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

Integral of the S.N.D.F.

  1. May 20, 2007 #1
    I fell very silly posting this, but I am making Normal Distribution tables. I tried to Integrate what seems to be quite a simple function:
    Well... I can't I have tried Integration by Parts, Algebraic Substitution etc. But I cannot do it! Can somebody help me?
  2. jcsd
  3. May 20, 2007 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

  4. May 20, 2007 #3
    Ok. But I don't intitutively see why... Can anyone answer this?

    So all tables are created as an approximation - e.g. the Simpson's Rule?
  5. May 23, 2007 #4
    Yes, they are numerical approximations. I can assure you they weren't calculated via Simpson's Rule.

    There is no elementary antiderivative of the pdf. Elementary being sums/products/compositions of 'nice' functions.
  6. May 23, 2007 #5

    Gib Z

    User Avatar
    Homework Helper

    It can be very hard to prove that a certain function has non-elementary antiderivatives. We know in this case that no known function has its derivative as that integral, so we defined one.
  7. May 23, 2007 #6


    User Avatar

    Staff: Mentor

    One could do it by numerical integration, but try a double integral, as in

    [tex]\int_{0}^{x} e^{-x^2} dx\,\int_{0}^{y} e^{-y^2} dy[/tex], where I is each integral.

    combine the two and use the transformation from Cartesian (x, y) to polar coordinates (r, [itex]\theta[/itex]).

    Limits (0,x) and (0, y) become (0, r) and (0, 2[itex]\pi[/itex])
    Last edited: May 23, 2007
  8. May 23, 2007 #7
    I am not too sure about Astronuc's suggestion. But anyway, can someone give me a reasonable function for an approximation. So far I have only come up with tanh(x), with a precding constant...
  9. May 24, 2007 #8


    User Avatar
    Science Advisor

    The double integral method is good for evaluating the definite integral of exp(-x^2/2) from -infinty to infinity, but it is of no use for integrating over finite limits.

    The double integral method works by transforming the square of the integral into a double integral over a region of the x,y plane. Since the element of area dx dy becomes [tex]r \, dr \, d\theta[/tex] then it follows that the inner of the two dimensional ([tex]dr \, d\theta[/tex]) integral becomes the trivial [tex]\int r e^{-r^2/2} dr[/tex].

    The catch is that the limits of the integration correspond to a square region of the x,y plane, so r is not a constant! Bascially the difficulty just gets tranferred to the outer [tex]d\theta[/tex] integral, so in general this is no solution.

    For the specific case of the integral from -infinity to +infinity however the double integral is over the entire x,y plane and therefore the difficulty with the rectangular limits vanishes. This is the standard method of proving that [tex]\int_{-\infty}^{+\infty} e^{-x^2/2} dx = \sqrt{2 \pi}[/tex]

    Goolging for 'erf approximations' gives several very good approximations in the first few hits. One nice simple one is :

    [tex]\frac{1}{\sqrt{2 \pi}} \int_x^{\infty} e^{-x^2/2} \, \simeq \frac{ e^{-x^2/2} } {1.64 x + \sqrt{0.76 x^2 + 4}} [/tex]
    Last edited: May 24, 2007
  10. May 25, 2007 #9
    I see......
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Integral of the S.N.D.F.
  1. Integral ? (Replies: 2)

  2. Integration with i. (Replies: 1)

  3. Beyond integration (Replies: 4)

  4. Numerical integrations (Replies: 2)

  5. What is integration? (Replies: 5)