What Steps Are Needed to Prove That a Number Is the Supremum?

  • Thread starter Thread starter semidevil
  • Start date Start date
Click For Summary

Homework Help Overview

The discussion revolves around proving that a number is the supremum, specifically focusing on the set {1 - 1/n} for natural numbers n. Participants explore the concept of supremum as the least upper bound and the necessary conditions to demonstrate that a number qualifies as such.

Discussion Character

  • Conceptual clarification, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants discuss the two main conditions needed to prove that 1 is the supremum: establishing that 1 is an upper bound and finding elements in the set that are arbitrarily close to 1. Questions arise about the clarity of the original poster's reasoning and the organization of their proof.

Discussion Status

Some participants provide guidance on the steps needed to demonstrate that 1 is the supremum, noting that the original poster has made progress but may need to clarify their justification and organization. Multiple interpretations of the proof steps are being explored, particularly regarding the relationship between upper bounds and the supremum.

Contextual Notes

There is a mention of needing to justify why 1 is considered an upper bound, as well as a discussion about the implications of different inequalities related to the elements in the set.

semidevil
Messages
156
Reaction score
2
ok, so supermum is the least upper bound, but when you write down on paer, what am I trying to show...does that make sense? I mean, what am I trying to get to that will show something is the supermum.

so for example: prove sup {1 - 1/n} = 1 for all n in N.

so I start by saying that I know that 1 is an upper bound...

then, for all epsilon > 0, there exist n st n > epsilon, or 1/n < epsilon.

ok, now what...how else do I need to go to show that 1 is supermum...

how do I know that I have shown that 1 is the supermum??
 
Physics news on Phys.org
To show 1 is a supremum you need to show 2 things,
a) that 1 is an upper bound for your set
b) that you can find points in your set arbitrarily close to 1, That is for every [tex]\epsilon>0[/tex] there is an [tex]x[/tex] in your set where [tex]1-\epsilon<x[/tex].

That's all. It appears you've done both of these steps, but your organization seems unclear. Do you understand how what you've done relates to the above?


An alternate to b) above, you can instead prove
b*)if A is any other upper bound for your set, then 1<=A
 
Show that the number is an upperbound AND any number smaller is NOT an upperbound.
 
shmoe said:
To show 1 is a supremum you need to show 2 things,
a) that 1 is an upper bound for your set
b) that you can find points in your set arbitrarily close to 1, That is for every [tex]\epsilon>0[/tex] there is an [tex]x[/tex] in your set where [tex]1-\epsilon<x[/tex].

That's all. It appears you've done both of these steps, but your organization seems unclear. Do you understand how what you've done relates to the above?


An alternate to b) above, you can instead prove
b*)if A is any other upper bound for your set, then 1<=A

I dotn know what I've done to solve the problem...can you tell me what I did? lol...how haev I already solved the problem?

btw, is 1 - e < x the same as 1/e < x?
 
semidevil said:
so I start by saying that I know that 1 is an upper bound...

This is condition a), though you should should give some justification as to why 1 is an upper bound.

semidevil said:
then, for all epsilon > 0, there exist n st n > epsilon, or 1/n < epsilon.

This is b), but it needs some explanation. To prove b), let [tex]\epsilon >0[/tex] and we want to find something in our set that's larger than [tex]1-\epsilon[/tex]. You've shown that there is an [tex]n\in\Bbb{N}[/tex] where [tex]1/n<\epsilon[/tex]. Now you know that [tex]1-1/n[/tex] is in your set and the last inequality tells us that [tex]1-1/n>1-\epsilon[/tex], and we've found an element in our set and condition b) is satisfied.


semidevil said:
btw, is 1 - e < x the same as 1/e < x?

No. Try e=1/2, x=1.
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
903
  • · Replies 2 ·
Replies
2
Views
986
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 7 ·
Replies
7
Views
3K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 16 ·
Replies
16
Views
4K
Replies
26
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K