Results 1 to 2 of 2

Math Help - Disease carrier problem with binomial dist.

  1. #1
    Super Member
    Joined
    Mar 2006
    Posts
    705
    Thanks
    2

    Disease carrier problem with binomial dist.

    Consider a test for a certain illness. Say that the probability of a false-negative is 0.2 and a false-positive is 0.01. Suppose that a disease carrier and 4 non-carriers get tested, what is the probability that we will get at least 2 positive result out of them? And what is the expected number of positive tests?

    My idea:

    So to get at least 2 positive result is saying the probability of having either:

    i. The disease carrier get positive and at least one of the non-carrier get false positive.

    ii. The disease carrier gets false negative and at least two of the non-carrier get false positive.

    Now, I believe I can use the binomial distribution for the non carrier part, implies:

    P( disease carrier pos and at least one non-carrier get pos) =  (.8) b(X \geq 1;4,0.01) = (.8) [ 1 - \frac {4!}{4!} (0.01)^0(0.04)^4 ]

    P( disease carrier neg and at least two non-carrier get pos) =  (.2) b(X \geq 2;4,0.01) = (.8) [ 1 - b(1;4,0.01) - b(0;4,0.01) ]

    Then do I just add them up?

    And how do I find the expected value? I know I need to do  E[X] = np , but I'm not so sure about the extra stuff that comes with the disease carrier.

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

  2. #2
    Flow Master
    mr fantastic's Avatar
    Joined
    Dec 2007
    From
    Zeitgeist
    Posts
    16,948
    Thanks
    5
    Quote Originally Posted by tttcomrader View Post
    Consider a test for a certain illness. Say that the probability of a false-negative is 0.2 and a false-positive is 0.01. Suppose that a disease carrier and 4 non-carriers get tested, what is the probability that we will get at least 2 positive result out of them? And what is the expected number of positive tests?

    My idea:

    So to get at least 2 positive result is saying the probability of having either:

    i. The disease carrier get positive and at least one of the non-carrier get false positive.

    ii. The disease carrier gets false negative and at least two of the non-carrier get false positive.

    Now, I believe I can use the binomial distribution for the non carrier part, implies:

    P( disease carrier pos and at least one non-carrier get pos) =  (.8) b(X \geq 1;4,0.01) = (.8) [ 1 - \frac {4!}{4!} (0.01)^0(0.04)^4 ]

    P( disease carrier neg and at least two non-carrier get pos) =  (.2) b(X \geq 2;4,0.01) = (.8) [ 1 - b(1;4,0.01) - b(0;4,0.01) ]

    Then do I just add them up?

    And how do I find the expected value? I know I need to do  E[X] = np , but I'm not so sure about the extra stuff that comes with the disease carrier.

    Thank you!
    It looks to me like there's missing information. eg. Pr(disease carrier) is unknown but required.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. bayes theorem disease problem
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: November 23rd 2010, 04:07 AM
  2. Normal approx to the binomial dist. questions
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: February 1st 2010, 07:44 PM
  3. Two Ind. Poissons, prove conditional dist of Y1 is binomial
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: December 14th 2009, 09:16 PM
  4. Cramer-Rao lower bound for Binomial dist
    Posted in the Advanced Statistics Forum
    Replies: 5
    Last Post: August 9th 2009, 08:24 AM
  5. Probability Teaser! Binomial dist or what?
    Posted in the Statistics Forum
    Replies: 2
    Last Post: October 19th 2008, 10:03 PM

Search Tags


/mathhelpforum @mathhelpforum