$$$$\equiv \forall x(x \notin m(A) \vee x \notin M(m(A)))$$ but if $x \notin m(A)$ I have an absurd because by hypothesis $A$ is bounded below ($m(A) \neq \emptyset$), if $x \notin M(m(A))$ I have an absurd because $A \neq \emptyset$ and $\forall r \in A (r \in M(m(A)))$ therefore $M(m(A))\neq \emptyset$.

It is correct?

$$m(A)=\{z|z \mbox{ is lower bound}\}$$$$M(A)=\{z|z \mbox{ is upper bound}\}$$

$\endgroup$
  • 2
    $\begingroup$ What is $m(A)$? $\endgroup$ – William Chang Mar 9 '14 at 23:55
  • 2
    $\begingroup$ Are you trying to prove that A has a least element, or that inf(A) exists? $\endgroup$ – user84413 Mar 10 '14 at 0:11
  • 2
    $\begingroup$ Well first of all the result is not true. The set of real numbers is a complete ordered field. $\{\frac 1 n\}$ for natural numbers $n$ is bounded below by $0$. But it has no least element. $\endgroup$ – Ishfaaq Mar 10 '14 at 1:56
  • $\begingroup$ oh, I'm sorry... I edited my post! It is correct? $\endgroup$ – mle Mar 10 '14 at 8:36
  • 2
    $\begingroup$ The result still is not true. You need the additional condition that $A$ is non-empty. Or else $\emptyset$ is a subset of $\Bbb R$ and is bounded below. But its infimum does not exist. $\endgroup$ – Ishfaaq Mar 10 '14 at 8:59
1
$\begingroup$

Right. Your result is equivalent to the completeness property. This is how you form your contradiction:- We know that $A$ is bounded below so $m(A) \neq \emptyset$. And since $A \neq \emptyset$ any element $a \in A$ is an upper bound for $m(A)$. Hence $\sup m(A) = c$ exists. By definition $c \in M(m(A))$. Suppose there is $a \in A$ such that $a \lt c$. This is a contradiction since $a$ is an upper bound for $m(A)$ and is less than its supremum. This gives us that $c \le a \;\; \forall a \in A$ and hence $c \in m(A)$.

Q. E. D.

There is an easier way to prove the result though. Consider the set $B = \{- a \ | \ a \in A\}$ and apply the supremum property to B.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.