In Bayes theory, Beta distribution is often introduced to make the assumption for parameter p which is from Binomial(N,p). My question is whether Beta distribution can model all probability or ratio in a random process or it should only be applied to a p under the context of a Binomial process?
Thanks heaps