Results 1 to 3 of 3
Like Tree2Thanks
  • 1 Post By HallsofIvy
  • 1 Post By romsek

Math Help - Paradox. Extracting probability density function from mean.

  1. #1
    Junior Member
    Joined
    Apr 2012
    From
    Yerevan
    Posts
    25
    Thanks
    1

    Smile Paradox. Extracting probability density function from mean.

    Is it possible to obtain probability density function of continuous distribution from it's mean?

    1. No. Probability density function contains more information about distribution than it's mean. Particularly one can obtain same mean for distributions with different probability density functions.

    2. Yes.
    <A>=∫−∞∞A∗pdfA(A)dA

    can be viewed as special case of Fredholm integral equation of the first kind
    (Fredholm Integral Equation of the First Kind -- from Wolfram MathWorld)
    where

    f(x)=<A>
    K(x−t)=A
    ϕ(t)=pdfA(A)

    hence by solving this for pdfA(A) we will obtain probability density function of the distribution from it's mean.

    I can not find where am I doing mistake in my reasoning. It should be in the second point I think.
    I will be thankful if you will provide solution for this paradox.

    Regards,
    Rafayel
    Last edited by Rafael; March 14th 2014 at 05:38 PM.
    Follow Math Help Forum on Facebook and Google+

  2. #2
    MHF Contributor

    Joined
    Apr 2005
    Posts
    16,396
    Thanks
    1846

    Re: Paradox. Extracting probability density function from mean.

    It should be obvious that you can't get the probability density function from its mean because many different probability densities have the same mean.

    The error in your (2) is that solutions to such a Fredholm equation are NOT unique.
    Thanks from Rafael
    Follow Math Help Forum on Facebook and Google+

  3. #3
    MHF Contributor
    Joined
    Nov 2013
    From
    California
    Posts
    2,789
    Thanks
    1147

    Re: Paradox. Extracting probability density function from mean.

    Quote Originally Posted by Rafael View Post
    Is it possible to obtain probability density function of continuous distribution from it's mean?

    1. No. Probability density function contains more information about distribution than it's mean. Particularly one can obtain same mean for distributions with different probability density functions.

    2. Yes.
    <A>=∫−∞∞A∗pdfA(A)dA

    can be viewed as special case of Fredholm integral equation of the first kind
    (Fredholm Integral Equation of the First Kind -- from Wolfram MathWorld)
    where

    f(x)=<A>
    K(x−t)=A
    ϕ(t)=pdfA(A)

    hence by solving this for pdfA(A) we will obtain probability density function of the distribution from it's mean.

    I can not find where am I doing mistake in my reasoning. It should be in the second point I think.
    I will be thankful if you will provide solution for this paradox.

    Regards,
    Rafayel
    Is it possible to find "a" probability density function from it's mean. Sure. Pick a density function of your choice and select parameters so that it's mean equals the number you are given.

    Is it possible to find "the" probability density function from it's mean. No, because it isn't unique. For any given number there are infinitely many density functions having this mean.
    Thanks from Rafael
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Replies: 3
    Last Post: September 8th 2012, 05:00 AM
  2. Replies: 4
    Last Post: October 27th 2010, 06:41 AM
  3. Probability density function
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: March 17th 2010, 11:42 AM
  4. Probability Density Function
    Posted in the Calculus Forum
    Replies: 4
    Last Post: May 29th 2008, 11:28 PM
  5. probability density function
    Posted in the Advanced Statistics Forum
    Replies: 2
    Last Post: February 23rd 2006, 10:46 AM

Search Tags


/mathhelpforum @mathhelpforum