Results 1 to 6 of 6

Math Help - conditional probability

  1. #1
    Junior Member
    Joined
    Aug 2010
    Posts
    31

    conditional probability

    Suppose that independent random variables
    X and Y have distributions N(4, 11) and

    N
    (6, 14), respectively.
    (
    a) Determine c such that Pr(X c) = 2Pr(X <c).
    (
    b) Find Pr{

    4
    (X Y )2 16}.
    (
    c) Suppose that Z is uniformly distributed over the interval [a, b]. Determine the

    values of
    a and b so that E(Y X) = E(Z) and Var(Y X) = Var(Z).

    I can't do part a or part c. If you could go through either in stages it would be most appreciated. Thanks
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Junior Member
    Joined
    Aug 2009
    Posts
    27
    a)  P(X>c) = 2P(X<c) \rightarrow 1 - P(X<c) = 2P(X<c) \rightarrow P(X<c) = 1/3

    Since  X \sim N(4,11) ,  Z = (X - 4)/\sqrt{11} \sim N(0,1)
    and then
      P(X<c)  = P(Z < (c - 4)/\sqrt{11}) = 1/3.
    Now you have to look for a standarized normal table and look
    for which the value of z such that P(Z < z) = 1/3.

    b) Here you just have to find the distribution of W = X - Y. It can be done
    using the Jacobian method, for example.

    c) From letter b) you know what is the distribution of W, so
    its expectation value and variance are also known. The formula to
    compute the expectation and variance of a uniform variable are easily
    obtained and then you have two equations with two unkown variables a and b.
    Solve the system.
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Junior Member
    Joined
    Aug 2010
    Posts
    31
    thanks i think that helped alot.

    For part c) do you mean:

    I use general formula's 0.5(a+b)=E(x)
    & (1/12)(b-a)^2 = V(x)

    to get two simultaneous equations, and then find values for a and c ?

    Does it not matter that part b asked for X-Y, where as part c asks Y-X ?
    Follow Math Help Forum on Facebook and Google+

  4. #4
    Junior Member
    Joined
    Aug 2009
    Posts
    27
    Sorry, I didn't notice that in c) it is Y - X.

    Actually, it is much simpler than what I have said.
    You can use the fact that
    E(X + Y) = E(X) + E(Y)
    and
    VAR(X + Y) = Var(X) + Var(Y)
    since X and Y are independents, and use the formulas you described.
    Follow Math Help Forum on Facebook and Google+

  5. #5
    Junior Member
    Joined
    Aug 2010
    Posts
    31
    cheers for that

    anychance you know how to answer part b and c of this:

    Suppose that a random variable X has distribution function
    F(x) :
    0 - x< 2,
    x/4-0.5 - 2 ≤ x ≤ 6,
    1 - x> 6.
    (i) Sketch the distribution function of X and find Pr(X >3).
    (ii) Find the density function of X.
    (iii) Find the mean and variance of X.

    Part (i) is obvious
    Part (ii) , would the answer be: f(x): 1/4 - (2 ≤ x ≤ 6) , 0 elsewhere
    Part (iii) i don't really understand, even though i've got the formula's

    Thanks again if you could help.
    Follow Math Help Forum on Facebook and Google+

  6. #6
    Junior Member
    Joined
    Aug 2009
    Posts
    27
    Yes, I think so.

    For part lll remember that expectation and variance are just integrals. Since you know the density, just integrate xf(x) under the domain of X to obtain the expectation value (or mean). The variance can be obtained using the appropriate formula.

    See you!
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Replies: 0
    Last Post: July 22nd 2011, 01:39 AM
  2. Conditional Probability using the Law of Total Probability
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: October 7th 2010, 03:01 AM
  3. conditional probability
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: June 20th 2010, 01:38 PM
  4. Conditional Probability
    Posted in the Statistics Forum
    Replies: 3
    Last Post: May 11th 2010, 10:14 AM
  5. Continuous probability - conditional probability
    Posted in the Statistics Forum
    Replies: 1
    Last Post: October 1st 2009, 01:21 AM

Search Tags


/mathhelpforum @mathhelpforum