Results 1 to 5 of 5

Math Help - Order statistics

  1. #1
    Junior Member
    Joined
    Nov 2008
    Posts
    53

    Order statistics

    Let Y1<Y2<Y3<Y4<Y5 denote the order statistics of a random sample of size 5 from a distribution having pdf f(x) = e^(-x), 0<x<inf, zero else where. Show that Z1=Y2 and Z2=Y4-Y2 are independent.(Hint: find the joint pdf of Y2 and Y4)

    I found the joint pdf g(Y2,Y4) and also joint pdf h(Z1,Z2)
    Now how should I proceed this?
    Follow Math Help Forum on Facebook and Google+

  2. #2
    MHF Contributor matheagle's Avatar
    Joined
    Feb 2009
    Posts
    2,763
    Thanks
    5
    Quote Originally Posted by ninano1205 View Post
    Let Y1<Y2<Y3<Y4<Y5 denote the order statistics of a random sample of size 5 from a distribution having pdf f(x) = e^(-x), 0<x<inf, zero else where. Show that Z1=Y2 and Z2=Y4-Y2 are independent.(Hint: find the joint pdf of Y2 and Y4)

    I found the joint pdf g(Y2,Y4) and also joint pdf h(Z1,Z2)
    Now how should I proceed this?
    You are done.
    If you have correctly found the joint density of Z1 and Z2
    and it factors into two functions, g(Z1) and h(Z2), which are NOT necessarily the marginals,
    AND the support is a rectangle, then the two rvs are independent. Most books leave out that fact.
    BUT, you can get the marginal densities of Z1 and Z2 and show that their product gives you the joint density, then you have certainly proved that Z1 and Z2 are independent.
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Junior Member
    Joined
    Nov 2008
    Posts
    53

    Question about your statement

    From your saying "which are NOT necessarily the marginals,
    AND the support is a rectangle, then the two rvs are independent."
    , you mean it's not necessary to find the marginals?
    Because I've seen this from books and lectures and didnt know the reason. They just found joint density and it seems they assume when this joint density can be factored as h(z1)g(z2), then it's independent.
    What's the reasoning for that?
    What should I look it up if I do not understand this?
    Thanks.
    Follow Math Help Forum on Facebook and Google+

  4. #4
    MHF Contributor matheagle's Avatar
    Joined
    Feb 2009
    Posts
    2,763
    Thanks
    5
    Quote Originally Posted by ninano1205 View Post
    From your saying "which are NOT necessarily the marginals,
    AND the support is a rectangle, then the two rvs are independent."
    , you mean it's not necessary to find the marginals?
    Because I've seen this from books and lectures and didnt know the reason. They just found joint density and it seems they assume when this joint density can be factored as h(z1)g(z2), then it's independent.
    What's the reasoning for that?
    What should I look it up if I do not understand this?
    Thanks.

    Its weird but most books leave that out
    You can see if the rvs are indep by just glancing at the joint density
    I teach out of walpole and also wackerly's book
    and both leave that out
    Last edited by mr fantastic; March 8th 2009 at 02:01 PM. Reason: Deleted an uncalled for remark.
    Follow Math Help Forum on Facebook and Google+

  5. #5
    Junior Member
    Joined
    Nov 2008
    Posts
    53

    Can anyone explain this?

    Joint density h(z1,z2) can be factored out as g(z1)q(z2)
    then z1 and z2 are indepent without finding marginals.
    Why is that?
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. order statistics
    Posted in the Advanced Statistics Forum
    Replies: 6
    Last Post: July 11th 2010, 09:19 AM
  2. Order statistics
    Posted in the Advanced Statistics Forum
    Replies: 10
    Last Post: April 4th 2010, 12:43 PM
  3. Order statistics
    Posted in the Advanced Statistics Forum
    Replies: 8
    Last Post: February 15th 2010, 09:25 PM
  4. Order statistics help
    Posted in the Advanced Statistics Forum
    Replies: 4
    Last Post: January 20th 2010, 01:41 PM
  5. Order Statistics!?
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: February 2nd 2009, 11:08 AM

Search Tags


/mathhelpforum @mathhelpforum