1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Improper integral Convergence theorem

  1. Jun 8, 2010 #1
    1. The problem statement, all variables and given/known data

    f(x) is a continuous and positive function when [tex] x\in[0,\infty)[/tex]. (#1)

    [tex] x_n [/tex] is a monotonic increasing sequence, [tex]x_0=0[/tex] [tex],x_n \rightarrow \infty[/tex]. (#2)

    Prove or contradict:

    [tex] \mbox{If } \sum_{n=0}^\infty \int_{x_n}^{x_(n+1)} f(x)dx \mbox{ is convergent (#3) then } \int_{0}^\infty f(x)dx \mbox{ is also convergent.}[/tex]


    3. The attempt

    [tex](*3)\ and\ by\ the\ Cauchy\ Criterion\ \Longrightarrow\ \forall\ \epsilon>0\ \exists\ N_1>0,\ so\ \forall\ m>k>N_1[/tex]

    [tex]\left \int_{x_(k+1)}^{x_(m+1)} f(x)dx=(*1)=| \sum_{n=0}^\infty \int_{x_n}^{x_(n+1)} f(x)dx|<\epsilon\mbox{ (*4)} \right [/tex]


    [tex](*2)\ \Longrightarrow\ \forall\ N_1>0\ \exists\ N>0\ so\ \forall\ n>N_1,\ x_n>N\ \ \ \ (*5) } [/tex]

    [tex](*4)\ and\ (*5)\ \Longrightarrow\ \forall\ m>k>N\ \int_{x_(k+1)}^{x_(m+1)} f(x)dx<\epsilon\\\Longrightarrow\ Cauchy\ Criterion\ \int_{0}^\infty f(x)dx\ is\ convergent. [/tex]

    It seems right to me, but I'm not sure...
    I think i also have vice versa proof.
     
    Last edited: Jun 9, 2010
  2. jcsd
  3. Jun 9, 2010 #2

    lanedance

    User Avatar
    Homework Helper

    how about this as a counter example....

    let
    [tex]f(x) = cos(2 \pi x) [/tex]
    [tex]x_n = n \in \mathds{N} [/tex]

    Then
    [tex] \int_{x_n}^{x_{n+1}} f(x)dx = \int_{x_n}^{x_{n+1}} cos(x)dx
    = 0 [/tex]

    But
    [tex] \int_{0}^{x}cos(2 \pi x)dx = -\frac{1}{\pi} sin(x) [/tex]

    though i guess it likely changes if it is true for any series...
     
    Last edited: Jun 9, 2010
  4. Jun 9, 2010 #3
    But f(x) supposed to be positive and continuous when [tex]x \in [0,\infty)[/tex]

    I'm actually taking about integral inside a series

    And if I am wrong where is my blunder in the proof?
     
  5. Jun 9, 2010 #4

    lanedance

    User Avatar
    Homework Helper

    ok good point, missed the positive - must be too late ;)
     
  6. Jun 9, 2010 #5
    This things happens sometimes, take another glance at my proof =)
    Thank you!
     
    Last edited: Jun 9, 2010
  7. Jun 10, 2010 #6

    Dick

    User Avatar
    Science Advisor
    Homework Helper

    Put an upper bound of N on the sum on the left side instead of infinity. Then you know the left side approaches a limit L as N->infinity. Put a definite upper bound of M on the right side integral. Then you want to show the integral also approaches L as M->infinity. You know for any integer N the sum is equal to the integral with an upper limit of x_N+1. So define F(M)=integral from 0 to M of f(x). You know lim F(x_n)->L as n->infinity. You just want to show F(M)->L as M->infinity. I.e. you just have to worry about the points M that are in between the x_i.
     
  8. Jun 10, 2010 #7
    I wasn't able to fully understand all the points.
    (The limit idea)

    ---------------------------------------------
    1. The problem statement, all variables and given/known data

    f(x) is a continuous and positive function when [tex] x\in[0,\infty)[/tex]. (*1)

    [tex] x_n [/tex] is a monotonic increasing sequence, [tex]x_0=0[/tex] [tex],x_n \rightarrow \infty[/tex]. (*2)

    Prove or contradict:

    [tex] \mbox{If } \sum_{n=0}^\infty \int_{x_n}^{x_(n+1)} f(x)dx \mbox{ is convergent (*3) then } \int_{0}^\infty f(x)dx \mbox{ is also convergent.}[/tex]
    -------------------------------------------------------------

    [tex]From\ (*3)\ \Rightarrow\ \forall\ \epsilon\ > 0\ \exists\ N_1>0\ so\ \forall\ m>k>N_1 [/tex]

    [tex]|\sum_{n=k+1}^{m} \int_{x_n}^{x_{n+1}} f(x)dx|=|\int_{x_{k+1}}^{x_{m+1}} f(x)dx|<\epsilon\ \ \ \ \ (*4) [/tex]

    [tex]From\ (*2)\ \Rightarrow\ \exists\ N_1>0\ so\ \forall\ n>N\ x_n>N\ \ \ \ \ (*5)[/tex]

    [tex]From\ (*4)\ and\ (*5)\ \Rightarrow\ \forall\ \epsilon>0\ and\ \forall\ s>t>N[/tex]

    [tex]|\int_{t}^{s}f(x)dx|<\epsilon [/tex]

    [tex]So\ by\ the\ Cauchy\ Criterion\ \int_{N}^\infty f(x)dx\ is\ convergent\ \ \ \ \ (*6) [/tex]

    [tex]From\ (*1)\ and\ (*6)\ \int_{0}^{\infty} f(x)dx\ is\ convergent [/tex]
     
    Last edited: Jun 10, 2010
  9. Jun 10, 2010 #8

    Dick

    User Avatar
    Science Advisor
    Homework Helper

    Did you understand any of the points? You 'proof' completely misses the point. Where did you use that f(x) is positive? You need to use that. Otherwise lanedance's example shows it's false.
     
  10. Jun 11, 2010 #9
    [tex]From\ (*3)\ and\ (*2)\ and\ (*1)\ \forall\ m:\ \ \ \ \ 0\leq S_m=\int_{0}^{m} f(x)dx\ \leq L[/tex]

    I understand now why my proof is wrong and all your points, thank you!
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook