For each problem clearly describe the conditional distribution of each coordinate given the others. Then describe the procedure for running Gibbs sampling to sample from the joint distribution. Assume that Gibbs sampling works for continuous densities as well as discrete distributions. Guess the conditional from the structure of the joint distribution. Avoid doing integration as much as possible. Use your knowledge of the all the named one dimensional distributions/ densities. a. Sample from the mixed joint pmf/pdf:

f(p, n) = p(1 - p)^-1, 0
You will need one integration to describe the density of p given N = n.

b. Sample from the joint density: f(p, q, r) =1/ Żpar, 0 < P, q ,r <1.

You will need one easy integration. What is special about this joint density?