Proof That {bn}n=1∞ Converges When {an}n=1∞ and {an + bn }n=1∞ Converge

  • Thread starter Thread starter Battlemage!
  • Start date Start date
  • Tags Tags
    Proof
Battlemage!
Messages
292
Reaction score
44

Homework Statement



Suppose {an}n=1 and {bn}n=1 are sequences such that {an}n=1 and {an + bn }n=1 converge.

Prove that {bn}n=1 converges.

Homework Equations



The definition of convergence.

The Attempt at a Solution



I am pretty new to mathematics that requires proof, so excuse me if I do something really stupid... but basically, is this a sufficient proof?1. Assume {an}n=1 converges to A (by hypothesis).

Then for ε/2 > 0 there is a positive integer N1 such that if n ≥ N1, then |an - A| < ε/2.​

2. Assume that {an + bn }n=1 converges to A + B (by hypothesis).

Then for ε > 0 there is a positive integer N = max{N1, N2} such that if n ≥ N, then | (an + bn) - (A + B) | < ε​

3. | (an + bn) - (A + B) | = | (an - A) + (bn - B) | < ε

4. Since by hypothesis |an - A| < ε/2, then

| (an - A) - (an - A) + (bn - B) | < ε - ε/2

| (bn - B) | < ε/2

if n ≥ N2 for some positive integer N2.​

5. But this is the definition of convergence, therefore {bn}n=1 converges (to B).
Thanks.
 
Last edited:
Physics news on Phys.org


You never actually say what N2 is.

Also, if a+b<e and a<e/2 that doesn't give you that b<e/2, so I'm not sure where that final inequality comes from. Subtracting |a_n-A| from the left hand side does not allow you to just move it inside the absolute value sign; and subtracting |a_n-A| from the right hand side, you can't replace it with \epsilon/2 and maintain the same inequality, since that makes the right hand side smaller, not larger
As a fast way to see a lot of results like this, once you have the standard summation and multiplication rules, you can use

b_n=(a_n+b_n)-a_n and use what you know about the summation of sequences
 
Last edited:


There's a problem in step 4 of the attempt. It doesn't follow from

| (an - A) + (bn - B) | < ε

that

| (an - A) - (an - A) + (bn - B) | < ε - ε/2

It's like saying |1-1.9|=0.9 < 1 so |1-1-1.9| < 1-1=0.
 


This looks okay, why not look at the algebra of limits? if a_{n}+b_{n}\rightarrow b} and a_{n}\rightarrow a then the sequence b_{n}=a_{n}+b_{n}-a_{n}\rightarrow b-a
 


Thanks everyone. I'm clearly missing something in the understanding of this material, so I'll take what you've said this weekend and dig through the book and see if I can spot the misunderstanding.
 
Prove $$\int\limits_0^{\sqrt2/4}\frac{1}{\sqrt{x-x^2}}\arcsin\sqrt{\frac{(x-1)\left(x-1+x\sqrt{9-16x}\right)}{1-2x}} \, \mathrm dx = \frac{\pi^2}{8}.$$ Let $$I = \int\limits_0^{\sqrt 2 / 4}\frac{1}{\sqrt{x-x^2}}\arcsin\sqrt{\frac{(x-1)\left(x-1+x\sqrt{9-16x}\right)}{1-2x}} \, \mathrm dx. \tag{1}$$ The representation integral of ##\arcsin## is $$\arcsin u = \int\limits_{0}^{1} \frac{\mathrm dt}{\sqrt{1-t^2}}, \qquad 0 \leqslant u \leqslant 1.$$ Plugging identity above into ##(1)## with ##u...
Back
Top