# Cauchy's theorem on limits

Tags:
1. Jul 9, 2015

### elliti123

Hello , i was just wondering if anyone could clarify one thing in this proof (its from Konrad Knopp book on infinite series) : If (x0,x1,...) is a null sequence, then the arithmetic means

xn'= x0+x1+x2+...+x/n+1 (n=1,2,3,....)

also forms a null sequence.

Proof: If ε >0 is given, then m can be so chosen, that for every n > m we have |xn| < ε/2 . For these n's, we have
|xn'| ≤ |x1+x2+x3+...+xm| / n+1 +(ε/2) (n-m /n+1)

since the numerator of the first fraction on the right hand side now contains a fixed number, we can further determine n0, so that for n > n0 that fraction remains < ε/2. But then, for every n > n0 , we have |xn'| < ε and our theorem is proved.

My question is: the chosen m in the proof as far as i know is a natural number changing according to what epsilon we give it so for example if the chosen m is 3 it might work for a particular ε but might not for another ε less than the other ε we have chosen first . So i have come to a conclusion that m or n0 that every n should be more than so the sequence converges to a real number is a function of epsilon therefore it changes whenever epsilon does. Now how exactly is the numerator they describe in the proof a fixed number?
Since the m changes whenever ε does then it is logical to infer that the summation of those terms would obviously change. And would you please explain the last part of the proof after the inequality i seem to have some vivid idea but i don't think i still get the last part. Thanks.

Last edited: Jul 9, 2015
2. Jul 9, 2015

### verty

$\epsilon$ is given, you choose a different m and n each time.

3. Jul 10, 2015

### verty

Oh sorry, the one thing you wanted to ask is how the numerator on the left is a fixed number. It isn't but m determines it. M is different each time. So that numerator is different each time. But you choose it to have a certain property, that everything else in the sequence is small. Then you make n large to make that fraction small. Then everything is small.

4. Jul 11, 2015

### elliti123

Haha there is a lot of small going on there.Anyways jokes aside so in this kind of proofs , i mean in general for epsilon proofs you actually do consider the epsilon you choose or give or even the epsilon itself to be a "fixed" positive number right?

5. Jul 11, 2015

### Shivaji Ahir

What Limit is all about, the value we found after evaluating limit what this value actually shows.