This is an example of the Gibbs Sampler that will sample from the joint distribution
h(x,y) when we are given only the conditionals f(x|y) and g(y|x). In this
particular application, we will be taking f(x|y) = (1 - y)^-1 for 0 < y < x < 1 and
g(y|x) = 3*y^2/x^3 for 0 < y < x < 1.
We will take X = Uniform(0,1), Y = Uniform(0,1) as our initial random vector (X,Y)
Enter the desired size for the bivariate sample:
Bivariate sample using Gibbs Sampler: Sample of X component is in top box, Y component in bottom box