Good gawd, we need to uniformise the notation in the manuscript!
I personally strong dislike the p(theta | y) ~~ p(y | theta)p(theta) notation because it overloads the letter p with wildly different things. In my writing I usually use p(theta|y) for the posterior, Pr(A|y) for the probability of an event A conditional on the data, f(y|theta) for the likelihood and \pi(theta) for the prior.
Whatever we choose, we should be consistent with it.
Good gawd, we need to uniformise the notation in the manuscript!
I personally strong dislike the
p(theta | y) ~~ p(y | theta)p(theta)notation because it overloads the letterpwith wildly different things. In my writing I usually usep(theta|y)for the posterior,Pr(A|y)for the probability of an eventAconditional on the data,f(y|theta)for the likelihood and\pi(theta)for the prior.Whatever we choose, we should be consistent with it.