Help with Limit Proof: A(n) & 1/A(n)

  • Thread starter Thread starter migetmonkee91
  • Start date Start date
  • Tags Tags
    Limit Proof
Click For Summary

Homework Help Overview

The discussion revolves around proving the limit of a sequence A(n) as n approaches infinity, specifically showing that if A(n) converges to A (where A is non-zero), then 1/A(n) converges to 1/A. Participants are exploring the epsilon-delta definition of limits in the context of sequences.

Discussion Character

  • Exploratory, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants discuss the manipulation of the epsilon definition and the conditions under which the limit of the reciprocal sequence converges. There are attempts to relate the convergence of A(n) to the convergence of 1/A(n) through various algebraic transformations.

Discussion Status

Some participants have offered algebraic insights and manipulations to approach the proof, while others express confusion about specific steps and the reasoning behind certain choices. There is an ongoing exploration of the implications of the conditions set for A(n) and the reciprocal sequence.

Contextual Notes

Participants are grappling with the formal definitions of limits and the necessary conditions for convergence, particularly in relation to the behavior of A(n) and its reciprocal. There is mention of needing to establish bounds and conditions for epsilon, as well as the significance of ensuring A(n) remains positive and bounded away from zero.

migetmonkee91
Messages
10
Reaction score
0

Homework Statement



The Lim as n goes to infinity of the sequence A(n) = A, A does not equal zero

Then prove Lim as n goes to infinity of 1/a(n) = 1/A

Homework Equations



How do i attack this? I am using the epsilon definition of course, but I have no idea how to manipulate this.

The Attempt at a Solution



So far i have looked at the formal definitions of the limits of both of the sequences, and seen if they could interact anyway, and so far, i cannot find much.

any help would be greatly appreciated!

I thought about letting Epsilon being fixed for a particular delta, but have no idea how that would work in the long run, and i also tried just plain manipulation.
 
Physics news on Phys.org
Let b_{n}=1/A_{n} and let b=1/A, now what are the conditions that b_n->b as n->infinity?
 
Ah, can't help but feeling like a complete idiot after that...thank you very much

Quickly though, does that really prove it, and why does it?

I know that the definition of the limit is The entire Epsilon > 0 |Xn-L| business, so is that why? It just doesn't seem like a thorough idea, so i just want to be sure.
 
Last edited:
My idea being i still have to PROVE that |bn-B| < epsilon, which i guess I'm still stuck on.
 
Look at:
<br /> \frac{1}{A_{n}}-\frac{1}{A}=-\frac{A_{n}-A}{A_{n}A}<br />
You know that there is an N>N_0 such that |A_{n}-A|<e. Since A_n converges it is possile to find an N such that A_N>A/2. Use these statements to complete the question.
 
I am still having trouble on this and i feel as if i am missing something completely obvious. I got now that |an-A / anA must be less than epsilon. So i know the numerator converges to A, but I have no idea what to do with the denominator.
 
Choose an N>N_0 such that |A_n-A|<A^2e/2 and A_n>A/2 then
<br /> \frac{|A_{n}-A|}{A_{n}A}&lt;\frac{A^{2}\varepsilon /2}{A^{2}/2}=\varepsilon<br />
Hence...
 
Where did the (A^2e/2) / A^2/2 come from?
 
It's a trick, to make:
<br /> \left|\frac{1}{A_{n}}-\frac{1}{A}\right| &lt;\varepsilon<br />
 
  • #10
Ah i see. Is there a particular reason that was chosen? Like why A^2/2, why would i need to try to find An -A > A/2
 
  • #11
I meant An > A/2? I just am more confused then ever.
 
  • #12
You have A_nA on the denoninator, so you have to estimate the A_n, you know that it is possible to find an N such that A_n>A/2, hence there will be a N such that A_nA>A^2/2 or 1/A_nA<2/A^2. Does this make things clearer?
 
  • #13
I'm sorry, can we just start from the beginning on this?

So you have |1/A_n - 1/A| = |-A_n-A/An(A)| due to the definition of a limit and all of that.

I understand this much. Now we understand A_n converges to A, and thus An-A will be less then the right side, so we have to prove A_n(A) also is smaller. This because A_n>A/2, there will be an N S.t. A_nA>A^2/2 sincey oure just multiplying both sides by A. Now this is where i am still confused. You took the reciprocal in your last post but it's A^2/2 In the first post you made. If it's possible for you to take it step by step in a single post that would be greatly helpful.
 
  • #14
You have to show that
<br /> \frac{|A_{n}-A|}{|A_{n}A|}&lt;\varepsilon<br />
We know that A_n converges and that we can find an N_1 in the natural numbers such that |A_n|>|A|/2. Also there is an N_2 such that |A_n-A|<e_{1}. Chose N=max{N_1,N_2}, then:
<br /> \frac{|A_{n}-A|}{|A_{n}A|}&lt;\frac{2\varepsilon_{1}}{A^{2}}=\varepsilon<br />
Choosing
<br /> \frac{2\varepsilon_{1}}{A^{2}}=\varepsilon<br />
Hence you have proved the theorem.
 
  • #15
The only thing now is why E1 x 2 / A^2 = Epsilon?
 
  • #16
It's just to make it look nice, you could leave it as it is and you would have still proven the theorem.
 

Similar threads

  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 12 ·
Replies
12
Views
2K
Replies
5
Views
2K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 20 ·
Replies
20
Views
3K
  • · Replies 9 ·
Replies
9
Views
4K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 5 ·
Replies
5
Views
1K