Does the series (below) converge absolutely, conditionally, or diverge?
\[\sum_{n=1}^{\infty}(-1)^{n+1}(0.1)^{n}\]
I know that it converges absolutely, but doesn't it fill all 3 criteria for the alternating series test?
I have checked online and people are only mentioning the absolute convergence
The answer seems to only be that it converges absolutely, but it also fulfills all 3 criteria for the alternating series test, so shouldn't it converge conditionally as well?
if the absolute value of series converges, then the actual series also converges.
Yes, I know, but can't a series converge conditionally as well?
If a series converges absolutely, then a series converges conditionally. the converse is not always true (e.g. \( \sum \)(-1)^n * 1/n ) therefore absolute convergence is a stronger condition
we say a series converges conditionally when the absolute value of series diverges but the actual series converges
So I dont even need to prove that it converges conditionally if it converges absolutely?
$$ \Large \rm \sum |a_n|~ converges \Rightarrow \sum a_n ~converges $$
I'm talking about conditionally. If a series converges absolutely then it converges conditionally as well?
yes
But some of the problems only converge absolutely
\[\sum\limits_{n=1}^{\infty} |a_n| \ge \sum\limits_{n=1}^{\infty} a_n\] If the left side series converges, then by comparison test the right side series also converges.
I know that. My confusion is with conditional convergence
@perl some of the series converge absolutely but not conditionally, so thats not true
perl is right, above comparison test tells you that if a series converges absolutely, then the actual series also converges.
conditionally
I am talking about conditional convergence
you seem to have two confusions regarding terminology, let me clear them up
conditional convergence is not regular convergence
And Perl said that if a function converges absolutely, then it converges conditionally as well, which I disproved in my homework like 3 times
could you show one example of your homework where you proved absolute convergence need not imply conditional convergence ?
\[\sum_{n=1}^{\infty}(-1)^{n+1}\frac{ \sin(n) }{ n^2 }\]
sin(n)/n^2 is not always positive, so it does not pass the alternating series test
But it passes the limit comparison test
You're correct about that series not passing the hypothesis of alternating series test. So that just means you cannot use alternating series test for that series. Thats all.
We haven't been given any other tests for conditional convergence
Alternating series test never tells you about divergence
But it does tell you whether a series converges conditionally or not
You cannot apply alternating series test here because the series doesn't meet the hypothesis of alternating series test. End of story.
go and use some other test
I wasn't given another conditional convergence test
That was THE only one
lets look at absolute value of series first
Ok
\[\sum_{n=1}^{\infty}\left| (-1)^{n+1}\frac{ \sin(n) }{ n^2 }\right|~~ =~~ \sum_{n=1}^{\infty}\left|\frac{ \sin(n) }{ n^2 }\right| \]
Yeah, and that converges absolutely by the limit comparison test
Since \(|\sin(n)|\le 1\), we have \[\sum_{n=1}^{\infty}\left|\frac{ \sin(n) }{ n^2 }\right| \le \sum_{n=1}^{\infty}\frac{ 1 }{ n^2 }\] By comparison with p-series test the left hand side series converges.
The absolute value of series converges, therefore the series also converges. And we say the series converges absolutely.
Ok
Absolute converges ====> actual series converges
So therefore it is supposed to converge conditionally, right?
Absolute convergence : when the "absolute value of series" converges, the "actual series" also converges and we say the series converges absolutely. Conditional convergence : when the "absolute value of series" diverges, but the "actual series" converges. we say the series only converges conditionally
mmm
it is wrong to say Absolute convergence ====> conditional convergence correct way to say is : Absolute convergence ====> the actual series converges
Perl said that the first one is true
Stick to the definitions and they will make sense
Ok. But a series can converge both absolutely AND conditionally, right?
tell me this, when do you say "a series converges conditionally" ?
also provide one example of a series that converges conditionally
A series converges conditionally if it passes the Alternating Series Test (it can fill the 3 criteria for the test)
\[\sum_{n=1}^{\infty}(-1)^{n+1}(0.1)^n\] converges conditionally
Wrong.
How so
To say "a series converges conditionally", you must do two things : 1) you must prove that the "absolute value of series diverges" 2) the actual series converges.
does the abolute value of your example series diverge ?
Taken directly from my book: "The series \[\sum_{n=1}^{\infty}(-1)^n+1Un \] converges if all three of the following conditions are satisfied: 1. Each Un is positive; 2. Un >= Un+1 for all n>=N, for some integer N; 3. limn->infty Un->0
It should be (-1)^(n+1)Un
yeah thats alternating series test
Which proves conditional convergence
which can be used to test convergence of "actual series"
Oh snap, I just found a small side-note section talking about conditional convergence only happening if it isn't absolutely convergent, but still convergent
To say "the seris converges conditionally", you must show that the "absolute value of series diverges"
Yep thats right
Oh my god. That means that I just spent like 2 hours doing my homework wrong
In my defense, my teacher never made that clear to us
Haha happens, this thing trips everyone in the start
Ok. So I have to use like the ratio test or the LCT or the DCT or etc to show that the absolute value doesn't converge??
which series are you talking about
Well any series if I want to prove that it converges conditionally (or doesn't if the absolute value doesn't diverge)
Yes. lets work a quick example
Ok cool
\[\sum\limits_{n=1}^{\infty}\dfrac{(-1)^n}{n}\] state whether the series is : 1) conditionally convergent, or 2) absolutely convergent
It can technically be both though, right?
Not specifically this one, but generally?
it can never be both.
look at definition of each of those terms again
Not on the same interval but I mean if a series has an interval of convergence, it could be conditionally convergent at the endpoints but absolutely convergent in the middle
lets not even talk about power series/interval of convergence yet
Ok, but I think that's the root of my confusion
the definitions of "absolute convergence" or "conditional convergence" have nothing to do with power series / alternating series
the definitions are more general, they are not based on alternating series test
you can tell whether a series is "absolutely convergenct" or "conditionally convergent" without using alternating series test.
Ok. But it is possible that if a series has an interval of convergence, that it exhibits both types of convergence. Not in the same places, but over the interval, both happen at different times. I got that the absolute value diverges BTW
Yes, in any given interval, a series is EITHER "absolutely convergent" or "conditionally convergent". It is wrong to say the series is both "absolutely convergent and conditionally convergent". It makes no sense
|dw:1429399637257:dw|
Join our real-time social learning platform and learn together with your friends!