1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Support of Continuous Conditional Density Functions (Probability)

  1. May 6, 2010 #1
    f(x,y) = x + y is the joint probability density function for continuous random variables X and Y. The support of this function is {0 < x < 1, 0 < y < 1}, which means it takes positive values over this region and zero elsewhere.

    g(x) = x + (1/2) is the probability density function of X, derived by integrating f(x,y) with respect to y, over 0 < y < 1. Since we "took y out of the equation", the support of this function should be 0 < x < 1.

    Question :

    What is the support of the conditional density function h(y|x) = f(x,y)/g(x)?

    My guess is that this function is well defined and takes non zero values only over region {0 < x < 1, 0 < y < 1} and so this is the support.

    But the book I am using states 0 < y < 1 instead. Is it a typo?
     
  2. jcsd
  3. May 7, 2010 #2
    I just realized my mistake :

    x is fixed since we are conditioning Y on a particular value of X = x. So the 0 < x < 1 is irrelevant. Or rather, the conditional probability density for a given X = x is a function of y only. So only the value of y matters in defining the support.

    I feel like an idiot lol.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Support of Continuous Conditional Density Functions (Probability)
Loading...