When can you not apply the alternating series test?

Click For Summary

Homework Help Overview

The discussion revolves around the application of the alternating series test to the series Ʃ(1 to infinity) ((-1)^n*n^n)/n!. Participants are exploring the conditions under which this test can be applied and the implications of failing to meet those conditions.

Discussion Character

  • Conceptual clarification, Assumption checking

Approaches and Questions Raised

  • Participants question the necessity of the terms being decreasing in absolute value for the alternating series test to be applicable. There is also a discussion on the limit of the terms as n approaches infinity and its relation to convergence.

Discussion Status

Some participants have clarified that the alternating series test is primarily used to test for convergence, and failure to meet its conditions does not imply divergence. There is an ongoing exploration of alternative tests for divergence when the conditions of the alternating series test are not satisfied.

Contextual Notes

Participants are considering the implications of the alternating series test in relation to other convergence tests and the specific characteristics of the series in question.

hahaha158
Messages
79
Reaction score
0

Homework Statement



I have a series

Ʃ(1 to infinity) ((-1)^n*n^n)/n!



Homework Equations





The Attempt at a Solution



apparently you cannot use the alternating series for this question, why is this? It has the (-1)^n, what else is needed to allow you to use the alternating series test?
 
Physics news on Phys.org
hahaha158 said:

Homework Statement



I have a series

Ʃ(1 to infinity) ((-1)^n*n^n)/n!

Homework Equations


The Attempt at a Solution



apparently you cannot use the alternating series for this question, why is this? It has the (-1)^n, what else is needed to allow you to use the alternating series test?

To apply the alternating series test you also need that the absolute value of the terms is decreasing. Yours aren't. E.g. (-1)^n*n doesn't converge.
 
Dick said:
To apply the alternating series test you also need that the absolute value of the terms is decreasing. Yours aren't. E.g. (-1)^n*n doesn't converge.

Would that not fall under one of the rules for alternating series (lim An as n->∞ must equal to 0 for the series to be convergent)? How can you determine a series to be divergent from this part of the test if you are not even able to apply it?
 
hahaha158 said:
Would that not fall under one of the rules for alternating series (lim An as n->∞ must equal to 0 for the series to be convergent)? How can you determine a series to be divergent from this part of the test if you are not even able to apply it?

If ##a_n## as a sequence does not converge to zero then the series ##\Sigma a_n## does not converge. Try that test.
 
Dick said:
If ##a_n## as a sequence does not converge to zero then the series ##\Sigma a_n## does not converge. Try that test.

Is there any way to do it only using the alternating series test?

From what i understand this is the alternating series test

Consider Ʃ(n=1 to ∞) (-1)^(n-1)bn where bn>=0. Suppose
I) b(n+1)<= bn eventually
II) lim as n ->∞ bn=0
Then Ʃ(n=1 to ∞) (-1)^(n-1)bn converges.

It seems like what you are saying is that you cannot test for divergence using the alternating series test, only convergence?

The reason i am asking this is beccause of this

http://imgur.com/kTd1Lbt

It has a choice where it can diverge and a choice where you can not apply alternating series test.

So is it true that for an alternating series test you can test for convergence, but if it does not follow the 2 conditions above, then you must take bn (or an) and apply it to a different test to see it if diverges or converges?
 
Yes, the alternating series test can only be applied to test for convergence. Failure of the alternating series test doesn't prove anything. Try another test. As I said, if ##a_n## doesn't converge to zero the series does NOT converge. That's a good first test for divergence.
 
Dick said:
Yes, the alternating series test can only be applied to test for convergence. Failure of the alternating series test doesn't prove anything. Try another test. As I said, if ##a_n## doesn't converge to zero the series does NOT converge. That's a good first test for divergence.

ok thanks for clarifying
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
1K
Replies
14
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
10K
  • · Replies 6 ·
Replies
6
Views
2K