Comparing Series Using the Comparison Test

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

Homework Help Overview

The discussion revolves around the convergence or divergence of various series, specifically using the Comparison Test. Participants explore the series n^2/(n^3+n^2) and other related expressions, questioning their behavior as n approaches infinity.

Discussion Character

  • Exploratory, Assumption checking, Mathematical reasoning

Approaches and Questions Raised

  • Participants discuss the limit of the series and its relation to known divergent series, questioning the validity of using limits alone to determine divergence. There is an exploration of simplifying expressions and finding appropriate comparisons with divergent sequences.

Discussion Status

The discussion is active, with participants providing insights and suggestions for approaching the problems. Some participants express uncertainty about certain inequalities and the rigor needed for their arguments, while others propose methods for comparison and approximation.

Contextual Notes

There are mentions of constraints regarding the use of Big O and Small O notation, as well as a focus on epsilon-delta arguments. Participants also highlight the need for rigorous proofs and the challenges of establishing bounds for higher-degree polynomials.

grossgermany
Messages
53
Reaction score
0

Homework Statement



Does the series n^2/(n^3+n^2) diverge?

Homework Equations



We know that 1/n diverges


The Attempt at a Solution



lim n^2/(n^3+n^2) =lim 1/n

Therefore intuitively it should diverge like 1/n

However, I am not very good at the Big O Small O notation. Can someone show me
1. A proof without using Big O or Small O , but purely epsilon delta, for all n>N type of argument
2. A proof involving Big O or Small O notation

Thanks
 
Physics news on Phys.org
You should be trying to show that n^2/(n^3+n^2) is greater that some divergent sequence. Showing that the limit goes to zero won't help you.
But you know that,
n^3+n^2 < 2n^3. :-)
 
You can also start by simplifying it to 1/(n+1)
 
Thanks for the reply. Here is a harder question:
(n^4-10n^3)/(n^5+100n^4)

It is not obvious that we can explicitly find an N such that for all n>N the above expression is larger than 1/n

My point is that is there any rigorous way to show that if lim an= lim bn
then series an diverges if and only if series bn diverges
 
Last edited:
grossgermany said:
Since it's not obvious that
n^3+100n^2< 2n^3

It's not? For sufficiently large n, the higher degree polynomials are always the largest.

edit:
grossgermany said:
My point is that is there any rigorous way to show that if lim an= lim bn
then series an diverges if and only if series bn diverges
You're showing that a sequence bn diverges while lim an > lim bn so of course an diverges as well.
 
grossgermany said:
Thanks for the reply. Here is a harder question:
(n^2-100n)/(n^3+100n^2)

It is not obvious that it is greater than 1/2n^3
Since it's not obvious that
n^3+100n^2< 2n^3

Mentallic said:
It's not? For sufficiently large n, the higher degree polynomials are always the largest.
Metallic is right.

n^3 + 100n^2 <2n^3

100n^2 < 2n^3 -n^3
100n^2 < n^3
100 <n ( since n is never zero we can do this )

All we care about is sufficiently large n.

Besides you could do the following
n^3 + 100n^2 < 101n^3

:)

My point is that is there any rigorous way to show that if lim an= lim bn
then series an diverges if and only if series bn diverges[

There is a theorem or test that says if
lim (an/bn) is finite and >0 and bn diverges then an diverges.
You're showing that a sequence bn diverges while lim an > lim bn so of course an diverges as well.
 
Sorry I changed my question to
(n^4-10n^3+6)/(n^5+100n^4)

There is no easy quintic formula so I made up this example to deter any easy way of explicity finding the N
 
It doesn't make a difference, if you can show that (n^4-10n^3+6)/(2n^5) diverges then you can hence show that (n^4-10n^3+6)/(n^5+100n^4) diverges.
 
What exactly is your question? You seem to be resisting any progress.

You seem to know a lot; perhaps you can help yourself answer your questions :-).
 
  • #10
All we are suggesting is that you should approximate your sequence by an easier one and then use an epsilon argument as freely as you want.
 
  • #11
Please check my answer:
for the comparison test, claim that there exists N such that for all n>=N,
(n^4-10n^3+6)/(n^5+100n^4+999) >(n^4-10n^3+6)/(2n^5)
Proof,need the following
n^5+100n^4+999<2n^5
n+100+999/n^4 <2n
100+999/n^4 <n

Therefore N=101?
 
Last edited:
  • #12
Is this one more difficult? How do we use the comparison test please to show the following?

sqrt[(n^8-10n^3+6)]/{sqrt[(n^7+100n^4+1)]*sqrt[n^3-500n^2+1]}
 
Last edited:

Similar threads

  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
Replies
9
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
2
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 13 ·
Replies
13
Views
4K