Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Bounding Argument for Comparison Test

  1. Jun 3, 2015 #1
    I have the sum,
    $$\sum_{n=1}^{\infty} \frac{1}{n^{3}}\sin(n \pi x) \text{, where }0 \leq x \leq 1$$
    I have to show that the series converges, so I'm going to use the Comparison Test.
    $$ \text{If }0 \leq a_n \leq b_n \text{ then}$$$$\text{If }\sum b_n \text{ converges then }\sum a_n \text{ must converges.}$$
    Applying the Comparison Test, I know I must show that,
    $$0 \leq \frac{1}{n^{3}}\sin(n \pi x) \leq \frac{1}{n^3} \text{, where }0 \leq x \leq 1$$
    (This is hard part), Since we know that x is between 0 and 1, I think we can bound it.
    $$0 \leq \frac{1}{n^{3}}\sin(n \pi (0)) \leq \frac{1}{n^{3}}\sin(n \pi (1)) \leq \frac{1}{n^3}$$
    Since 0 times anything is 0, and 1 times anything is 1, we get
    $$ 0 \leq \frac{1}{n^{3}}\sin(0) \leq \frac{1}{n^{3}}\sin(n \pi ) \leq \frac{1}{n^3}$$
    Since sine of 0 is 0, we get
    $$ 0 \leq 0 \leq \frac{1}{n^{3}}\sin(n \pi ) \leq \frac{1}{n^3}$$
    And since sine of n times pi alternates between sine of 1 pi and 2pi, sine equals 0.
    $$ 0 \leq 0 \leq 0 \leq \frac{1}{n^3}$$
    And since this holds true, and we know that 1 divided by n cubed converges due to the p-series, the sum must therefore converge.

    I feel like this isn't correct, because if x = 1/2 and n = 3, we get sine of 3pi/2, which equals -1 and then this falls apart. If possible can someone nudge me in the right direction? Much thanks!
  2. jcsd
  3. Jun 4, 2015 #2


    User Avatar
    Gold Member

    Indeed the term you are summing are not always positive, so your proof is incorrect (your argument about ## \sin(n\pi x) ## is not justified, sine is not an increasing function)

    Instead, try to show that your series is absolutely convergent.
  4. Jun 4, 2015 #3


    User Avatar
    Science Advisor

    Basis of proof:
    [itex]0\le \frac{|sinx|}{n^3} \le \frac{1}{n^3}[/itex]
  5. Jun 4, 2015 #4
    I didn't even think about about absolute convergence, since ratio and root test failed. I think I can take it from here, I just gotta show that it is absolutely convergent using the comparison test and then that will be it. Much Thanks!
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Similar Threads for Bounding Argument Comparison
I Independent arguments of dilogarithms