Series convergence- why is my proof incorrect?

In summary, when proving series convergence, it is important to carefully check each step of your proof and avoid common mistakes such as mixing up terms or misapplying convergence tests. Other helpful tips include using theorems and definitions correctly, providing clear explanations, and double-checking your work. It is also possible for a series to converge even if its terms do not approach zero, as long as certain conditions are met, such as in the Alternating Series Test.
  • #1
JThompson
95
0

Homework Statement



Prove that if [itex]\sum{|a_{n}|}[/itex] converges and [itex](b_{n})[/itex] is a bounded sequence, then [itex]\sum a_{n}b_{n}[/itex] converges.

Homework Equations



Comparison Test part (i): Let [itex]\sum a_{n}[/itex] be a series where [itex]a_{n}\geq 0[/itex] for all n. If [itex]\sum a_{n}[/itex] converges and [itex]|b_{n}|\leq a_{n}[/itex] for all n, then [itex]\sum b_{n}[/itex] converges.

or Cauchy criterion

The Attempt at a Solution



Upon seeing this problem, I immediately thought of a proof using the Comparison Test which seemed easier that using the Cauchy criterion.
Proof:

Since [itex](b_{n})[/itex] is bounded, [itex]\exists M\in\mathbb{R}[/itex] with [itex]|b_{n}|\leq M[/itex] for all n. Then [itex]|a_{n}b_{n}|\leq M|a_{n}|[/itex].
Since [itex]\sum |a_{n}|[/itex] converges, [itex]\sum M|a_{n}|[/itex] converges (we proved this previously).
Since [itex]|a_{n}b_{n}|\leq M|a_{n}|[/itex] and [itex]\sum M|a_{n}|[/itex] converges, [itex]\sum a_{n}b_{n}[/itex] converges by the Comparison Test.

There were no marks indicating which part was incorrect, but I received half credit for the problem and my professor's comment was, "[itex]\sum |a_{n}b_{n}|[/itex] converges, use Cauchy criterion." Where does this proof run afoul?
 
Physics news on Phys.org
  • #2
As your professor's note indicates, you showed that Σ |anbn| converges. You didn't show Σ anbn converges yet.
 
  • #3
But according to the Comparison Test

JThompson said:
Comparison Test part (i): Let [itex]\sum a_{n}[/itex] be a series where [itex]a_{n}\geq 0[/itex] for all n. If [itex]\sum a_{n}[/itex] converges and [itex]|b_{n}|\leq a_{n}[/itex] for all n, then [itex]\sum b_{n}[/itex] converges.
[itex]M|a_{n}|\geq 0[/itex] is obvious since [itex]M\geq 0[/itex]. I proved that [itex]\sum M|a_{n}|[/itex] converges and that [itex]|a_{n}b_{n}|\leq M|a_{n}|[/itex]. It seems to me by the Comparison Test above that [itex]\sum a_{n}b_{n}[/itex]. The final series in the Comparison Test is not an absolute value, so why can I not do the same in this problem? How am I applying the Comparison Test incorrectly?
 
  • #4
Yeah, you're right. Perhaps someone else can spot what we're both missing, or maybe your professor is wrong. You could go ask for clarification.
 
  • #5
I'll ask him if no one responds- it's not urgent- but I have an easier time articulating questions online (because I'm shy), and written responses are easier to understand than verbal responses.
 
  • #6
vela said:
As your professor's note indicates, you showed that Σ |anbn| converges. You didn't show Σ anbn converges yet.
Yes he did.
If

[tex]
\sum |a_nb_n|
[/tex]

then

[tex] \sum a_n b_n [/tex]

converges (absolute convergence)
 
  • #7
statdad said:
Yes he did.
If

[tex]
\sum |a_nb_n|
[/tex]

then

[tex] \sum a_n b_n [/tex]

converges (absolute convergence)
D'oh! I hate it when I miss obvious stuff like that.

I've asked for others to take a look at this thread to see why the professor didn't like the proof.
 
  • #8
Did the problem statement tell you to use the Cauchy criteria?
 
  • #9
The only part of the problem statement that I did not post above was the book's hint.

Hint: Use Theorem 14.4

Theorem 14.4 states that a series converges iff it satisfies the Cauchy Criterion.
A hint is a suggestion, not a requirement. At least, I assumed as much.
 
  • #10
I see nothing wrong with your proof. Using the Cauchy criterion would only obfuscate it, in my opinion.
 

Related to Series convergence- why is my proof incorrect?

1. Why is my proof of series convergence incorrect?

The most common reason for an incorrect proof of series convergence is a mistake in the application of the convergence tests. Make sure to carefully check each step of your proof and ensure that you are using the correct test for the given series.

2. What are some common mistakes to avoid when proving series convergence?

Some common mistakes to avoid when proving series convergence include mixing up the terms in the series, assuming that a series is convergent without proper justification, and misapplying the convergence tests.

3. How can I check my proof for series convergence?

To check your proof for series convergence, you can use other convergence tests or try to find a counterexample. It can also be helpful to ask a colleague or professor to review your proof and provide feedback.

4. What are some helpful tips for proving series convergence?

When proving series convergence, it is important to carefully read and understand the problem, use theorems and definitions correctly, and provide clear and concise explanations for each step of your proof. It can also be helpful to double-check your work and use multiple approaches to verify your answer.

5. Can a series converge even if its terms do not approach zero?

Yes, it is possible for a series to converge even if its terms do not approach zero. This is known as the Alternating Series Test, which states that if the terms of a series alternate in sign and decrease in absolute value, then the series converges. However, note that this is just one of many convergence tests and other conditions must also be met for a series to converge.

Similar threads

  • Calculus and Beyond Homework Help
Replies
2
Views
261
  • Calculus and Beyond Homework Help
Replies
2
Views
737
  • Calculus and Beyond Homework Help
Replies
5
Views
533
  • Calculus and Beyond Homework Help
Replies
14
Views
1K
  • Calculus and Beyond Homework Help
Replies
7
Views
735
  • Calculus and Beyond Homework Help
Replies
1
Views
327
  • Calculus and Beyond Homework Help
Replies
4
Views
389
  • Calculus and Beyond Homework Help
Replies
8
Views
1K
  • Calculus and Beyond Homework Help
Replies
2
Views
122
  • Calculus and Beyond Homework Help
Replies
3
Views
456
Back
Top