# infimum and supremums

Printable View

• January 15th 2008, 08:49 PM
alexmin
infimum and supremums
We showed in class that sqr root 2 is not rational, i.e., that there

is no rational number y such that y^2= 2. The aim of this

problem is to show that there exists a positive real number

y such that y^2= 2.

Let A = {x is in R | x^2< 2}.

(a) Show that A is subset of [−2, 2].

(b) Suppose that y is the least upper bound of A. Show

that

i) y >= 1;

ii)y^2 >= 2;

iii)y^2 <= 2;

iv)y^2 = 2.

[Hint: Use proofs by contradiction for parts ii)–iii).]
• January 15th 2008, 11:14 PM
Jhevon
Quote:

Originally Posted by alexmin
We showed in class that sqr root 2 is not rational, i.e., that there

is no rational number y such that y^2= 2. The aim of this

problem is to show that there exists a positive real number

y such that y^2= 2.

Let A = {x is in R | x^2< 2}.

(a) Show that A is subset of [−2, 2].

we denote the solution to $a^2 = 2$ to be $a = \pm \sqrt{2}$. then clearly, for all $x \in A$, $x \in (-\sqrt {2}, \sqrt {2})$

it can be shown (by examples if necessary) that $|\sqrt{2}| < 2$. thus $A = (-\sqrt{2}, \sqrt{2}) \subset [-2,2]$

(am i begging the question here? i hate doing proofs for things that seem "obvious")

Quote:

(b) Suppose that y is the least upper bound of A. Show

that

i) y >= 1;
well, if you accept what is in (a), this is immediate. i guess you can show it by example, or come up with some approximation to $\sqrt{2}$ or just some number greater than 1 but less than $\sqrt{2}$, and then it would be in A.

Quote:

ii)y^2 >= 2;
assume to the contrary that y^2 < 2. then by the denseness of Q, we can find some rational x in A such that y^2 < x^2 < 2. a contradiction

Quote:

iii)y^2 <= 2;
assume to the contrary that y^2 > 2. clearly we have that y^2 > 2 > x^2 for all x in A. but that means that there is some upper bound of the set A that is less than y, therefore, y cannot be the supremum as stated. there's our contradiction

Quote:

iv)y^2 = 2.
i'm not in the mood for any fancy arguments at this point, i never liked this question to begin with. so i'd just slap on the abstract property of real numbers that says: if a and b are real numbers, with a <= b and b <= a then a = b
• January 16th 2008, 03:07 AM
JaneBennet
(b)(i) is straightforward: use the definition of upper bound. $y$ is an upper bound of $A\ \Rightarrow\ \forall\,x\in A, x\leq y$. In particular, $1\leq y$ since $1\in A$.

b(ii)
If $y^2<2$, we can find a real number $x$ such that $y^2. Since $x$ is positive, we can write $x=a^2$ for some positive real number $a$. Thus $y^2. But $a^2<2\ \Rightarrow\ a\in A\ \Rightarrow$ (given that $y$ is an upper bound of $A$) $a\leq y\ \Rightarrow\ a^2\leq y^2$. That’s our contradiction.

b(iii)
If $y^2>2$, take a real number $x$ such that $2. As before we can write $x=a^2\ (a>0)$, so $2. $a^2 is not an upper bound of $A$ since $y$ is the least upper bound $\Rightarrow\ \exists\,b\in A$ such that $a. But $b\in A\ \Rightarrow\ b^2<2$; then $a. Contradiction again.

Note that b(ii) is true for all upper bounds of $A$, not just the least upper bound. b(iii) only holds for the least upper bound.