Results 1 to 6 of 6

Math Help - [SOLVED] Gaussian Integral with Polynomial Term

  1. #1
    Newbie
    Joined
    Mar 2010
    Posts
    3

    [SOLVED] Gaussian Integral with Polynomial Term

    Any help on how to calculate the following would be highly appreciated (I am familiar with the "standard" way of calculating the Gaussian integral in polar coordinates):



    Apologies for a rather straightforward question and thanks in advance.
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Super Member Deadstar's Avatar
    Joined
    Oct 2007
    Posts
    722
    YES I GOT IT!

    Sorry... This has been boggling my mind for a wee while.

    This integral involves the following lemma...

    \int_{-\infty}^{\infty} F(u) dx = \int_{-\infty}^{\infty} F(x)  dx where u = x-\frac{1}{x}.

    So we have...

    \int_{-\infty}^{\infty} e^{-\tfrac{1}{2}\left(x^2 + \tfrac{1}{x^2}\right)}dx

    =\frac{1}{e} \int_{-\infty}^{\infty} e^{-\tfrac{1}{2}\left(x -  \tfrac{1}{x}\right)^2}dx

    WHY..?

    Because e^{-\tfrac{1}{2}\left(x-\tfrac{1}{x}\right)^2} = e^{-\tfrac{1}{2}\left(x^2 - 2 + \tfrac{1}{x^2}\right)} = e^{-\tfrac{1}{2}\left(x^2 + \tfrac{1}{x^2}\right)}e

    So we need an extra e^{-1} outside the integral!

    So continuing we get...

    =\frac{1}{e} \int_{-\infty}^{\infty} e^{-\tfrac{1}{2}\left(x -   \tfrac{1}{x} \right)^2}dx

    =\frac{1}{e} \int_{-\infty}^{\infty} e^{-\tfrac{1}{2}x^2}dx using Lemma!

    =\frac{1}{e} \sqrt{2} \sqrt{\pi}

    = \frac{\sqrt{2\pi}}{e}

    Epic Win!

    Additional calculations

    Showing that

    =\frac{1}{e} \int_{-\infty}^{\infty} e^{-\tfrac{1}{2}x^2}dx = \frac{1}{e} \sqrt{2} \sqrt{\pi}

    Let z = \frac{x}{\sqrt{2}}, then dx = \sqrt{2}dz.

    Hence the integral becomes...

    \frac{\sqrt{2}}{e} \int_{-\infty}^{\infty} e^{-z^2}dz

    =\frac{\sqrt{2 \pi}}{e}
    Last edited by Deadstar; March 31st 2010 at 08:51 AM.
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Newbie
    Joined
    Mar 2010
    Posts
    3
    Thanks, Deadstar. Any chance you could expand a bit/point me towards the proof of the lemma?

    Quote Originally Posted by Deadstar View Post
    YES I GOT IT!

    Sorry... This has been boggling my mind for a wee while.

    This integral involves the following lemma...

    \int_{-\infty}^{\infty} F(u) dx = \int_{-\infty}^{\infty} F(x)  dx where u = x-\frac{1}{x}.

    So we have...

    \int_{-\infty}^{\infty} e^{-\tfrac{1}{2}\left(x^2 + \tfrac{1}{x^2}\right)}dx

    =\frac{1}{e} \int_{-\infty}^{\infty} e^{-\tfrac{1}{2}\left(x -  \tfrac{1}{x}\right)^2}dx

    WHY..?

    Because e^{-\tfrac{1}{2}\left(x-\tfrac{1}{x}\right)^2} = e^{-\tfrac{1}{2}\left(x^2 - 2 + \tfrac{1}{x^2}\right)} = e^{-\tfrac{1}{2}\left(x^2 + \tfrac{1}{x^2}\right)}e

    So we need an extra e^{-1} outside the integral!

    So continuing we get...

    =\frac{1}{e} \int_{-\infty}^{\infty} e^{-\tfrac{1}{2}\left(x -   \tfrac{1}{x} \right)^2}dx

    =\frac{1}{e} \int_{-\infty}^{\infty} e^{-\tfrac{1}{2}x^2}dx using Lemma!

    =\frac{1}{e} \sqrt{2} \sqrt{\pi}

    = \frac{\sqrt{2\pi}}{e}

    Epic Win!

    Additional calculations

    Showing that

    =\frac{1}{e} \int_{-\infty}^{\infty} e^{-\tfrac{1}{2}x^2}dx = \frac{1}{e} \sqrt{2} \sqrt{\pi}

    Let z = \frac{x}{\sqrt{2}}, then dx = \sqrt{2}dz.

    Hence the integral becomes...

    \frac{\sqrt{2}}{e} \int_{-\infty}^{\infty} e^{-z^2}dz

    =\frac{\sqrt{2 \pi}}{e}
    Follow Math Help Forum on Facebook and Google+

  4. #4
    Super Member Deadstar's Avatar
    Joined
    Oct 2007
    Posts
    722
    EDIT: HAHAHA I just realized you were after the proof, not an explanation of how I used it... I'll post it up next post, I don't understand it fully though...

    Sure. Bare in mind I just learned this Lemma so I hope I'm describing it correctly!

    So, the idea is, if we have an integral of the form \int_{-\infty}^{\infty} F(u) dx where u=x - \frac{1}{x}


    This is equal to the integral \int_{-\infty}^{\infty} F(x) dx.

    Where we replace the expression x - \frac{1}{x} by just x.

    So an example... (I think)

    \int_{-\infty}^{\infty} \frac{1}{x - \frac{1}{x}} dx = \int_{-\infty}^{\infty} \frac{1}{x} dx.

    As by the lemma... We can replace the expression x - \frac{1}{x} by just x.

    So... In the example you posted... I just rewrote x^2 + \frac{1}{x^2} as \left(x - \frac{1}{x}\right)^2 and included an e^{-1} outside the integral to cancel out the extra e created.

    So...

    e^{-\frac{1}{2}\left(x^2 + \frac{1}{x^2}\right)}

    = \frac{e}{e}e^{-\frac{1}{2}x^2}e^{-\frac{1}{2x^2}}

    = \frac{1}{e}e^{-\frac{1}{2}x^2} e^1 e^{-\frac{1}{2x^2}}

    = \frac{1}{e}e^{-\frac{1}{2}x^2 + 1 - \frac{1}{2x^2}}

    = \frac{1}{e}e^{-\frac{1}{2}\left(x^2 - 2 + \frac{1}{x^2}\right)}

    = \frac{1}{e}e^{-\frac{1}{2}\left(x - \frac{1}{x}\right)^2}

    However, I did it the opposite way round and just multiplied the integral by the inverse of the extra term I had.

    But anyway... So we have the integral

    \int_{-\infty}^{\infty} \frac{1}{e}e^{-\frac{1}{2} \left(x - \frac{1}{x}\right)^2} = \frac{1}{e}\int_{-\infty}^{\infty} e^{-\frac{1}{2}\left(x -  \frac{1}{x}\right)^2}

    and by the Lemma, we can replace the x - \frac{1}{x} term by x.

    Hence we are left with...

    \frac{1}{e} \int_{-\infty}^{\infty} e^{-\frac{1}{2}x^2} which is much more workable as shown in my additional calculations.

    This clear things up? I'm still learning about it as I say so hopefully one of the mods could step in and say whether I'm doing things right.
    Follow Math Help Forum on Facebook and Google+

  5. #5
    Super Member
    Joined
    Jan 2009
    Posts
    715
    Almost a year ago , someone posted it as a challenge . I think this thread has been passed into oblivion ...

    http://www.mathhelpforum.com/math-he...ation-1-a.html
    Follow Math Help Forum on Facebook and Google+

  6. #6
    Newbie
    Joined
    Mar 2010
    Posts
    3
    Thanks for the link, simplependulum!
    And in general - really valuable input guys - appreciate it.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. The degree- term in the Maclaurin polynomial
    Posted in the Calculus Forum
    Replies: 2
    Last Post: May 27th 2011, 01:52 AM
  2. four term polynomial with gcf
    Posted in the Algebra Forum
    Replies: 4
    Last Post: October 9th 2010, 08:39 AM
  3. Replies: 9
    Last Post: June 4th 2010, 09:48 AM
  4. Replies: 0
    Last Post: February 21st 2009, 10:40 AM
  5. [SOLVED] Ratio of the second term to the first term
    Posted in the Math Topics Forum
    Replies: 4
    Last Post: September 26th 2008, 07:13 AM

Search Tags


/mathhelpforum @mathhelpforum