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!

Homework Help: Comparisons of imporper integrals

  1. Feb 15, 2008 #1
    1. The problem statement, all variables and given/known data
    Find the upper bound for

    [tex]\int[/tex] (from 3 to infinity) e^(-x^2)

    [hint: e^(-x^2) < e^(-3x) for x>3]

    3. The attempt at a solution

    I don't understand what they mean by "upper bound"
    Are they asking for the upper bound for the area below e^(-x^2)
    for 3<x<infinity ?

    I knew that the function converges by graphing the function
    y= e^(-x^2) and y=e^(-x) and comparing between those two,
    but I'm not sure what the problem is asking for.
    Pleas help me witht the start!
    1. The problem statement, all variables and given/known data

    2. Relevant equations

    3. The attempt at a solution
  2. jcsd
  3. Feb 15, 2008 #2
    If for two functions [itex]f,\,g[/itex] you have [itex]f(x)<g(x),\, \forall x>x_0[/itex] what can you say about the integrals [itex]\int_{x_0}^\alpha f(x)\,d\,x,\,\int_{x_0}^\alpha g(x)\,d\,x[/itex]?
  4. Feb 15, 2008 #3
    That [itex]\int_{x_0}^\alpha f(x)\,d\,x,\ < \int_{x_0}^\alpha g(x)\,d\,x[/itex] and that if
    \int_{x_0}^\alpha g(x)\,d\,x[/itex] converges, then [itex]\int_{x_0}^\alpha f(x)\,d\,x,\ converges as well?

    But how do I find upper bound knowing this?
  5. Feb 15, 2008 #4
    I meant "integral of g(x)" for the first one and "integral for f(X)" for the second one. Sorry about the mess
  6. Feb 15, 2008 #5
    Let [itex]f(x)=e^{-x^2},\,g(x)=e^{-3\,x}[/itex]. Can you calculate
    [tex]\int_3^\infty g(x)\,d\,x[/tex]?
  7. Feb 15, 2008 #6
    Yes, I calculaed it and
    lim[b to infinity] (-1/3 e^(-5b)) (1/3)
    so the first part goes to zero when you plug in infinity for b
    therefore it convererges to 1/3

    .. so.. is it right that since f(x) is smaller than g(x) it converges as well
    and therefore the possible upper bound should be 1/3 (or smaller than 1/3) ?
  8. Feb 15, 2008 #7
    I think the result is
    [tex]\int_3^\infty e^{-3\,x}\,d\,x=\frac{1}{3}\,e^{-9}[/tex] :smile:
    And the answer is that one upper bound is [itex]\frac{1}{3}\,e^{-9}[/itex], i.e.

    [tex]\int_3^\infty e^{-x^2}\,d\,x<\frac{1}{3}\,e^{-9}[/tex]
    Last edited: Feb 15, 2008
  9. Feb 15, 2008 #8
    ooops,, I guess I made a calculation mistake somewhere in the middle, I'll figure it out what I did wrong. and THANK YOU for your help!!
  10. Feb 15, 2008 #9
    Your are faster in typing! :smile:
    The result I posted it's the correct one, and one upper bound is [itex]\frac{1}{3}\,e^{-9}[/itex], or every other number bigger than that, not smaller!

    Glad I helped :smile:
  11. Feb 15, 2008 #10
    I noticed that you meant 1/3 e^(-9)
    :) Thanks again!
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook